Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,61 @@
+---
+library_name: transformers
+license: other
+base_model: google/gemma-2-9b
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: gemma-2-9b-cpt
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# gemma-2-9b-cpt
+
+This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b) on the pretrain202404 dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 128
+- total_eval_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 2.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.45.0.dev0
+- Pytorch 2.4.1+cu121
+- Datasets 2.20.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.999043011324366,
+    "total_flos": 3.076816830283645e+19,
+    "train_loss": 2.254816828382776,
+    "train_runtime": 45442.7326,
+    "train_samples_per_second": 6.622,
+    "train_steps_per_second": 0.052
+}

config.json

@@ -0,0 +1,34 @@
+{
+  "_name_or_path": "google/gemma-2-9b",
+  "architectures": [
+    "Gemma2ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attn_logit_softcapping": 50.0,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": 1,
+  "final_logit_softcapping": 30.0,
+  "head_dim": 256,
+  "hidden_act": "gelu_pytorch_tanh",
+  "hidden_activation": "gelu_pytorch_tanh",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 8192,
+  "model_type": "gemma2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 42,
+  "num_key_value_heads": 8,
+  "pad_token_id": 0,
+  "query_pre_attn_scalar": 256,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "sliding_window_size": 4096,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.0.dev0",
+  "use_cache": false,
+  "vocab_size": 256000
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": 1,
+  "pad_token_id": 0,
+  "transformers_version": "4.45.0.dev0"
+}

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:262bada8b45458f4275b39fa2e4917a8e188d0ac48d6dbf00829c1a61b1cd7d4
+size 4903351912

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70e16f2ab897f6197851ce529f9db894c7f1238c4a838fe15fbd9e662e4d7a69
+size 4947570872

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87f20634c29d2462d731ee4de3383600a89924eb6ffa8ba1509b094931d0d736
+size 4962221464

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fde82793319fae5aac0a5e5724f6eb5ddb11db407f2caaccad40b620edbbf225
+size 3670322200

model.safetensors.index.json

@@ -0,0 +1,471 @@
+{
+  "metadata": {
+    "total_size": 18483411968
+  },
+  "weight_map": {
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.post_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.pre_feedforward_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.32.post_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.32.pre_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.post_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.pre_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.post_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.pre_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.post_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.pre_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.post_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.pre_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.post_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.pre_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.post_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.pre_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.post_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.pre_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.post_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.pre_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.post_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.pre_feedforward_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00004-of-00004.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,34 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003

tokenizer_config.json

@@ -0,0 +1,2015 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "split_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.999043011324366,
+    "total_flos": 3.076816830283645e+19,
+    "train_loss": 2.254816828382776,
+    "train_runtime": 45442.7326,
+    "train_samples_per_second": 6.622,
+    "train_steps_per_second": 0.052
+}

trainer_log.jsonl

@@ -0,0 +1,471 @@
+{"current_steps": 5, "total_steps": 2350, "loss": 7.8552, "learning_rate": 1.0638297872340427e-06, "epoch": 0.0042532830028178, "percentage": 0.21, "elapsed_time": "0:01:55", "remaining_time": "15:05:28"}
+{"current_steps": 10, "total_steps": 2350, "loss": 6.1537, "learning_rate": 2.1276595744680853e-06, "epoch": 0.0085065660056356, "percentage": 0.43, "elapsed_time": "0:03:33", "remaining_time": "13:52:33"}
+{"current_steps": 15, "total_steps": 2350, "loss": 4.3481, "learning_rate": 3.1914893617021277e-06, "epoch": 0.0127598490084534, "percentage": 0.64, "elapsed_time": "0:05:11", "remaining_time": "13:27:56"}
+{"current_steps": 20, "total_steps": 2350, "loss": 3.4135, "learning_rate": 4.255319148936171e-06, "epoch": 0.0170131320112712, "percentage": 0.85, "elapsed_time": "0:06:48", "remaining_time": "13:13:21"}
+{"current_steps": 25, "total_steps": 2350, "loss": 2.9841, "learning_rate": 5.319148936170213e-06, "epoch": 0.021266415014089, "percentage": 1.06, "elapsed_time": "0:08:25", "remaining_time": "13:04:13"}
+{"current_steps": 30, "total_steps": 2350, "loss": 2.8886, "learning_rate": 6.3829787234042555e-06, "epoch": 0.0255196980169068, "percentage": 1.28, "elapsed_time": "0:10:02", "remaining_time": "12:57:06"}
+{"current_steps": 35, "total_steps": 2350, "loss": 2.7333, "learning_rate": 7.446808510638298e-06, "epoch": 0.0297729810197246, "percentage": 1.49, "elapsed_time": "0:11:39", "remaining_time": "12:51:38"}
+{"current_steps": 40, "total_steps": 2350, "loss": 2.7314, "learning_rate": 8.510638297872341e-06, "epoch": 0.0340262640225424, "percentage": 1.7, "elapsed_time": "0:13:16", "remaining_time": "12:46:42"}
+{"current_steps": 45, "total_steps": 2350, "loss": 2.7024, "learning_rate": 9.574468085106383e-06, "epoch": 0.0382795470253602, "percentage": 1.91, "elapsed_time": "0:14:53", "remaining_time": "12:42:34"}
+{"current_steps": 50, "total_steps": 2350, "loss": 2.7374, "learning_rate": 1.0638297872340426e-05, "epoch": 0.042532830028178, "percentage": 2.13, "elapsed_time": "0:16:30", "remaining_time": "12:39:14"}
+{"current_steps": 55, "total_steps": 2350, "loss": 2.6406, "learning_rate": 1.170212765957447e-05, "epoch": 0.0467861130309958, "percentage": 2.34, "elapsed_time": "0:18:06", "remaining_time": "12:35:46"}
+{"current_steps": 60, "total_steps": 2350, "loss": 2.658, "learning_rate": 1.2765957446808511e-05, "epoch": 0.0510393960338136, "percentage": 2.55, "elapsed_time": "0:19:43", "remaining_time": "12:33:05"}
+{"current_steps": 65, "total_steps": 2350, "loss": 2.6501, "learning_rate": 1.3829787234042554e-05, "epoch": 0.0552926790366314, "percentage": 2.77, "elapsed_time": "0:21:20", "remaining_time": "12:30:16"}
+{"current_steps": 70, "total_steps": 2350, "loss": 2.6481, "learning_rate": 1.4893617021276596e-05, "epoch": 0.0595459620394492, "percentage": 2.98, "elapsed_time": "0:22:57", "remaining_time": "12:27:35"}
+{"current_steps": 75, "total_steps": 2350, "loss": 2.6195, "learning_rate": 1.595744680851064e-05, "epoch": 0.063799245042267, "percentage": 3.19, "elapsed_time": "0:24:33", "remaining_time": "12:24:58"}
+{"current_steps": 80, "total_steps": 2350, "loss": 2.6704, "learning_rate": 1.7021276595744682e-05, "epoch": 0.0680525280450848, "percentage": 3.4, "elapsed_time": "0:26:09", "remaining_time": "12:22:22"}
+{"current_steps": 85, "total_steps": 2350, "loss": 2.6392, "learning_rate": 1.8085106382978724e-05, "epoch": 0.0723058110479026, "percentage": 3.62, "elapsed_time": "0:27:45", "remaining_time": "12:19:48"}
+{"current_steps": 90, "total_steps": 2350, "loss": 2.6517, "learning_rate": 1.9148936170212766e-05, "epoch": 0.0765590940507204, "percentage": 3.83, "elapsed_time": "0:29:21", "remaining_time": "12:17:21"}
+{"current_steps": 95, "total_steps": 2350, "loss": 2.6277, "learning_rate": 2.0212765957446807e-05, "epoch": 0.0808123770535382, "percentage": 4.04, "elapsed_time": "0:30:58", "remaining_time": "12:15:07"}
+{"current_steps": 100, "total_steps": 2350, "loss": 2.6178, "learning_rate": 2.1276595744680852e-05, "epoch": 0.085065660056356, "percentage": 4.26, "elapsed_time": "0:32:34", "remaining_time": "12:12:48"}
+{"current_steps": 105, "total_steps": 2350, "loss": 2.6114, "learning_rate": 2.2340425531914894e-05, "epoch": 0.0893189430591738, "percentage": 4.47, "elapsed_time": "0:34:09", "remaining_time": "12:10:29"}
+{"current_steps": 110, "total_steps": 2350, "loss": 2.583, "learning_rate": 2.340425531914894e-05, "epoch": 0.0935722260619916, "percentage": 4.68, "elapsed_time": "0:35:46", "remaining_time": "12:08:23"}
+{"current_steps": 115, "total_steps": 2350, "loss": 2.585, "learning_rate": 2.446808510638298e-05, "epoch": 0.0978255090648094, "percentage": 4.89, "elapsed_time": "0:37:21", "remaining_time": "12:06:08"}
+{"current_steps": 120, "total_steps": 2350, "loss": 2.5928, "learning_rate": 2.5531914893617022e-05, "epoch": 0.1020787920676272, "percentage": 5.11, "elapsed_time": "0:38:57", "remaining_time": "12:04:07"}
+{"current_steps": 125, "total_steps": 2350, "loss": 2.5695, "learning_rate": 2.6595744680851064e-05, "epoch": 0.106332075070445, "percentage": 5.32, "elapsed_time": "0:40:33", "remaining_time": "12:02:03"}
+{"current_steps": 130, "total_steps": 2350, "loss": 2.5822, "learning_rate": 2.765957446808511e-05, "epoch": 0.1105853580732628, "percentage": 5.53, "elapsed_time": "0:42:09", "remaining_time": "11:59:58"}
+{"current_steps": 135, "total_steps": 2350, "loss": 2.6068, "learning_rate": 2.8723404255319154e-05, "epoch": 0.1148386410760806, "percentage": 5.74, "elapsed_time": "0:43:45", "remaining_time": "11:57:54"}
+{"current_steps": 140, "total_steps": 2350, "loss": 2.6135, "learning_rate": 2.9787234042553192e-05, "epoch": 0.1190919240788984, "percentage": 5.96, "elapsed_time": "0:45:21", "remaining_time": "11:55:56"}
+{"current_steps": 145, "total_steps": 2350, "loss": 2.6274, "learning_rate": 3.085106382978723e-05, "epoch": 0.1233452070817162, "percentage": 6.17, "elapsed_time": "0:46:57", "remaining_time": "11:54:01"}
+{"current_steps": 150, "total_steps": 2350, "loss": 2.5456, "learning_rate": 3.191489361702128e-05, "epoch": 0.127598490084534, "percentage": 6.38, "elapsed_time": "0:48:32", "remaining_time": "11:52:01"}
+{"current_steps": 155, "total_steps": 2350, "loss": 2.5587, "learning_rate": 3.2978723404255317e-05, "epoch": 0.1318517730873518, "percentage": 6.6, "elapsed_time": "0:50:08", "remaining_time": "11:50:03"}
+{"current_steps": 160, "total_steps": 2350, "loss": 2.5825, "learning_rate": 3.4042553191489365e-05, "epoch": 0.1361050560901696, "percentage": 6.81, "elapsed_time": "0:51:44", "remaining_time": "11:48:11"}
+{"current_steps": 165, "total_steps": 2350, "loss": 2.5997, "learning_rate": 3.5106382978723407e-05, "epoch": 0.1403583390929874, "percentage": 7.02, "elapsed_time": "0:53:20", "remaining_time": "11:46:27"}
+{"current_steps": 170, "total_steps": 2350, "loss": 2.5429, "learning_rate": 3.617021276595745e-05, "epoch": 0.1446116220958052, "percentage": 7.23, "elapsed_time": "0:54:57", "remaining_time": "11:44:42"}
+{"current_steps": 175, "total_steps": 2350, "loss": 2.5906, "learning_rate": 3.723404255319149e-05, "epoch": 0.148864905098623, "percentage": 7.45, "elapsed_time": "0:56:34", "remaining_time": "11:43:03"}
+{"current_steps": 180, "total_steps": 2350, "loss": 2.5513, "learning_rate": 3.829787234042553e-05, "epoch": 0.1531181881014408, "percentage": 7.66, "elapsed_time": "0:58:10", "remaining_time": "11:41:23"}
+{"current_steps": 185, "total_steps": 2350, "loss": 2.582, "learning_rate": 3.936170212765958e-05, "epoch": 0.1573714711042586, "percentage": 7.87, "elapsed_time": "0:59:46", "remaining_time": "11:39:33"}
+{"current_steps": 190, "total_steps": 2350, "loss": 2.6428, "learning_rate": 4.0425531914893614e-05, "epoch": 0.1616247541070764, "percentage": 8.09, "elapsed_time": "1:01:23", "remaining_time": "11:38:00"}
+{"current_steps": 195, "total_steps": 2350, "loss": 2.5858, "learning_rate": 4.148936170212766e-05, "epoch": 0.1658780371098942, "percentage": 8.3, "elapsed_time": "1:02:59", "remaining_time": "11:36:11"}
+{"current_steps": 200, "total_steps": 2350, "loss": 2.571, "learning_rate": 4.2553191489361704e-05, "epoch": 0.170131320112712, "percentage": 8.51, "elapsed_time": "1:04:35", "remaining_time": "11:34:23"}
+{"current_steps": 205, "total_steps": 2350, "loss": 2.5948, "learning_rate": 4.3617021276595746e-05, "epoch": 0.1743846031155298, "percentage": 8.72, "elapsed_time": "1:06:15", "remaining_time": "11:33:14"}
+{"current_steps": 210, "total_steps": 2350, "loss": 2.5763, "learning_rate": 4.468085106382979e-05, "epoch": 0.1786378861183476, "percentage": 8.94, "elapsed_time": "1:07:53", "remaining_time": "11:31:47"}
+{"current_steps": 215, "total_steps": 2350, "loss": 2.5975, "learning_rate": 4.574468085106383e-05, "epoch": 0.1828911691211654, "percentage": 9.15, "elapsed_time": "1:09:30", "remaining_time": "11:30:16"}
+{"current_steps": 220, "total_steps": 2350, "loss": 2.5627, "learning_rate": 4.680851063829788e-05, "epoch": 0.1871444521239832, "percentage": 9.36, "elapsed_time": "1:11:08", "remaining_time": "11:28:44"}
+{"current_steps": 225, "total_steps": 2350, "loss": 2.6303, "learning_rate": 4.787234042553192e-05, "epoch": 0.191397735126801, "percentage": 9.57, "elapsed_time": "1:12:45", "remaining_time": "11:27:08"}
+{"current_steps": 230, "total_steps": 2350, "loss": 2.614, "learning_rate": 4.893617021276596e-05, "epoch": 0.1956510181296188, "percentage": 9.79, "elapsed_time": "1:14:22", "remaining_time": "11:25:29"}
+{"current_steps": 235, "total_steps": 2350, "loss": 2.6077, "learning_rate": 5e-05, "epoch": 0.1999043011324366, "percentage": 10.0, "elapsed_time": "1:15:59", "remaining_time": "11:23:57"}
+{"current_steps": 240, "total_steps": 2350, "loss": 2.5803, "learning_rate": 4.988179669030733e-05, "epoch": 0.2041575841352544, "percentage": 10.21, "elapsed_time": "1:17:37", "remaining_time": "11:22:23"}
+{"current_steps": 245, "total_steps": 2350, "loss": 2.579, "learning_rate": 4.976359338061466e-05, "epoch": 0.2084108671380722, "percentage": 10.43, "elapsed_time": "1:19:13", "remaining_time": "11:20:43"}
+{"current_steps": 250, "total_steps": 2350, "loss": 2.5945, "learning_rate": 4.964539007092199e-05, "epoch": 0.21266415014089, "percentage": 10.64, "elapsed_time": "1:20:50", "remaining_time": "11:19:00"}
+{"current_steps": 255, "total_steps": 2350, "loss": 2.6439, "learning_rate": 4.9527186761229313e-05, "epoch": 0.2169174331437078, "percentage": 10.85, "elapsed_time": "1:22:26", "remaining_time": "11:17:20"}
+{"current_steps": 260, "total_steps": 2350, "loss": 2.5977, "learning_rate": 4.940898345153664e-05, "epoch": 0.2211707161465256, "percentage": 11.06, "elapsed_time": "1:24:03", "remaining_time": "11:15:40"}
+{"current_steps": 265, "total_steps": 2350, "loss": 2.6129, "learning_rate": 4.929078014184397e-05, "epoch": 0.2254239991493434, "percentage": 11.28, "elapsed_time": "1:25:39", "remaining_time": "11:13:56"}
+{"current_steps": 270, "total_steps": 2350, "loss": 2.5694, "learning_rate": 4.91725768321513e-05, "epoch": 0.2296772821521612, "percentage": 11.49, "elapsed_time": "1:27:15", "remaining_time": "11:12:15"}
+{"current_steps": 275, "total_steps": 2350, "loss": 2.5778, "learning_rate": 4.905437352245863e-05, "epoch": 0.233930565154979, "percentage": 11.7, "elapsed_time": "1:28:51", "remaining_time": "11:10:30"}
+{"current_steps": 280, "total_steps": 2350, "loss": 2.6525, "learning_rate": 4.893617021276596e-05, "epoch": 0.2381838481577968, "percentage": 11.91, "elapsed_time": "1:30:27", "remaining_time": "11:08:46"}
+{"current_steps": 285, "total_steps": 2350, "loss": 2.5845, "learning_rate": 4.8817966903073283e-05, "epoch": 0.2424371311606146, "percentage": 12.13, "elapsed_time": "1:32:03", "remaining_time": "11:07:03"}
+{"current_steps": 290, "total_steps": 2350, "loss": 2.6101, "learning_rate": 4.869976359338061e-05, "epoch": 0.2466904141634324, "percentage": 12.34, "elapsed_time": "1:33:39", "remaining_time": "11:05:18"}
+{"current_steps": 295, "total_steps": 2350, "loss": 2.5458, "learning_rate": 4.858156028368794e-05, "epoch": 0.2509436971662502, "percentage": 12.55, "elapsed_time": "1:35:15", "remaining_time": "11:03:37"}
+{"current_steps": 300, "total_steps": 2350, "loss": 2.5742, "learning_rate": 4.846335697399527e-05, "epoch": 0.255196980169068, "percentage": 12.77, "elapsed_time": "1:36:51", "remaining_time": "11:01:53"}
+{"current_steps": 305, "total_steps": 2350, "loss": 2.5727, "learning_rate": 4.83451536643026e-05, "epoch": 0.2594502631718858, "percentage": 12.98, "elapsed_time": "1:38:28", "remaining_time": "11:00:15"}
+{"current_steps": 310, "total_steps": 2350, "loss": 2.5882, "learning_rate": 4.822695035460993e-05, "epoch": 0.2637035461747036, "percentage": 13.19, "elapsed_time": "1:40:03", "remaining_time": "10:58:29"}
+{"current_steps": 315, "total_steps": 2350, "loss": 2.5671, "learning_rate": 4.810874704491726e-05, "epoch": 0.2679568291775214, "percentage": 13.4, "elapsed_time": "1:41:40", "remaining_time": "10:56:51"}
+{"current_steps": 320, "total_steps": 2350, "loss": 2.622, "learning_rate": 4.799054373522459e-05, "epoch": 0.2722101121803392, "percentage": 13.62, "elapsed_time": "1:43:16", "remaining_time": "10:55:07"}
+{"current_steps": 325, "total_steps": 2350, "loss": 2.5931, "learning_rate": 4.787234042553192e-05, "epoch": 0.276463395183157, "percentage": 13.83, "elapsed_time": "1:44:52", "remaining_time": "10:53:27"}
+{"current_steps": 330, "total_steps": 2350, "loss": 2.5471, "learning_rate": 4.775413711583925e-05, "epoch": 0.2807166781859748, "percentage": 14.04, "elapsed_time": "1:46:28", "remaining_time": "10:51:47"}
+{"current_steps": 335, "total_steps": 2350, "loss": 2.5485, "learning_rate": 4.763593380614658e-05, "epoch": 0.2849699611887926, "percentage": 14.26, "elapsed_time": "1:48:05", "remaining_time": "10:50:06"}
+{"current_steps": 340, "total_steps": 2350, "loss": 2.5652, "learning_rate": 4.751773049645391e-05, "epoch": 0.2892232441916104, "percentage": 14.47, "elapsed_time": "1:49:40", "remaining_time": "10:48:25"}
+{"current_steps": 345, "total_steps": 2350, "loss": 2.617, "learning_rate": 4.739952718676123e-05, "epoch": 0.2934765271944282, "percentage": 14.68, "elapsed_time": "1:51:16", "remaining_time": "10:46:41"}
+{"current_steps": 350, "total_steps": 2350, "loss": 2.5512, "learning_rate": 4.728132387706856e-05, "epoch": 0.297729810197246, "percentage": 14.89, "elapsed_time": "1:52:52", "remaining_time": "10:44:59"}
+{"current_steps": 355, "total_steps": 2350, "loss": 2.5682, "learning_rate": 4.716312056737589e-05, "epoch": 0.3019830932000638, "percentage": 15.11, "elapsed_time": "1:54:28", "remaining_time": "10:43:16"}
+{"current_steps": 360, "total_steps": 2350, "loss": 2.5709, "learning_rate": 4.704491725768322e-05, "epoch": 0.3062363762028816, "percentage": 15.32, "elapsed_time": "1:56:04", "remaining_time": "10:41:35"}
+{"current_steps": 365, "total_steps": 2350, "loss": 2.5834, "learning_rate": 4.692671394799055e-05, "epoch": 0.3104896592056994, "percentage": 15.53, "elapsed_time": "1:57:39", "remaining_time": "10:39:53"}
+{"current_steps": 370, "total_steps": 2350, "loss": 2.5736, "learning_rate": 4.680851063829788e-05, "epoch": 0.3147429422085172, "percentage": 15.74, "elapsed_time": "1:59:15", "remaining_time": "10:38:10"}
+{"current_steps": 375, "total_steps": 2350, "loss": 2.5321, "learning_rate": 4.669030732860521e-05, "epoch": 0.31899622521133497, "percentage": 15.96, "elapsed_time": "2:00:50", "remaining_time": "10:36:28"}
+{"current_steps": 380, "total_steps": 2350, "loss": 2.5701, "learning_rate": 4.657210401891253e-05, "epoch": 0.3232495082141528, "percentage": 16.17, "elapsed_time": "2:02:27", "remaining_time": "10:34:48"}
+{"current_steps": 385, "total_steps": 2350, "loss": 2.553, "learning_rate": 4.645390070921986e-05, "epoch": 0.3275027912169706, "percentage": 16.38, "elapsed_time": "2:04:03", "remaining_time": "10:33:09"}
+{"current_steps": 390, "total_steps": 2350, "loss": 2.5242, "learning_rate": 4.633569739952719e-05, "epoch": 0.3317560742197884, "percentage": 16.6, "elapsed_time": "2:05:38", "remaining_time": "10:31:26"}
+{"current_steps": 395, "total_steps": 2350, "loss": 2.5279, "learning_rate": 4.621749408983452e-05, "epoch": 0.3360093572226062, "percentage": 16.81, "elapsed_time": "2:07:15", "remaining_time": "10:29:50"}
+{"current_steps": 400, "total_steps": 2350, "loss": 2.607, "learning_rate": 4.609929078014185e-05, "epoch": 0.340262640225424, "percentage": 17.02, "elapsed_time": "2:08:51", "remaining_time": "10:28:09"}
+{"current_steps": 405, "total_steps": 2350, "loss": 2.5556, "learning_rate": 4.598108747044918e-05, "epoch": 0.3445159232282418, "percentage": 17.23, "elapsed_time": "2:10:29", "remaining_time": "10:26:40"}
+{"current_steps": 410, "total_steps": 2350, "loss": 2.5468, "learning_rate": 4.58628841607565e-05, "epoch": 0.3487692062310596, "percentage": 17.45, "elapsed_time": "2:12:05", "remaining_time": "10:25:02"}
+{"current_steps": 415, "total_steps": 2350, "loss": 2.5511, "learning_rate": 4.574468085106383e-05, "epoch": 0.3530224892338774, "percentage": 17.66, "elapsed_time": "2:13:41", "remaining_time": "10:23:22"}
+{"current_steps": 420, "total_steps": 2350, "loss": 2.5438, "learning_rate": 4.562647754137116e-05, "epoch": 0.3572757722366952, "percentage": 17.87, "elapsed_time": "2:15:17", "remaining_time": "10:21:41"}
+{"current_steps": 425, "total_steps": 2350, "loss": 2.5264, "learning_rate": 4.550827423167849e-05, "epoch": 0.361529055239513, "percentage": 18.09, "elapsed_time": "2:16:52", "remaining_time": "10:19:59"}
+{"current_steps": 430, "total_steps": 2350, "loss": 2.5365, "learning_rate": 4.539007092198582e-05, "epoch": 0.3657823382423308, "percentage": 18.3, "elapsed_time": "2:18:28", "remaining_time": "10:18:17"}
+{"current_steps": 435, "total_steps": 2350, "loss": 2.5596, "learning_rate": 4.527186761229315e-05, "epoch": 0.3700356212451486, "percentage": 18.51, "elapsed_time": "2:20:03", "remaining_time": "10:16:36"}
+{"current_steps": 440, "total_steps": 2350, "loss": 2.4894, "learning_rate": 4.5153664302600477e-05, "epoch": 0.3742889042479664, "percentage": 18.72, "elapsed_time": "2:21:39", "remaining_time": "10:14:53"}
+{"current_steps": 445, "total_steps": 2350, "loss": 2.5441, "learning_rate": 4.50354609929078e-05, "epoch": 0.3785421872507842, "percentage": 18.94, "elapsed_time": "2:23:14", "remaining_time": "10:13:12"}
+{"current_steps": 450, "total_steps": 2350, "loss": 2.5379, "learning_rate": 4.491725768321513e-05, "epoch": 0.382795470253602, "percentage": 19.15, "elapsed_time": "2:24:49", "remaining_time": "10:11:30"}
+{"current_steps": 455, "total_steps": 2350, "loss": 2.5194, "learning_rate": 4.479905437352246e-05, "epoch": 0.3870487532564198, "percentage": 19.36, "elapsed_time": "2:26:25", "remaining_time": "10:09:49"}
+{"current_steps": 460, "total_steps": 2350, "loss": 2.5767, "learning_rate": 4.468085106382979e-05, "epoch": 0.3913020362592376, "percentage": 19.57, "elapsed_time": "2:28:00", "remaining_time": "10:08:08"}
+{"current_steps": 465, "total_steps": 2350, "loss": 2.5485, "learning_rate": 4.456264775413712e-05, "epoch": 0.3955553192620554, "percentage": 19.79, "elapsed_time": "2:29:37", "remaining_time": "10:06:30"}
+{"current_steps": 470, "total_steps": 2350, "loss": 2.547, "learning_rate": 4.4444444444444447e-05, "epoch": 0.3998086022648732, "percentage": 20.0, "elapsed_time": "2:31:12", "remaining_time": "10:04:51"}
+{"current_steps": 475, "total_steps": 2350, "loss": 2.5278, "learning_rate": 4.432624113475177e-05, "epoch": 0.404061885267691, "percentage": 20.21, "elapsed_time": "2:32:48", "remaining_time": "10:03:09"}
+{"current_steps": 480, "total_steps": 2350, "loss": 2.5245, "learning_rate": 4.4208037825059106e-05, "epoch": 0.4083151682705088, "percentage": 20.43, "elapsed_time": "2:34:23", "remaining_time": "10:01:29"}
+{"current_steps": 485, "total_steps": 2350, "loss": 2.5318, "learning_rate": 4.4089834515366435e-05, "epoch": 0.4125684512733266, "percentage": 20.64, "elapsed_time": "2:35:59", "remaining_time": "9:59:52"}
+{"current_steps": 490, "total_steps": 2350, "loss": 2.4747, "learning_rate": 4.3971631205673764e-05, "epoch": 0.4168217342761444, "percentage": 20.85, "elapsed_time": "2:37:35", "remaining_time": "9:58:12"}
+{"current_steps": 495, "total_steps": 2350, "loss": 2.4824, "learning_rate": 4.3853427895981094e-05, "epoch": 0.4210750172789622, "percentage": 21.06, "elapsed_time": "2:39:11", "remaining_time": "9:56:33"}
+{"current_steps": 500, "total_steps": 2350, "loss": 2.5236, "learning_rate": 4.373522458628842e-05, "epoch": 0.42532830028178, "percentage": 21.28, "elapsed_time": "2:40:46", "remaining_time": "9:54:53"}
+{"current_steps": 505, "total_steps": 2350, "loss": 2.5287, "learning_rate": 4.3617021276595746e-05, "epoch": 0.4295815832845978, "percentage": 21.49, "elapsed_time": "2:42:23", "remaining_time": "9:53:16"}
+{"current_steps": 510, "total_steps": 2350, "loss": 2.4732, "learning_rate": 4.3498817966903076e-05, "epoch": 0.4338348662874156, "percentage": 21.7, "elapsed_time": "2:44:01", "remaining_time": "9:51:46"}
+{"current_steps": 515, "total_steps": 2350, "loss": 2.5246, "learning_rate": 4.3380614657210405e-05, "epoch": 0.4380881492902334, "percentage": 21.91, "elapsed_time": "2:45:39", "remaining_time": "9:50:15"}
+{"current_steps": 520, "total_steps": 2350, "loss": 2.4755, "learning_rate": 4.3262411347517734e-05, "epoch": 0.4423414322930512, "percentage": 22.13, "elapsed_time": "2:47:16", "remaining_time": "9:48:40"}
+{"current_steps": 525, "total_steps": 2350, "loss": 2.5074, "learning_rate": 4.3144208037825064e-05, "epoch": 0.44659471529586897, "percentage": 22.34, "elapsed_time": "2:48:53", "remaining_time": "9:47:04"}
+{"current_steps": 530, "total_steps": 2350, "loss": 2.5037, "learning_rate": 4.302600472813239e-05, "epoch": 0.4508479982986868, "percentage": 22.55, "elapsed_time": "2:50:29", "remaining_time": "9:45:28"}
+{"current_steps": 535, "total_steps": 2350, "loss": 2.5041, "learning_rate": 4.2907801418439716e-05, "epoch": 0.4551012813015046, "percentage": 22.77, "elapsed_time": "2:52:06", "remaining_time": "9:43:51"}
+{"current_steps": 540, "total_steps": 2350, "loss": 2.5235, "learning_rate": 4.2789598108747046e-05, "epoch": 0.4593545643043224, "percentage": 22.98, "elapsed_time": "2:53:42", "remaining_time": "9:42:14"}
+{"current_steps": 545, "total_steps": 2350, "loss": 2.5049, "learning_rate": 4.2671394799054375e-05, "epoch": 0.4636078473071402, "percentage": 23.19, "elapsed_time": "2:55:19", "remaining_time": "9:40:38"}
+{"current_steps": 550, "total_steps": 2350, "loss": 2.5086, "learning_rate": 4.2553191489361704e-05, "epoch": 0.467861130309958, "percentage": 23.4, "elapsed_time": "2:56:55", "remaining_time": "9:39:02"}
+{"current_steps": 555, "total_steps": 2350, "loss": 2.5231, "learning_rate": 4.2434988179669034e-05, "epoch": 0.4721144133127758, "percentage": 23.62, "elapsed_time": "2:58:31", "remaining_time": "9:37:24"}
+{"current_steps": 560, "total_steps": 2350, "loss": 2.4703, "learning_rate": 4.231678486997636e-05, "epoch": 0.4763676963155936, "percentage": 23.83, "elapsed_time": "3:00:07", "remaining_time": "9:35:46"}
+{"current_steps": 565, "total_steps": 2350, "loss": 2.4774, "learning_rate": 4.219858156028369e-05, "epoch": 0.4806209793184114, "percentage": 24.04, "elapsed_time": "3:01:43", "remaining_time": "9:34:08"}
+{"current_steps": 570, "total_steps": 2350, "loss": 2.5054, "learning_rate": 4.2080378250591016e-05, "epoch": 0.4848742623212292, "percentage": 24.26, "elapsed_time": "3:03:19", "remaining_time": "9:32:29"}
+{"current_steps": 575, "total_steps": 2350, "loss": 2.4538, "learning_rate": 4.1962174940898345e-05, "epoch": 0.489127545324047, "percentage": 24.47, "elapsed_time": "3:04:55", "remaining_time": "9:30:52"}
+{"current_steps": 580, "total_steps": 2350, "loss": 2.5135, "learning_rate": 4.1843971631205674e-05, "epoch": 0.4933808283268648, "percentage": 24.68, "elapsed_time": "3:06:31", "remaining_time": "9:29:14"}
+{"current_steps": 585, "total_steps": 2350, "loss": 2.4991, "learning_rate": 4.1725768321513004e-05, "epoch": 0.4976341113296826, "percentage": 24.89, "elapsed_time": "3:08:07", "remaining_time": "9:27:36"}
+{"current_steps": 590, "total_steps": 2350, "loss": 2.522, "learning_rate": 4.1607565011820333e-05, "epoch": 0.5018873943325004, "percentage": 25.11, "elapsed_time": "3:09:44", "remaining_time": "9:25:59"}
+{"current_steps": 595, "total_steps": 2350, "loss": 2.4663, "learning_rate": 4.148936170212766e-05, "epoch": 0.5061406773353182, "percentage": 25.32, "elapsed_time": "3:11:21", "remaining_time": "9:24:24"}
+{"current_steps": 600, "total_steps": 2350, "loss": 2.4471, "learning_rate": 4.1371158392434986e-05, "epoch": 0.510393960338136, "percentage": 25.53, "elapsed_time": "3:12:57", "remaining_time": "9:22:47"}
+{"current_steps": 605, "total_steps": 2350, "loss": 2.5032, "learning_rate": 4.1252955082742315e-05, "epoch": 0.5146472433409538, "percentage": 25.74, "elapsed_time": "3:14:33", "remaining_time": "9:21:09"}
+{"current_steps": 610, "total_steps": 2350, "loss": 2.4995, "learning_rate": 4.1134751773049644e-05, "epoch": 0.5189005263437716, "percentage": 25.96, "elapsed_time": "3:16:11", "remaining_time": "9:19:36"}
+{"current_steps": 615, "total_steps": 2350, "loss": 2.4837, "learning_rate": 4.1016548463356974e-05, "epoch": 0.5231538093465894, "percentage": 26.17, "elapsed_time": "3:17:47", "remaining_time": "9:17:59"}
+{"current_steps": 620, "total_steps": 2350, "loss": 2.484, "learning_rate": 4.0898345153664303e-05, "epoch": 0.5274070923494072, "percentage": 26.38, "elapsed_time": "3:19:23", "remaining_time": "9:16:23"}
+{"current_steps": 625, "total_steps": 2350, "loss": 2.4921, "learning_rate": 4.078014184397163e-05, "epoch": 0.531660375352225, "percentage": 26.6, "elapsed_time": "3:20:59", "remaining_time": "9:14:44"}
+{"current_steps": 630, "total_steps": 2350, "loss": 2.5048, "learning_rate": 4.066193853427896e-05, "epoch": 0.5359136583550428, "percentage": 26.81, "elapsed_time": "3:22:35", "remaining_time": "9:13:05"}
+{"current_steps": 635, "total_steps": 2350, "loss": 2.4843, "learning_rate": 4.0543735224586285e-05, "epoch": 0.5401669413578606, "percentage": 27.02, "elapsed_time": "3:24:11", "remaining_time": "9:11:29"}
+{"current_steps": 640, "total_steps": 2350, "loss": 2.4262, "learning_rate": 4.0425531914893614e-05, "epoch": 0.5444202243606784, "percentage": 27.23, "elapsed_time": "3:25:48", "remaining_time": "9:09:52"}
+{"current_steps": 645, "total_steps": 2350, "loss": 2.4656, "learning_rate": 4.030732860520095e-05, "epoch": 0.5486735073634962, "percentage": 27.45, "elapsed_time": "3:27:24", "remaining_time": "9:08:15"}
+{"current_steps": 650, "total_steps": 2350, "loss": 2.4967, "learning_rate": 4.018912529550828e-05, "epoch": 0.552926790366314, "percentage": 27.66, "elapsed_time": "3:29:00", "remaining_time": "9:06:38"}
+{"current_steps": 655, "total_steps": 2350, "loss": 2.4475, "learning_rate": 4.007092198581561e-05, "epoch": 0.5571800733691318, "percentage": 27.87, "elapsed_time": "3:30:36", "remaining_time": "9:05:01"}
+{"current_steps": 660, "total_steps": 2350, "loss": 2.4688, "learning_rate": 3.995271867612293e-05, "epoch": 0.5614333563719496, "percentage": 28.09, "elapsed_time": "3:32:12", "remaining_time": "9:03:23"}
+{"current_steps": 665, "total_steps": 2350, "loss": 2.492, "learning_rate": 3.983451536643026e-05, "epoch": 0.5656866393747674, "percentage": 28.3, "elapsed_time": "3:33:48", "remaining_time": "9:01:45"}
+{"current_steps": 670, "total_steps": 2350, "loss": 2.4977, "learning_rate": 3.971631205673759e-05, "epoch": 0.5699399223775852, "percentage": 28.51, "elapsed_time": "3:35:23", "remaining_time": "9:00:06"}
+{"current_steps": 675, "total_steps": 2350, "loss": 2.4755, "learning_rate": 3.959810874704492e-05, "epoch": 0.574193205380403, "percentage": 28.72, "elapsed_time": "3:37:01", "remaining_time": "8:58:31"}
+{"current_steps": 680, "total_steps": 2350, "loss": 2.513, "learning_rate": 3.947990543735225e-05, "epoch": 0.5784464883832208, "percentage": 28.94, "elapsed_time": "3:38:36", "remaining_time": "8:56:53"}
+{"current_steps": 685, "total_steps": 2350, "loss": 2.4759, "learning_rate": 3.936170212765958e-05, "epoch": 0.5826997713860386, "percentage": 29.15, "elapsed_time": "3:40:12", "remaining_time": "8:55:15"}
+{"current_steps": 690, "total_steps": 2350, "loss": 2.4487, "learning_rate": 3.924349881796691e-05, "epoch": 0.5869530543888564, "percentage": 29.36, "elapsed_time": "3:41:48", "remaining_time": "8:53:36"}
+{"current_steps": 695, "total_steps": 2350, "loss": 2.4394, "learning_rate": 3.912529550827423e-05, "epoch": 0.5912063373916742, "percentage": 29.57, "elapsed_time": "3:43:23", "remaining_time": "8:51:56"}
+{"current_steps": 700, "total_steps": 2350, "loss": 2.4777, "learning_rate": 3.900709219858156e-05, "epoch": 0.595459620394492, "percentage": 29.79, "elapsed_time": "3:44:58", "remaining_time": "8:50:18"}
+{"current_steps": 705, "total_steps": 2350, "loss": 2.4709, "learning_rate": 3.888888888888889e-05, "epoch": 0.5997129033973098, "percentage": 30.0, "elapsed_time": "3:46:34", "remaining_time": "8:48:39"}
+{"current_steps": 710, "total_steps": 2350, "loss": 2.4775, "learning_rate": 3.877068557919622e-05, "epoch": 0.6039661864001276, "percentage": 30.21, "elapsed_time": "3:48:09", "remaining_time": "8:47:01"}
+{"current_steps": 715, "total_steps": 2350, "loss": 2.4405, "learning_rate": 3.865248226950355e-05, "epoch": 0.6082194694029454, "percentage": 30.43, "elapsed_time": "3:49:44", "remaining_time": "8:45:22"}
+{"current_steps": 720, "total_steps": 2350, "loss": 2.3798, "learning_rate": 3.853427895981088e-05, "epoch": 0.6124727524057632, "percentage": 30.64, "elapsed_time": "3:51:19", "remaining_time": "8:43:42"}
+{"current_steps": 725, "total_steps": 2350, "loss": 2.4553, "learning_rate": 3.84160756501182e-05, "epoch": 0.616726035408581, "percentage": 30.85, "elapsed_time": "3:52:55", "remaining_time": "8:42:03"}
+{"current_steps": 730, "total_steps": 2350, "loss": 2.4596, "learning_rate": 3.829787234042553e-05, "epoch": 0.6209793184113988, "percentage": 31.06, "elapsed_time": "3:54:30", "remaining_time": "8:40:25"}
+{"current_steps": 735, "total_steps": 2350, "loss": 2.494, "learning_rate": 3.817966903073286e-05, "epoch": 0.6252326014142165, "percentage": 31.28, "elapsed_time": "3:56:05", "remaining_time": "8:38:46"}
+{"current_steps": 740, "total_steps": 2350, "loss": 2.3969, "learning_rate": 3.806146572104019e-05, "epoch": 0.6294858844170343, "percentage": 31.49, "elapsed_time": "3:57:41", "remaining_time": "8:37:08"}
+{"current_steps": 745, "total_steps": 2350, "loss": 2.4438, "learning_rate": 3.794326241134752e-05, "epoch": 0.6337391674198521, "percentage": 31.7, "elapsed_time": "3:59:16", "remaining_time": "8:35:28"}
+{"current_steps": 750, "total_steps": 2350, "loss": 2.4191, "learning_rate": 3.782505910165485e-05, "epoch": 0.6379924504226699, "percentage": 31.91, "elapsed_time": "4:00:51", "remaining_time": "8:33:50"}
+{"current_steps": 755, "total_steps": 2350, "loss": 2.4433, "learning_rate": 3.770685579196218e-05, "epoch": 0.6422457334254879, "percentage": 32.13, "elapsed_time": "4:02:27", "remaining_time": "8:32:11"}
+{"current_steps": 760, "total_steps": 2350, "loss": 2.4597, "learning_rate": 3.75886524822695e-05, "epoch": 0.6464990164283057, "percentage": 32.34, "elapsed_time": "4:04:02", "remaining_time": "8:30:34"}
+{"current_steps": 765, "total_steps": 2350, "loss": 2.4633, "learning_rate": 3.747044917257683e-05, "epoch": 0.6507522994311234, "percentage": 32.55, "elapsed_time": "4:05:38", "remaining_time": "8:28:56"}
+{"current_steps": 770, "total_steps": 2350, "loss": 2.4522, "learning_rate": 3.735224586288416e-05, "epoch": 0.6550055824339412, "percentage": 32.77, "elapsed_time": "4:07:13", "remaining_time": "8:27:17"}
+{"current_steps": 775, "total_steps": 2350, "loss": 2.4274, "learning_rate": 3.723404255319149e-05, "epoch": 0.659258865436759, "percentage": 32.98, "elapsed_time": "4:08:48", "remaining_time": "8:25:38"}
+{"current_steps": 780, "total_steps": 2350, "loss": 2.4793, "learning_rate": 3.711583924349882e-05, "epoch": 0.6635121484395768, "percentage": 33.19, "elapsed_time": "4:10:23", "remaining_time": "8:24:00"}
+{"current_steps": 785, "total_steps": 2350, "loss": 2.4104, "learning_rate": 3.699763593380615e-05, "epoch": 0.6677654314423946, "percentage": 33.4, "elapsed_time": "4:11:58", "remaining_time": "8:22:21"}
+{"current_steps": 790, "total_steps": 2350, "loss": 2.4158, "learning_rate": 3.687943262411347e-05, "epoch": 0.6720187144452124, "percentage": 33.62, "elapsed_time": "4:13:34", "remaining_time": "8:20:44"}
+{"current_steps": 795, "total_steps": 2350, "loss": 2.4096, "learning_rate": 3.67612293144208e-05, "epoch": 0.6762719974480302, "percentage": 33.83, "elapsed_time": "4:15:10", "remaining_time": "8:19:07"}
+{"current_steps": 800, "total_steps": 2350, "loss": 2.4041, "learning_rate": 3.664302600472813e-05, "epoch": 0.680525280450848, "percentage": 34.04, "elapsed_time": "4:16:46", "remaining_time": "8:17:30"}
+{"current_steps": 805, "total_steps": 2350, "loss": 2.4548, "learning_rate": 3.6524822695035466e-05, "epoch": 0.6847785634536658, "percentage": 34.26, "elapsed_time": "4:18:22", "remaining_time": "8:15:52"}
+{"current_steps": 810, "total_steps": 2350, "loss": 2.395, "learning_rate": 3.6406619385342796e-05, "epoch": 0.6890318464564836, "percentage": 34.47, "elapsed_time": "4:19:57", "remaining_time": "8:14:14"}
+{"current_steps": 815, "total_steps": 2350, "loss": 2.4343, "learning_rate": 3.6288416075650125e-05, "epoch": 0.6932851294593014, "percentage": 34.68, "elapsed_time": "4:21:35", "remaining_time": "8:12:40"}
+{"current_steps": 820, "total_steps": 2350, "loss": 2.4177, "learning_rate": 3.617021276595745e-05, "epoch": 0.6975384124621192, "percentage": 34.89, "elapsed_time": "4:23:10", "remaining_time": "8:11:02"}
+{"current_steps": 825, "total_steps": 2350, "loss": 2.3848, "learning_rate": 3.605200945626478e-05, "epoch": 0.701791695464937, "percentage": 35.11, "elapsed_time": "4:24:45", "remaining_time": "8:09:24"}
+{"current_steps": 830, "total_steps": 2350, "loss": 2.4164, "learning_rate": 3.593380614657211e-05, "epoch": 0.7060449784677548, "percentage": 35.32, "elapsed_time": "4:26:22", "remaining_time": "8:07:48"}
+{"current_steps": 835, "total_steps": 2350, "loss": 2.4032, "learning_rate": 3.5815602836879437e-05, "epoch": 0.7102982614705726, "percentage": 35.53, "elapsed_time": "4:27:58", "remaining_time": "8:06:12"}
+{"current_steps": 840, "total_steps": 2350, "loss": 2.3989, "learning_rate": 3.5697399527186766e-05, "epoch": 0.7145515444733904, "percentage": 35.74, "elapsed_time": "4:29:33", "remaining_time": "8:04:34"}
+{"current_steps": 845, "total_steps": 2350, "loss": 2.3684, "learning_rate": 3.5579196217494095e-05, "epoch": 0.7188048274762082, "percentage": 35.96, "elapsed_time": "4:31:10", "remaining_time": "8:02:59"}
+{"current_steps": 850, "total_steps": 2350, "loss": 2.4081, "learning_rate": 3.546099290780142e-05, "epoch": 0.723058110479026, "percentage": 36.17, "elapsed_time": "4:32:47", "remaining_time": "8:01:24"}
+{"current_steps": 855, "total_steps": 2350, "loss": 2.3928, "learning_rate": 3.534278959810875e-05, "epoch": 0.7273113934818438, "percentage": 36.38, "elapsed_time": "4:34:24", "remaining_time": "7:59:48"}
+{"current_steps": 860, "total_steps": 2350, "loss": 2.4376, "learning_rate": 3.522458628841608e-05, "epoch": 0.7315646764846616, "percentage": 36.6, "elapsed_time": "4:36:00", "remaining_time": "7:58:12"}
+{"current_steps": 865, "total_steps": 2350, "loss": 2.4199, "learning_rate": 3.5106382978723407e-05, "epoch": 0.7358179594874794, "percentage": 36.81, "elapsed_time": "4:37:36", "remaining_time": "7:56:34"}
+{"current_steps": 870, "total_steps": 2350, "loss": 2.4299, "learning_rate": 3.4988179669030736e-05, "epoch": 0.7400712424902972, "percentage": 37.02, "elapsed_time": "4:39:12", "remaining_time": "7:54:58"}
+{"current_steps": 875, "total_steps": 2350, "loss": 2.3908, "learning_rate": 3.4869976359338065e-05, "epoch": 0.744324525493115, "percentage": 37.23, "elapsed_time": "4:40:48", "remaining_time": "7:53:21"}
+{"current_steps": 880, "total_steps": 2350, "loss": 2.3904, "learning_rate": 3.4751773049645395e-05, "epoch": 0.7485778084959328, "percentage": 37.45, "elapsed_time": "4:42:25", "remaining_time": "7:51:46"}
+{"current_steps": 885, "total_steps": 2350, "loss": 2.3385, "learning_rate": 3.463356973995272e-05, "epoch": 0.7528310914987506, "percentage": 37.66, "elapsed_time": "4:44:00", "remaining_time": "7:50:08"}
+{"current_steps": 890, "total_steps": 2350, "loss": 2.3727, "learning_rate": 3.451536643026005e-05, "epoch": 0.7570843745015684, "percentage": 37.87, "elapsed_time": "4:45:36", "remaining_time": "7:48:31"}
+{"current_steps": 895, "total_steps": 2350, "loss": 2.3772, "learning_rate": 3.4397163120567377e-05, "epoch": 0.7613376575043862, "percentage": 38.09, "elapsed_time": "4:47:12", "remaining_time": "7:46:54"}
+{"current_steps": 900, "total_steps": 2350, "loss": 2.4052, "learning_rate": 3.4278959810874706e-05, "epoch": 0.765590940507204, "percentage": 38.3, "elapsed_time": "4:48:47", "remaining_time": "7:45:17"}
+{"current_steps": 905, "total_steps": 2350, "loss": 2.3386, "learning_rate": 3.4160756501182035e-05, "epoch": 0.7698442235100218, "percentage": 38.51, "elapsed_time": "4:50:23", "remaining_time": "7:43:40"}
+{"current_steps": 910, "total_steps": 2350, "loss": 2.3723, "learning_rate": 3.4042553191489365e-05, "epoch": 0.7740975065128396, "percentage": 38.72, "elapsed_time": "4:51:59", "remaining_time": "7:42:03"}
+{"current_steps": 915, "total_steps": 2350, "loss": 2.3892, "learning_rate": 3.392434988179669e-05, "epoch": 0.7783507895156574, "percentage": 38.94, "elapsed_time": "4:53:36", "remaining_time": "7:40:27"}
+{"current_steps": 920, "total_steps": 2350, "loss": 2.4115, "learning_rate": 3.380614657210402e-05, "epoch": 0.7826040725184752, "percentage": 39.15, "elapsed_time": "4:55:11", "remaining_time": "7:38:50"}
+{"current_steps": 925, "total_steps": 2350, "loss": 2.377, "learning_rate": 3.3687943262411347e-05, "epoch": 0.786857355521293, "percentage": 39.36, "elapsed_time": "4:56:47", "remaining_time": "7:37:13"}
+{"current_steps": 930, "total_steps": 2350, "loss": 2.3725, "learning_rate": 3.3569739952718676e-05, "epoch": 0.7911106385241108, "percentage": 39.57, "elapsed_time": "4:58:24", "remaining_time": "7:35:37"}
+{"current_steps": 935, "total_steps": 2350, "loss": 2.3879, "learning_rate": 3.3451536643026005e-05, "epoch": 0.7953639215269286, "percentage": 39.79, "elapsed_time": "5:00:00", "remaining_time": "7:34:01"}
+{"current_steps": 940, "total_steps": 2350, "loss": 2.3657, "learning_rate": 3.3333333333333335e-05, "epoch": 0.7996172045297464, "percentage": 40.0, "elapsed_time": "5:01:36", "remaining_time": "7:32:24"}
+{"current_steps": 945, "total_steps": 2350, "loss": 2.3563, "learning_rate": 3.3215130023640664e-05, "epoch": 0.8038704875325642, "percentage": 40.21, "elapsed_time": "5:03:12", "remaining_time": "7:30:48"}
+{"current_steps": 950, "total_steps": 2350, "loss": 2.3464, "learning_rate": 3.309692671394799e-05, "epoch": 0.808123770535382, "percentage": 40.43, "elapsed_time": "5:04:48", "remaining_time": "7:29:11"}
+{"current_steps": 955, "total_steps": 2350, "loss": 2.4217, "learning_rate": 3.2978723404255317e-05, "epoch": 0.8123770535381998, "percentage": 40.64, "elapsed_time": "5:06:24", "remaining_time": "7:27:34"}
+{"current_steps": 960, "total_steps": 2350, "loss": 2.3891, "learning_rate": 3.2860520094562646e-05, "epoch": 0.8166303365410176, "percentage": 40.85, "elapsed_time": "5:07:59", "remaining_time": "7:25:57"}
+{"current_steps": 965, "total_steps": 2350, "loss": 2.3703, "learning_rate": 3.2742316784869975e-05, "epoch": 0.8208836195438354, "percentage": 41.06, "elapsed_time": "5:09:35", "remaining_time": "7:24:20"}
+{"current_steps": 970, "total_steps": 2350, "loss": 2.3914, "learning_rate": 3.262411347517731e-05, "epoch": 0.8251369025466532, "percentage": 41.28, "elapsed_time": "5:11:11", "remaining_time": "7:22:43"}
+{"current_steps": 975, "total_steps": 2350, "loss": 2.3948, "learning_rate": 3.2505910165484634e-05, "epoch": 0.829390185549471, "percentage": 41.49, "elapsed_time": "5:12:47", "remaining_time": "7:21:06"}
+{"current_steps": 980, "total_steps": 2350, "loss": 2.397, "learning_rate": 3.2387706855791964e-05, "epoch": 0.8336434685522888, "percentage": 41.7, "elapsed_time": "5:14:22", "remaining_time": "7:19:29"}
+{"current_steps": 985, "total_steps": 2350, "loss": 2.379, "learning_rate": 3.226950354609929e-05, "epoch": 0.8378967515551066, "percentage": 41.91, "elapsed_time": "5:15:57", "remaining_time": "7:17:51"}
+{"current_steps": 990, "total_steps": 2350, "loss": 2.366, "learning_rate": 3.215130023640662e-05, "epoch": 0.8421500345579244, "percentage": 42.13, "elapsed_time": "5:17:32", "remaining_time": "7:16:13"}
+{"current_steps": 995, "total_steps": 2350, "loss": 2.3653, "learning_rate": 3.203309692671395e-05, "epoch": 0.8464033175607422, "percentage": 42.34, "elapsed_time": "5:19:08", "remaining_time": "7:14:37"}
+{"current_steps": 1000, "total_steps": 2350, "loss": 2.3757, "learning_rate": 3.191489361702128e-05, "epoch": 0.85065660056356, "percentage": 42.55, "elapsed_time": "5:20:44", "remaining_time": "7:12:59"}
+{"current_steps": 1005, "total_steps": 2350, "loss": 2.3504, "learning_rate": 3.179669030732861e-05, "epoch": 0.8549098835663778, "percentage": 42.77, "elapsed_time": "5:24:01", "remaining_time": "7:13:38"}
+{"current_steps": 1010, "total_steps": 2350, "loss": 2.367, "learning_rate": 3.1678486997635934e-05, "epoch": 0.8591631665691956, "percentage": 42.98, "elapsed_time": "5:25:38", "remaining_time": "7:12:01"}
+{"current_steps": 1015, "total_steps": 2350, "loss": 2.3557, "learning_rate": 3.156028368794326e-05, "epoch": 0.8634164495720134, "percentage": 43.19, "elapsed_time": "5:27:15", "remaining_time": "7:10:26"}
+{"current_steps": 1020, "total_steps": 2350, "loss": 2.3673, "learning_rate": 3.144208037825059e-05, "epoch": 0.8676697325748312, "percentage": 43.4, "elapsed_time": "5:28:51", "remaining_time": "7:08:48"}
+{"current_steps": 1025, "total_steps": 2350, "loss": 2.3407, "learning_rate": 3.132387706855792e-05, "epoch": 0.871923015577649, "percentage": 43.62, "elapsed_time": "5:30:26", "remaining_time": "7:07:09"}
+{"current_steps": 1030, "total_steps": 2350, "loss": 2.3363, "learning_rate": 3.120567375886525e-05, "epoch": 0.8761762985804668, "percentage": 43.83, "elapsed_time": "5:32:02", "remaining_time": "7:05:31"}
+{"current_steps": 1035, "total_steps": 2350, "loss": 2.3894, "learning_rate": 3.108747044917258e-05, "epoch": 0.8804295815832845, "percentage": 44.04, "elapsed_time": "5:33:38", "remaining_time": "7:03:53"}
+{"current_steps": 1040, "total_steps": 2350, "loss": 2.3537, "learning_rate": 3.0969267139479904e-05, "epoch": 0.8846828645861023, "percentage": 44.26, "elapsed_time": "5:35:13", "remaining_time": "7:02:15"}
+{"current_steps": 1045, "total_steps": 2350, "loss": 2.3588, "learning_rate": 3.085106382978723e-05, "epoch": 0.8889361475889201, "percentage": 44.47, "elapsed_time": "5:36:49", "remaining_time": "7:00:37"}
+{"current_steps": 1050, "total_steps": 2350, "loss": 2.3798, "learning_rate": 3.073286052009456e-05, "epoch": 0.8931894305917379, "percentage": 44.68, "elapsed_time": "5:38:24", "remaining_time": "6:58:59"}
+{"current_steps": 1055, "total_steps": 2350, "loss": 2.364, "learning_rate": 3.061465721040189e-05, "epoch": 0.8974427135945559, "percentage": 44.89, "elapsed_time": "5:40:00", "remaining_time": "6:57:20"}
+{"current_steps": 1060, "total_steps": 2350, "loss": 2.3922, "learning_rate": 3.0496453900709222e-05, "epoch": 0.9016959965973737, "percentage": 45.11, "elapsed_time": "5:41:35", "remaining_time": "6:55:42"}
+{"current_steps": 1065, "total_steps": 2350, "loss": 2.3189, "learning_rate": 3.0378250591016548e-05, "epoch": 0.9059492796001914, "percentage": 45.32, "elapsed_time": "5:43:10", "remaining_time": "6:54:04"}
+{"current_steps": 1070, "total_steps": 2350, "loss": 2.3871, "learning_rate": 3.0260047281323877e-05, "epoch": 0.9102025626030092, "percentage": 45.53, "elapsed_time": "5:44:45", "remaining_time": "6:52:25"}
+{"current_steps": 1075, "total_steps": 2350, "loss": 2.3712, "learning_rate": 3.0141843971631207e-05, "epoch": 0.914455845605827, "percentage": 45.74, "elapsed_time": "5:46:22", "remaining_time": "6:50:49"}
+{"current_steps": 1080, "total_steps": 2350, "loss": 2.3451, "learning_rate": 3.0023640661938536e-05, "epoch": 0.9187091286086448, "percentage": 45.96, "elapsed_time": "5:47:58", "remaining_time": "6:49:11"}
+{"current_steps": 1085, "total_steps": 2350, "loss": 2.2947, "learning_rate": 2.9905437352245862e-05, "epoch": 0.9229624116114626, "percentage": 46.17, "elapsed_time": "5:49:33", "remaining_time": "6:47:32"}
+{"current_steps": 1090, "total_steps": 2350, "loss": 2.3742, "learning_rate": 2.9787234042553192e-05, "epoch": 0.9272156946142804, "percentage": 46.38, "elapsed_time": "5:51:08", "remaining_time": "6:45:54"}
+{"current_steps": 1095, "total_steps": 2350, "loss": 2.3482, "learning_rate": 2.966903073286052e-05, "epoch": 0.9314689776170982, "percentage": 46.6, "elapsed_time": "5:52:44", "remaining_time": "6:44:16"}
+{"current_steps": 1100, "total_steps": 2350, "loss": 2.3385, "learning_rate": 2.9550827423167847e-05, "epoch": 0.935722260619916, "percentage": 46.81, "elapsed_time": "5:54:19", "remaining_time": "6:42:38"}
+{"current_steps": 1105, "total_steps": 2350, "loss": 2.3676, "learning_rate": 2.9432624113475177e-05, "epoch": 0.9399755436227338, "percentage": 47.02, "elapsed_time": "5:55:55", "remaining_time": "6:41:00"}
+{"current_steps": 1110, "total_steps": 2350, "loss": 2.3245, "learning_rate": 2.9314420803782506e-05, "epoch": 0.9442288266255516, "percentage": 47.23, "elapsed_time": "5:57:30", "remaining_time": "6:39:22"}
+{"current_steps": 1115, "total_steps": 2350, "loss": 2.346, "learning_rate": 2.9196217494089832e-05, "epoch": 0.9484821096283694, "percentage": 47.45, "elapsed_time": "5:59:05", "remaining_time": "6:37:43"}
+{"current_steps": 1120, "total_steps": 2350, "loss": 2.3601, "learning_rate": 2.9078014184397162e-05, "epoch": 0.9527353926311872, "percentage": 47.66, "elapsed_time": "6:00:40", "remaining_time": "6:36:05"}
+{"current_steps": 1125, "total_steps": 2350, "loss": 2.3, "learning_rate": 2.895981087470449e-05, "epoch": 0.956988675634005, "percentage": 47.87, "elapsed_time": "6:02:15", "remaining_time": "6:34:27"}
+{"current_steps": 1130, "total_steps": 2350, "loss": 2.3347, "learning_rate": 2.8841607565011824e-05, "epoch": 0.9612419586368228, "percentage": 48.09, "elapsed_time": "6:03:50", "remaining_time": "6:32:49"}
+{"current_steps": 1135, "total_steps": 2350, "loss": 2.3142, "learning_rate": 2.8723404255319154e-05, "epoch": 0.9654952416396406, "percentage": 48.3, "elapsed_time": "6:05:25", "remaining_time": "6:31:10"}
+{"current_steps": 1140, "total_steps": 2350, "loss": 2.3444, "learning_rate": 2.8605200945626483e-05, "epoch": 0.9697485246424584, "percentage": 48.51, "elapsed_time": "6:07:00", "remaining_time": "6:29:32"}
+{"current_steps": 1145, "total_steps": 2350, "loss": 2.332, "learning_rate": 2.848699763593381e-05, "epoch": 0.9740018076452762, "percentage": 48.72, "elapsed_time": "6:08:35", "remaining_time": "6:27:54"}
+{"current_steps": 1150, "total_steps": 2350, "loss": 2.3412, "learning_rate": 2.836879432624114e-05, "epoch": 0.978255090648094, "percentage": 48.94, "elapsed_time": "6:10:10", "remaining_time": "6:26:16"}
+{"current_steps": 1155, "total_steps": 2350, "loss": 2.3245, "learning_rate": 2.8250591016548468e-05, "epoch": 0.9825083736509118, "percentage": 49.15, "elapsed_time": "6:11:45", "remaining_time": "6:24:38"}
+{"current_steps": 1160, "total_steps": 2350, "loss": 2.3507, "learning_rate": 2.8132387706855794e-05, "epoch": 0.9867616566537296, "percentage": 49.36, "elapsed_time": "6:13:21", "remaining_time": "6:23:00"}
+{"current_steps": 1165, "total_steps": 2350, "loss": 2.3284, "learning_rate": 2.8014184397163124e-05, "epoch": 0.9910149396565474, "percentage": 49.57, "elapsed_time": "6:14:56", "remaining_time": "6:21:22"}
+{"current_steps": 1170, "total_steps": 2350, "loss": 2.3622, "learning_rate": 2.7895981087470453e-05, "epoch": 0.9952682226593652, "percentage": 49.79, "elapsed_time": "6:16:32", "remaining_time": "6:19:45"}
+{"current_steps": 1175, "total_steps": 2350, "loss": 2.3499, "learning_rate": 2.777777777777778e-05, "epoch": 0.999521505662183, "percentage": 50.0, "elapsed_time": "6:18:07", "remaining_time": "6:18:07"}
+{"current_steps": 1180, "total_steps": 2350, "loss": 2.0719, "learning_rate": 2.765957446808511e-05, "epoch": 1.0037747886650008, "percentage": 50.21, "elapsed_time": "6:19:45", "remaining_time": "6:16:32"}
+{"current_steps": 1185, "total_steps": 2350, "loss": 2.0455, "learning_rate": 2.7541371158392438e-05, "epoch": 1.0080280716678185, "percentage": 50.43, "elapsed_time": "6:21:22", "remaining_time": "6:14:56"}
+{"current_steps": 1190, "total_steps": 2350, "loss": 2.0007, "learning_rate": 2.7423167848699764e-05, "epoch": 1.0122813546706364, "percentage": 50.64, "elapsed_time": "6:22:58", "remaining_time": "6:13:19"}
+{"current_steps": 1195, "total_steps": 2350, "loss": 2.0265, "learning_rate": 2.7304964539007094e-05, "epoch": 1.016534637673454, "percentage": 50.85, "elapsed_time": "6:24:34", "remaining_time": "6:11:41"}
+{"current_steps": 1200, "total_steps": 2350, "loss": 1.9858, "learning_rate": 2.7186761229314423e-05, "epoch": 1.020787920676272, "percentage": 51.06, "elapsed_time": "6:26:09", "remaining_time": "6:10:04"}
+{"current_steps": 1205, "total_steps": 2350, "loss": 2.0112, "learning_rate": 2.7068557919621753e-05, "epoch": 1.02504120367909, "percentage": 51.28, "elapsed_time": "6:27:47", "remaining_time": "6:08:28"}
+{"current_steps": 1210, "total_steps": 2350, "loss": 2.0078, "learning_rate": 2.695035460992908e-05, "epoch": 1.0292944866819076, "percentage": 51.49, "elapsed_time": "6:29:23", "remaining_time": "6:06:52"}
+{"current_steps": 1215, "total_steps": 2350, "loss": 2.0072, "learning_rate": 2.6832151300236408e-05, "epoch": 1.0335477696847255, "percentage": 51.7, "elapsed_time": "6:31:01", "remaining_time": "6:05:16"}
+{"current_steps": 1220, "total_steps": 2350, "loss": 1.9904, "learning_rate": 2.6713947990543738e-05, "epoch": 1.0378010526875432, "percentage": 51.91, "elapsed_time": "6:32:37", "remaining_time": "6:03:39"}
+{"current_steps": 1225, "total_steps": 2350, "loss": 2.0318, "learning_rate": 2.6595744680851064e-05, "epoch": 1.042054335690361, "percentage": 52.13, "elapsed_time": "6:34:13", "remaining_time": "6:02:02"}
+{"current_steps": 1230, "total_steps": 2350, "loss": 2.0214, "learning_rate": 2.6477541371158393e-05, "epoch": 1.0463076186931788, "percentage": 52.34, "elapsed_time": "6:35:49", "remaining_time": "6:00:25"}
+{"current_steps": 1235, "total_steps": 2350, "loss": 2.0233, "learning_rate": 2.6359338061465723e-05, "epoch": 1.0505609016959967, "percentage": 52.55, "elapsed_time": "6:37:25", "remaining_time": "5:58:48"}
+{"current_steps": 1240, "total_steps": 2350, "loss": 1.9697, "learning_rate": 2.624113475177305e-05, "epoch": 1.0548141846988144, "percentage": 52.77, "elapsed_time": "6:39:01", "remaining_time": "5:57:11"}
+{"current_steps": 1245, "total_steps": 2350, "loss": 1.9899, "learning_rate": 2.6122931442080378e-05, "epoch": 1.0590674677016323, "percentage": 52.98, "elapsed_time": "6:40:37", "remaining_time": "5:55:34"}
+{"current_steps": 1250, "total_steps": 2350, "loss": 2.0306, "learning_rate": 2.6004728132387708e-05, "epoch": 1.06332075070445, "percentage": 53.19, "elapsed_time": "6:42:13", "remaining_time": "5:53:57"}
+{"current_steps": 1255, "total_steps": 2350, "loss": 1.9851, "learning_rate": 2.5886524822695034e-05, "epoch": 1.0675740337072679, "percentage": 53.4, "elapsed_time": "6:43:50", "remaining_time": "5:52:21"}
+{"current_steps": 1260, "total_steps": 2350, "loss": 2.0191, "learning_rate": 2.5768321513002363e-05, "epoch": 1.0718273167100856, "percentage": 53.62, "elapsed_time": "6:45:26", "remaining_time": "5:50:44"}
+{"current_steps": 1265, "total_steps": 2350, "loss": 2.0307, "learning_rate": 2.5650118203309693e-05, "epoch": 1.0760805997129035, "percentage": 53.83, "elapsed_time": "6:47:03", "remaining_time": "5:49:08"}
+{"current_steps": 1270, "total_steps": 2350, "loss": 2.0119, "learning_rate": 2.5531914893617022e-05, "epoch": 1.0803338827157212, "percentage": 54.04, "elapsed_time": "6:48:39", "remaining_time": "5:47:31"}
+{"current_steps": 1275, "total_steps": 2350, "loss": 1.9794, "learning_rate": 2.5413711583924348e-05, "epoch": 1.084587165718539, "percentage": 54.26, "elapsed_time": "6:50:15", "remaining_time": "5:45:54"}
+{"current_steps": 1280, "total_steps": 2350, "loss": 2.0091, "learning_rate": 2.5295508274231678e-05, "epoch": 1.0888404487213568, "percentage": 54.47, "elapsed_time": "6:51:53", "remaining_time": "5:44:18"}
+{"current_steps": 1285, "total_steps": 2350, "loss": 1.9977, "learning_rate": 2.5177304964539007e-05, "epoch": 1.0930937317241747, "percentage": 54.68, "elapsed_time": "6:53:29", "remaining_time": "5:42:41"}
+{"current_steps": 1290, "total_steps": 2350, "loss": 2.0054, "learning_rate": 2.5059101654846333e-05, "epoch": 1.0973470147269924, "percentage": 54.89, "elapsed_time": "6:55:04", "remaining_time": "5:41:04"}
+{"current_steps": 1295, "total_steps": 2350, "loss": 2.0273, "learning_rate": 2.4940898345153666e-05, "epoch": 1.1016002977298103, "percentage": 55.11, "elapsed_time": "6:56:40", "remaining_time": "5:39:27"}
+{"current_steps": 1300, "total_steps": 2350, "loss": 1.9994, "learning_rate": 2.4822695035460995e-05, "epoch": 1.105853580732628, "percentage": 55.32, "elapsed_time": "6:58:16", "remaining_time": "5:37:50"}
+{"current_steps": 1305, "total_steps": 2350, "loss": 2.0084, "learning_rate": 2.470449172576832e-05, "epoch": 1.1101068637354459, "percentage": 55.53, "elapsed_time": "6:59:52", "remaining_time": "5:36:13"}
+{"current_steps": 1310, "total_steps": 2350, "loss": 2.0087, "learning_rate": 2.458628841607565e-05, "epoch": 1.1143601467382636, "percentage": 55.74, "elapsed_time": "7:01:28", "remaining_time": "5:34:35"}
+{"current_steps": 1315, "total_steps": 2350, "loss": 1.9766, "learning_rate": 2.446808510638298e-05, "epoch": 1.1186134297410815, "percentage": 55.96, "elapsed_time": "7:03:03", "remaining_time": "5:32:58"}
+{"current_steps": 1320, "total_steps": 2350, "loss": 2.0408, "learning_rate": 2.4349881796690306e-05, "epoch": 1.1228667127438992, "percentage": 56.17, "elapsed_time": "7:04:39", "remaining_time": "5:31:21"}
+{"current_steps": 1325, "total_steps": 2350, "loss": 1.9739, "learning_rate": 2.4231678486997636e-05, "epoch": 1.127119995746717, "percentage": 56.38, "elapsed_time": "7:06:14", "remaining_time": "5:29:44"}
+{"current_steps": 1330, "total_steps": 2350, "loss": 2.005, "learning_rate": 2.4113475177304965e-05, "epoch": 1.1313732787495347, "percentage": 56.6, "elapsed_time": "7:07:50", "remaining_time": "5:28:07"}
+{"current_steps": 1335, "total_steps": 2350, "loss": 2.0161, "learning_rate": 2.3995271867612295e-05, "epoch": 1.1356265617523527, "percentage": 56.81, "elapsed_time": "7:09:26", "remaining_time": "5:26:29"}
+{"current_steps": 1340, "total_steps": 2350, "loss": 1.995, "learning_rate": 2.3877068557919624e-05, "epoch": 1.1398798447551703, "percentage": 57.02, "elapsed_time": "7:11:01", "remaining_time": "5:24:52"}
+{"current_steps": 1345, "total_steps": 2350, "loss": 2.0458, "learning_rate": 2.3758865248226954e-05, "epoch": 1.1441331277579883, "percentage": 57.23, "elapsed_time": "7:12:37", "remaining_time": "5:23:15"}
+{"current_steps": 1350, "total_steps": 2350, "loss": 1.9776, "learning_rate": 2.364066193853428e-05, "epoch": 1.148386410760806, "percentage": 57.45, "elapsed_time": "7:14:13", "remaining_time": "5:21:38"}
+{"current_steps": 1355, "total_steps": 2350, "loss": 1.9791, "learning_rate": 2.352245862884161e-05, "epoch": 1.1526396937636239, "percentage": 57.66, "elapsed_time": "7:15:49", "remaining_time": "5:20:01"}
+{"current_steps": 1360, "total_steps": 2350, "loss": 1.9759, "learning_rate": 2.340425531914894e-05, "epoch": 1.1568929767664415, "percentage": 57.87, "elapsed_time": "7:17:24", "remaining_time": "5:18:24"}
+{"current_steps": 1365, "total_steps": 2350, "loss": 1.9886, "learning_rate": 2.3286052009456265e-05, "epoch": 1.1611462597692594, "percentage": 58.09, "elapsed_time": "7:19:00", "remaining_time": "5:16:47"}
+{"current_steps": 1370, "total_steps": 2350, "loss": 2.0073, "learning_rate": 2.3167848699763594e-05, "epoch": 1.1653995427720771, "percentage": 58.3, "elapsed_time": "7:20:35", "remaining_time": "5:15:10"}
+{"current_steps": 1375, "total_steps": 2350, "loss": 1.9885, "learning_rate": 2.3049645390070924e-05, "epoch": 1.169652825774895, "percentage": 58.51, "elapsed_time": "7:22:11", "remaining_time": "5:13:33"}
+{"current_steps": 1380, "total_steps": 2350, "loss": 1.9871, "learning_rate": 2.293144208037825e-05, "epoch": 1.1739061087777127, "percentage": 58.72, "elapsed_time": "7:23:46", "remaining_time": "5:11:55"}
+{"current_steps": 1385, "total_steps": 2350, "loss": 1.9838, "learning_rate": 2.281323877068558e-05, "epoch": 1.1781593917805306, "percentage": 58.94, "elapsed_time": "7:25:22", "remaining_time": "5:10:18"}
+{"current_steps": 1390, "total_steps": 2350, "loss": 2.0428, "learning_rate": 2.269503546099291e-05, "epoch": 1.1824126747833483, "percentage": 59.15, "elapsed_time": "7:26:58", "remaining_time": "5:08:41"}
+{"current_steps": 1395, "total_steps": 2350, "loss": 2.0274, "learning_rate": 2.2576832151300238e-05, "epoch": 1.1866659577861662, "percentage": 59.36, "elapsed_time": "7:28:33", "remaining_time": "5:07:04"}
+{"current_steps": 1400, "total_steps": 2350, "loss": 2.0031, "learning_rate": 2.2458628841607564e-05, "epoch": 1.190919240788984, "percentage": 59.57, "elapsed_time": "7:30:09", "remaining_time": "5:05:27"}
+{"current_steps": 1405, "total_steps": 2350, "loss": 2.0222, "learning_rate": 2.2340425531914894e-05, "epoch": 1.1951725237918018, "percentage": 59.79, "elapsed_time": "7:31:45", "remaining_time": "5:03:50"}
+{"current_steps": 1410, "total_steps": 2350, "loss": 2.0086, "learning_rate": 2.2222222222222223e-05, "epoch": 1.1994258067946195, "percentage": 60.0, "elapsed_time": "7:33:21", "remaining_time": "5:02:14"}
+{"current_steps": 1415, "total_steps": 2350, "loss": 1.9803, "learning_rate": 2.2104018912529553e-05, "epoch": 1.2036790897974374, "percentage": 60.21, "elapsed_time": "7:34:56", "remaining_time": "5:00:37"}
+{"current_steps": 1420, "total_steps": 2350, "loss": 1.986, "learning_rate": 2.1985815602836882e-05, "epoch": 1.2079323728002551, "percentage": 60.43, "elapsed_time": "7:36:32", "remaining_time": "4:59:00"}
+{"current_steps": 1425, "total_steps": 2350, "loss": 2.0073, "learning_rate": 2.186761229314421e-05, "epoch": 1.212185655803073, "percentage": 60.64, "elapsed_time": "7:38:08", "remaining_time": "4:57:23"}
+{"current_steps": 1430, "total_steps": 2350, "loss": 1.9992, "learning_rate": 2.1749408983451538e-05, "epoch": 1.2164389388058907, "percentage": 60.85, "elapsed_time": "7:39:44", "remaining_time": "4:55:46"}
+{"current_steps": 1435, "total_steps": 2350, "loss": 2.004, "learning_rate": 2.1631205673758867e-05, "epoch": 1.2206922218087086, "percentage": 61.06, "elapsed_time": "7:41:19", "remaining_time": "4:54:09"}
+{"current_steps": 1440, "total_steps": 2350, "loss": 1.9966, "learning_rate": 2.1513002364066197e-05, "epoch": 1.2249455048115263, "percentage": 61.28, "elapsed_time": "7:42:55", "remaining_time": "4:52:32"}
+{"current_steps": 1445, "total_steps": 2350, "loss": 2.0202, "learning_rate": 2.1394799054373523e-05, "epoch": 1.2291987878143442, "percentage": 61.49, "elapsed_time": "7:44:30", "remaining_time": "4:50:55"}
+{"current_steps": 1450, "total_steps": 2350, "loss": 2.0352, "learning_rate": 2.1276595744680852e-05, "epoch": 1.233452070817162, "percentage": 61.7, "elapsed_time": "7:46:05", "remaining_time": "4:49:17"}
+{"current_steps": 1455, "total_steps": 2350, "loss": 1.9897, "learning_rate": 2.115839243498818e-05, "epoch": 1.2377053538199798, "percentage": 61.91, "elapsed_time": "7:47:40", "remaining_time": "4:47:40"}
+{"current_steps": 1460, "total_steps": 2350, "loss": 1.9795, "learning_rate": 2.1040189125295508e-05, "epoch": 1.2419586368227975, "percentage": 62.13, "elapsed_time": "7:49:15", "remaining_time": "4:46:03"}
+{"current_steps": 1465, "total_steps": 2350, "loss": 1.9816, "learning_rate": 2.0921985815602837e-05, "epoch": 1.2462119198256154, "percentage": 62.34, "elapsed_time": "7:50:51", "remaining_time": "4:44:26"}
+{"current_steps": 1470, "total_steps": 2350, "loss": 2.011, "learning_rate": 2.0803782505910167e-05, "epoch": 1.2504652028284333, "percentage": 62.55, "elapsed_time": "7:52:26", "remaining_time": "4:42:49"}
+{"current_steps": 1475, "total_steps": 2350, "loss": 1.9725, "learning_rate": 2.0685579196217493e-05, "epoch": 1.254718485831251, "percentage": 62.77, "elapsed_time": "7:54:01", "remaining_time": "4:41:11"}
+{"current_steps": 1480, "total_steps": 2350, "loss": 2.0059, "learning_rate": 2.0567375886524822e-05, "epoch": 1.2589717688340687, "percentage": 62.98, "elapsed_time": "7:55:36", "remaining_time": "4:39:34"}
+{"current_steps": 1485, "total_steps": 2350, "loss": 1.9543, "learning_rate": 2.0449172576832152e-05, "epoch": 1.2632250518368866, "percentage": 63.19, "elapsed_time": "7:57:11", "remaining_time": "4:37:57"}
+{"current_steps": 1490, "total_steps": 2350, "loss": 2.0051, "learning_rate": 2.033096926713948e-05, "epoch": 1.2674783348397045, "percentage": 63.4, "elapsed_time": "7:58:46", "remaining_time": "4:36:20"}
+{"current_steps": 1495, "total_steps": 2350, "loss": 1.9752, "learning_rate": 2.0212765957446807e-05, "epoch": 1.2717316178425222, "percentage": 63.62, "elapsed_time": "8:00:21", "remaining_time": "4:34:43"}
+{"current_steps": 1500, "total_steps": 2350, "loss": 1.9795, "learning_rate": 2.009456264775414e-05, "epoch": 1.2759849008453399, "percentage": 63.83, "elapsed_time": "8:01:56", "remaining_time": "4:33:06"}
+{"current_steps": 1505, "total_steps": 2350, "loss": 2.0009, "learning_rate": 1.9976359338061466e-05, "epoch": 1.2802381838481578, "percentage": 64.04, "elapsed_time": "8:03:31", "remaining_time": "4:31:29"}
+{"current_steps": 1510, "total_steps": 2350, "loss": 1.994, "learning_rate": 1.9858156028368796e-05, "epoch": 1.2844914668509757, "percentage": 64.26, "elapsed_time": "8:05:07", "remaining_time": "4:29:52"}
+{"current_steps": 1515, "total_steps": 2350, "loss": 1.9667, "learning_rate": 1.9739952718676125e-05, "epoch": 1.2887447498537934, "percentage": 64.47, "elapsed_time": "8:06:45", "remaining_time": "4:28:16"}
+{"current_steps": 1520, "total_steps": 2350, "loss": 1.974, "learning_rate": 1.9621749408983455e-05, "epoch": 1.292998032856611, "percentage": 64.68, "elapsed_time": "8:08:22", "remaining_time": "4:26:40"}
+{"current_steps": 1525, "total_steps": 2350, "loss": 2.0098, "learning_rate": 1.950354609929078e-05, "epoch": 1.297251315859429, "percentage": 64.89, "elapsed_time": "8:09:59", "remaining_time": "4:25:04"}
+{"current_steps": 1530, "total_steps": 2350, "loss": 1.9415, "learning_rate": 1.938534278959811e-05, "epoch": 1.301504598862247, "percentage": 65.11, "elapsed_time": "8:11:35", "remaining_time": "4:23:28"}
+{"current_steps": 1535, "total_steps": 2350, "loss": 1.9867, "learning_rate": 1.926713947990544e-05, "epoch": 1.3057578818650646, "percentage": 65.32, "elapsed_time": "8:13:12", "remaining_time": "4:21:52"}
+{"current_steps": 1540, "total_steps": 2350, "loss": 2.0018, "learning_rate": 1.9148936170212766e-05, "epoch": 1.3100111648678823, "percentage": 65.53, "elapsed_time": "8:14:49", "remaining_time": "4:20:16"}
+{"current_steps": 1545, "total_steps": 2350, "loss": 1.9627, "learning_rate": 1.9030732860520095e-05, "epoch": 1.3142644478707002, "percentage": 65.74, "elapsed_time": "8:16:27", "remaining_time": "4:18:40"}
+{"current_steps": 1550, "total_steps": 2350, "loss": 2.0021, "learning_rate": 1.8912529550827425e-05, "epoch": 1.318517730873518, "percentage": 65.96, "elapsed_time": "8:18:03", "remaining_time": "4:17:03"}
+{"current_steps": 1555, "total_steps": 2350, "loss": 1.9893, "learning_rate": 1.879432624113475e-05, "epoch": 1.3227710138763358, "percentage": 66.17, "elapsed_time": "8:19:40", "remaining_time": "4:15:27"}
+{"current_steps": 1560, "total_steps": 2350, "loss": 1.9839, "learning_rate": 1.867612293144208e-05, "epoch": 1.3270242968791535, "percentage": 66.38, "elapsed_time": "8:21:16", "remaining_time": "4:13:51"}
+{"current_steps": 1565, "total_steps": 2350, "loss": 1.9724, "learning_rate": 1.855791962174941e-05, "epoch": 1.3312775798819714, "percentage": 66.6, "elapsed_time": "8:22:52", "remaining_time": "4:12:14"}
+{"current_steps": 1570, "total_steps": 2350, "loss": 1.9775, "learning_rate": 1.8439716312056736e-05, "epoch": 1.3355308628847893, "percentage": 66.81, "elapsed_time": "8:24:29", "remaining_time": "4:10:38"}
+{"current_steps": 1575, "total_steps": 2350, "loss": 1.991, "learning_rate": 1.8321513002364065e-05, "epoch": 1.339784145887607, "percentage": 67.02, "elapsed_time": "8:26:05", "remaining_time": "4:09:01"}
+{"current_steps": 1580, "total_steps": 2350, "loss": 1.9752, "learning_rate": 1.8203309692671398e-05, "epoch": 1.3440374288904249, "percentage": 67.23, "elapsed_time": "8:27:41", "remaining_time": "4:07:25"}
+{"current_steps": 1585, "total_steps": 2350, "loss": 2.0044, "learning_rate": 1.8085106382978724e-05, "epoch": 1.3482907118932426, "percentage": 67.45, "elapsed_time": "8:29:17", "remaining_time": "4:05:48"}
+{"current_steps": 1590, "total_steps": 2350, "loss": 1.9499, "learning_rate": 1.7966903073286054e-05, "epoch": 1.3525439948960605, "percentage": 67.66, "elapsed_time": "8:30:53", "remaining_time": "4:04:11"}
+{"current_steps": 1595, "total_steps": 2350, "loss": 2.0067, "learning_rate": 1.7848699763593383e-05, "epoch": 1.3567972778988782, "percentage": 67.87, "elapsed_time": "8:32:28", "remaining_time": "4:02:35"}
+{"current_steps": 1600, "total_steps": 2350, "loss": 1.9673, "learning_rate": 1.773049645390071e-05, "epoch": 1.361050560901696, "percentage": 68.09, "elapsed_time": "8:34:04", "remaining_time": "4:00:58"}
+{"current_steps": 1605, "total_steps": 2350, "loss": 2.021, "learning_rate": 1.761229314420804e-05, "epoch": 1.3653038439045138, "percentage": 68.3, "elapsed_time": "8:35:40", "remaining_time": "3:59:21"}
+{"current_steps": 1610, "total_steps": 2350, "loss": 1.952, "learning_rate": 1.7494089834515368e-05, "epoch": 1.3695571269073317, "percentage": 68.51, "elapsed_time": "8:37:15", "remaining_time": "3:57:44"}
+{"current_steps": 1615, "total_steps": 2350, "loss": 2.0264, "learning_rate": 1.7375886524822697e-05, "epoch": 1.3738104099101494, "percentage": 68.72, "elapsed_time": "8:38:52", "remaining_time": "3:56:08"}
+{"current_steps": 1620, "total_steps": 2350, "loss": 1.949, "learning_rate": 1.7257683215130024e-05, "epoch": 1.3780636929129673, "percentage": 68.94, "elapsed_time": "8:40:28", "remaining_time": "3:54:32"}
+{"current_steps": 1625, "total_steps": 2350, "loss": 2.0119, "learning_rate": 1.7139479905437353e-05, "epoch": 1.382316975915785, "percentage": 69.15, "elapsed_time": "8:42:04", "remaining_time": "3:52:55"}
+{"current_steps": 1630, "total_steps": 2350, "loss": 1.9519, "learning_rate": 1.7021276595744682e-05, "epoch": 1.3865702589186029, "percentage": 69.36, "elapsed_time": "8:43:40", "remaining_time": "3:51:18"}
+{"current_steps": 1635, "total_steps": 2350, "loss": 1.9907, "learning_rate": 1.690307328605201e-05, "epoch": 1.3908235419214205, "percentage": 69.57, "elapsed_time": "8:45:16", "remaining_time": "3:49:42"}
+{"current_steps": 1640, "total_steps": 2350, "loss": 1.9697, "learning_rate": 1.6784869976359338e-05, "epoch": 1.3950768249242385, "percentage": 69.79, "elapsed_time": "8:46:53", "remaining_time": "3:48:06"}
+{"current_steps": 1645, "total_steps": 2350, "loss": 2.0108, "learning_rate": 1.6666666666666667e-05, "epoch": 1.3993301079270561, "percentage": 70.0, "elapsed_time": "8:48:29", "remaining_time": "3:46:29"}
+{"current_steps": 1650, "total_steps": 2350, "loss": 1.9791, "learning_rate": 1.6548463356973994e-05, "epoch": 1.403583390929874, "percentage": 70.21, "elapsed_time": "8:50:04", "remaining_time": "3:44:53"}
+{"current_steps": 1655, "total_steps": 2350, "loss": 1.943, "learning_rate": 1.6430260047281323e-05, "epoch": 1.4078366739326917, "percentage": 70.43, "elapsed_time": "8:51:40", "remaining_time": "3:43:16"}
+{"current_steps": 1660, "total_steps": 2350, "loss": 1.9913, "learning_rate": 1.6312056737588656e-05, "epoch": 1.4120899569355096, "percentage": 70.64, "elapsed_time": "8:53:16", "remaining_time": "3:41:39"}
+{"current_steps": 1665, "total_steps": 2350, "loss": 1.9857, "learning_rate": 1.6193853427895982e-05, "epoch": 1.4163432399383273, "percentage": 70.85, "elapsed_time": "8:54:51", "remaining_time": "3:40:02"}
+{"current_steps": 1670, "total_steps": 2350, "loss": 1.9915, "learning_rate": 1.607565011820331e-05, "epoch": 1.4205965229411452, "percentage": 71.06, "elapsed_time": "8:56:27", "remaining_time": "3:38:26"}
+{"current_steps": 1675, "total_steps": 2350, "loss": 1.981, "learning_rate": 1.595744680851064e-05, "epoch": 1.424849805943963, "percentage": 71.28, "elapsed_time": "8:58:02", "remaining_time": "3:36:49"}
+{"current_steps": 1680, "total_steps": 2350, "loss": 1.9546, "learning_rate": 1.5839243498817967e-05, "epoch": 1.4291030889467808, "percentage": 71.49, "elapsed_time": "8:59:38", "remaining_time": "3:35:12"}
+{"current_steps": 1685, "total_steps": 2350, "loss": 1.9395, "learning_rate": 1.5721040189125296e-05, "epoch": 1.4333563719495985, "percentage": 71.7, "elapsed_time": "9:01:13", "remaining_time": "3:33:35"}
+{"current_steps": 1690, "total_steps": 2350, "loss": 1.9626, "learning_rate": 1.5602836879432626e-05, "epoch": 1.4376096549524164, "percentage": 71.91, "elapsed_time": "9:02:48", "remaining_time": "3:31:59"}
+{"current_steps": 1695, "total_steps": 2350, "loss": 1.9989, "learning_rate": 1.5484633569739952e-05, "epoch": 1.4418629379552341, "percentage": 72.13, "elapsed_time": "9:04:23", "remaining_time": "3:30:22"}
+{"current_steps": 1700, "total_steps": 2350, "loss": 1.9689, "learning_rate": 1.536643026004728e-05, "epoch": 1.446116220958052, "percentage": 72.34, "elapsed_time": "9:05:59", "remaining_time": "3:28:45"}
+{"current_steps": 1705, "total_steps": 2350, "loss": 1.9641, "learning_rate": 1.5248226950354611e-05, "epoch": 1.4503695039608697, "percentage": 72.55, "elapsed_time": "9:07:34", "remaining_time": "3:27:08"}
+{"current_steps": 1710, "total_steps": 2350, "loss": 1.951, "learning_rate": 1.5130023640661939e-05, "epoch": 1.4546227869636876, "percentage": 72.77, "elapsed_time": "9:09:09", "remaining_time": "3:25:31"}
+{"current_steps": 1715, "total_steps": 2350, "loss": 1.9763, "learning_rate": 1.5011820330969268e-05, "epoch": 1.4588760699665053, "percentage": 72.98, "elapsed_time": "9:10:44", "remaining_time": "3:23:55"}
+{"current_steps": 1720, "total_steps": 2350, "loss": 1.9631, "learning_rate": 1.4893617021276596e-05, "epoch": 1.4631293529693232, "percentage": 73.19, "elapsed_time": "9:12:19", "remaining_time": "3:22:18"}
+{"current_steps": 1725, "total_steps": 2350, "loss": 2.0039, "learning_rate": 1.4775413711583924e-05, "epoch": 1.467382635972141, "percentage": 73.4, "elapsed_time": "9:13:55", "remaining_time": "3:20:41"}
+{"current_steps": 1730, "total_steps": 2350, "loss": 1.9483, "learning_rate": 1.4657210401891253e-05, "epoch": 1.4716359189749588, "percentage": 73.62, "elapsed_time": "9:15:31", "remaining_time": "3:19:05"}
+{"current_steps": 1735, "total_steps": 2350, "loss": 1.9422, "learning_rate": 1.4539007092198581e-05, "epoch": 1.4758892019777767, "percentage": 73.83, "elapsed_time": "9:17:07", "remaining_time": "3:17:28"}
+{"current_steps": 1740, "total_steps": 2350, "loss": 1.9517, "learning_rate": 1.4420803782505912e-05, "epoch": 1.4801424849805944, "percentage": 74.04, "elapsed_time": "9:18:42", "remaining_time": "3:15:52"}
+{"current_steps": 1745, "total_steps": 2350, "loss": 1.9792, "learning_rate": 1.4302600472813242e-05, "epoch": 1.484395767983412, "percentage": 74.26, "elapsed_time": "9:20:18", "remaining_time": "3:14:15"}
+{"current_steps": 1750, "total_steps": 2350, "loss": 2.0064, "learning_rate": 1.418439716312057e-05, "epoch": 1.48864905098623, "percentage": 74.47, "elapsed_time": "9:21:53", "remaining_time": "3:12:39"}
+{"current_steps": 1755, "total_steps": 2350, "loss": 1.9686, "learning_rate": 1.4066193853427897e-05, "epoch": 1.492902333989048, "percentage": 74.68, "elapsed_time": "9:23:29", "remaining_time": "3:11:02"}
+{"current_steps": 1760, "total_steps": 2350, "loss": 1.9391, "learning_rate": 1.3947990543735227e-05, "epoch": 1.4971556169918656, "percentage": 74.89, "elapsed_time": "9:25:04", "remaining_time": "3:09:25"}
+{"current_steps": 1765, "total_steps": 2350, "loss": 1.9603, "learning_rate": 1.3829787234042554e-05, "epoch": 1.5014088999946833, "percentage": 75.11, "elapsed_time": "9:26:40", "remaining_time": "3:07:49"}
+{"current_steps": 1770, "total_steps": 2350, "loss": 1.9378, "learning_rate": 1.3711583924349882e-05, "epoch": 1.5056621829975012, "percentage": 75.32, "elapsed_time": "9:28:15", "remaining_time": "3:06:12"}
+{"current_steps": 1775, "total_steps": 2350, "loss": 1.9399, "learning_rate": 1.3593380614657212e-05, "epoch": 1.5099154660003191, "percentage": 75.53, "elapsed_time": "9:29:50", "remaining_time": "3:04:35"}
+{"current_steps": 1780, "total_steps": 2350, "loss": 1.9586, "learning_rate": 1.347517730496454e-05, "epoch": 1.5141687490031368, "percentage": 75.74, "elapsed_time": "9:31:25", "remaining_time": "3:02:58"}
+{"current_steps": 1785, "total_steps": 2350, "loss": 1.9742, "learning_rate": 1.3356973995271869e-05, "epoch": 1.5184220320059545, "percentage": 75.96, "elapsed_time": "9:33:00", "remaining_time": "3:01:22"}
+{"current_steps": 1790, "total_steps": 2350, "loss": 1.9867, "learning_rate": 1.3238770685579197e-05, "epoch": 1.5226753150087724, "percentage": 76.17, "elapsed_time": "9:34:35", "remaining_time": "2:59:45"}
+{"current_steps": 1795, "total_steps": 2350, "loss": 1.9653, "learning_rate": 1.3120567375886524e-05, "epoch": 1.5269285980115903, "percentage": 76.38, "elapsed_time": "9:36:10", "remaining_time": "2:58:08"}
+{"current_steps": 1800, "total_steps": 2350, "loss": 1.9553, "learning_rate": 1.3002364066193854e-05, "epoch": 1.531181881014408, "percentage": 76.6, "elapsed_time": "9:37:45", "remaining_time": "2:56:32"}
+{"current_steps": 1805, "total_steps": 2350, "loss": 1.9441, "learning_rate": 1.2884160756501182e-05, "epoch": 1.5354351640172257, "percentage": 76.81, "elapsed_time": "9:39:21", "remaining_time": "2:54:55"}
+{"current_steps": 1810, "total_steps": 2350, "loss": 1.9559, "learning_rate": 1.2765957446808511e-05, "epoch": 1.5396884470200436, "percentage": 77.02, "elapsed_time": "9:40:56", "remaining_time": "2:53:19"}
+{"current_steps": 1815, "total_steps": 2350, "loss": 1.9375, "learning_rate": 1.2647754137115839e-05, "epoch": 1.5439417300228615, "percentage": 77.23, "elapsed_time": "9:42:31", "remaining_time": "2:51:42"}
+{"current_steps": 1820, "total_steps": 2350, "loss": 1.9919, "learning_rate": 1.2529550827423167e-05, "epoch": 1.5481950130256792, "percentage": 77.45, "elapsed_time": "9:44:08", "remaining_time": "2:50:06"}
+{"current_steps": 1825, "total_steps": 2350, "loss": 1.9703, "learning_rate": 1.2411347517730498e-05, "epoch": 1.5524482960284969, "percentage": 77.66, "elapsed_time": "9:45:43", "remaining_time": "2:48:29"}
+{"current_steps": 1830, "total_steps": 2350, "loss": 1.9232, "learning_rate": 1.2293144208037825e-05, "epoch": 1.5567015790313148, "percentage": 77.87, "elapsed_time": "9:47:19", "remaining_time": "2:46:53"}
+{"current_steps": 1835, "total_steps": 2350, "loss": 1.9505, "learning_rate": 1.2174940898345153e-05, "epoch": 1.5609548620341327, "percentage": 78.09, "elapsed_time": "9:48:55", "remaining_time": "2:45:16"}
+{"current_steps": 1840, "total_steps": 2350, "loss": 1.9262, "learning_rate": 1.2056737588652483e-05, "epoch": 1.5652081450369504, "percentage": 78.3, "elapsed_time": "9:50:30", "remaining_time": "2:43:40"}
+{"current_steps": 1845, "total_steps": 2350, "loss": 1.9686, "learning_rate": 1.1938534278959812e-05, "epoch": 1.569461428039768, "percentage": 78.51, "elapsed_time": "9:52:05", "remaining_time": "2:42:03"}
+{"current_steps": 1850, "total_steps": 2350, "loss": 1.9462, "learning_rate": 1.182033096926714e-05, "epoch": 1.573714711042586, "percentage": 78.72, "elapsed_time": "9:53:41", "remaining_time": "2:40:27"}
+{"current_steps": 1855, "total_steps": 2350, "loss": 1.973, "learning_rate": 1.170212765957447e-05, "epoch": 1.5779679940454039, "percentage": 78.94, "elapsed_time": "9:55:21", "remaining_time": "2:38:52"}
+{"current_steps": 1860, "total_steps": 2350, "loss": 1.9319, "learning_rate": 1.1583924349881797e-05, "epoch": 1.5822212770482216, "percentage": 79.15, "elapsed_time": "9:56:59", "remaining_time": "2:37:16"}
+{"current_steps": 1865, "total_steps": 2350, "loss": 1.9633, "learning_rate": 1.1465721040189125e-05, "epoch": 1.5864745600510393, "percentage": 79.36, "elapsed_time": "9:58:37", "remaining_time": "2:35:40"}
+{"current_steps": 1870, "total_steps": 2350, "loss": 1.9293, "learning_rate": 1.1347517730496454e-05, "epoch": 1.5907278430538572, "percentage": 79.57, "elapsed_time": "10:00:15", "remaining_time": "2:34:04"}
+{"current_steps": 1875, "total_steps": 2350, "loss": 1.9791, "learning_rate": 1.1229314420803782e-05, "epoch": 1.594981126056675, "percentage": 79.79, "elapsed_time": "10:01:53", "remaining_time": "2:32:28"}
+{"current_steps": 1880, "total_steps": 2350, "loss": 1.9372, "learning_rate": 1.1111111111111112e-05, "epoch": 1.5992344090594928, "percentage": 80.0, "elapsed_time": "10:03:31", "remaining_time": "2:30:52"}
+{"current_steps": 1885, "total_steps": 2350, "loss": 1.9623, "learning_rate": 1.0992907801418441e-05, "epoch": 1.6034876920623105, "percentage": 80.21, "elapsed_time": "10:05:07", "remaining_time": "2:29:16"}
+{"current_steps": 1890, "total_steps": 2350, "loss": 1.9176, "learning_rate": 1.0874704491725769e-05, "epoch": 1.6077409750651284, "percentage": 80.43, "elapsed_time": "10:06:44", "remaining_time": "2:27:40"}
+{"current_steps": 1895, "total_steps": 2350, "loss": 1.9251, "learning_rate": 1.0756501182033098e-05, "epoch": 1.6119942580679463, "percentage": 80.64, "elapsed_time": "10:08:21", "remaining_time": "2:26:04"}
+{"current_steps": 1900, "total_steps": 2350, "loss": 1.9609, "learning_rate": 1.0638297872340426e-05, "epoch": 1.616247541070764, "percentage": 80.85, "elapsed_time": "10:09:57", "remaining_time": "2:24:27"}
+{"current_steps": 1905, "total_steps": 2350, "loss": 1.9602, "learning_rate": 1.0520094562647754e-05, "epoch": 1.6205008240735816, "percentage": 81.06, "elapsed_time": "10:11:35", "remaining_time": "2:22:51"}
+{"current_steps": 1910, "total_steps": 2350, "loss": 1.9371, "learning_rate": 1.0401891252955083e-05, "epoch": 1.6247541070763996, "percentage": 81.28, "elapsed_time": "10:13:11", "remaining_time": "2:21:15"}
+{"current_steps": 1915, "total_steps": 2350, "loss": 1.9796, "learning_rate": 1.0283687943262411e-05, "epoch": 1.6290073900792175, "percentage": 81.49, "elapsed_time": "10:14:48", "remaining_time": "2:19:39"}
+{"current_steps": 1920, "total_steps": 2350, "loss": 1.943, "learning_rate": 1.016548463356974e-05, "epoch": 1.6332606730820352, "percentage": 81.7, "elapsed_time": "10:16:24", "remaining_time": "2:18:02"}
+{"current_steps": 1925, "total_steps": 2350, "loss": 1.933, "learning_rate": 1.004728132387707e-05, "epoch": 1.6375139560848528, "percentage": 81.91, "elapsed_time": "10:18:00", "remaining_time": "2:16:26"}
+{"current_steps": 1930, "total_steps": 2350, "loss": 1.9473, "learning_rate": 9.929078014184398e-06, "epoch": 1.6417672390876707, "percentage": 82.13, "elapsed_time": "10:19:37", "remaining_time": "2:14:50"}
+{"current_steps": 1935, "total_steps": 2350, "loss": 1.9453, "learning_rate": 9.810874704491727e-06, "epoch": 1.6460205220904887, "percentage": 82.34, "elapsed_time": "10:21:13", "remaining_time": "2:13:14"}
+{"current_steps": 1940, "total_steps": 2350, "loss": 1.9359, "learning_rate": 9.692671394799055e-06, "epoch": 1.6502738050933063, "percentage": 82.55, "elapsed_time": "10:22:49", "remaining_time": "2:11:37"}
+{"current_steps": 1945, "total_steps": 2350, "loss": 1.9362, "learning_rate": 9.574468085106383e-06, "epoch": 1.654527088096124, "percentage": 82.77, "elapsed_time": "10:24:25", "remaining_time": "2:10:01"}
+{"current_steps": 1950, "total_steps": 2350, "loss": 1.9583, "learning_rate": 9.456264775413712e-06, "epoch": 1.658780371098942, "percentage": 82.98, "elapsed_time": "10:26:01", "remaining_time": "2:08:24"}
+{"current_steps": 1955, "total_steps": 2350, "loss": 1.9398, "learning_rate": 9.33806146572104e-06, "epoch": 1.6630336541017599, "percentage": 83.19, "elapsed_time": "10:27:37", "remaining_time": "2:06:48"}
+{"current_steps": 1960, "total_steps": 2350, "loss": 1.9631, "learning_rate": 9.219858156028368e-06, "epoch": 1.6672869371045778, "percentage": 83.4, "elapsed_time": "10:29:14", "remaining_time": "2:05:12"}
+{"current_steps": 1965, "total_steps": 2350, "loss": 1.9779, "learning_rate": 9.101654846335699e-06, "epoch": 1.6715402201073954, "percentage": 83.62, "elapsed_time": "10:30:49", "remaining_time": "2:03:35"}
+{"current_steps": 1970, "total_steps": 2350, "loss": 1.9397, "learning_rate": 8.983451536643027e-06, "epoch": 1.6757935031102131, "percentage": 83.83, "elapsed_time": "10:32:25", "remaining_time": "2:01:59"}
+{"current_steps": 1975, "total_steps": 2350, "loss": 1.9447, "learning_rate": 8.865248226950355e-06, "epoch": 1.680046786113031, "percentage": 84.04, "elapsed_time": "10:34:01", "remaining_time": "2:00:23"}
+{"current_steps": 1980, "total_steps": 2350, "loss": 1.9395, "learning_rate": 8.747044917257684e-06, "epoch": 1.684300069115849, "percentage": 84.26, "elapsed_time": "10:35:37", "remaining_time": "1:58:46"}
+{"current_steps": 1985, "total_steps": 2350, "loss": 1.9629, "learning_rate": 8.628841607565012e-06, "epoch": 1.6885533521186666, "percentage": 84.47, "elapsed_time": "10:37:13", "remaining_time": "1:57:10"}
+{"current_steps": 1990, "total_steps": 2350, "loss": 1.9608, "learning_rate": 8.510638297872341e-06, "epoch": 1.6928066351214843, "percentage": 84.68, "elapsed_time": "10:38:49", "remaining_time": "1:55:33"}
+{"current_steps": 1995, "total_steps": 2350, "loss": 1.9572, "learning_rate": 8.392434988179669e-06, "epoch": 1.6970599181243022, "percentage": 84.89, "elapsed_time": "10:40:25", "remaining_time": "1:53:57"}
+{"current_steps": 2000, "total_steps": 2350, "loss": 1.9234, "learning_rate": 8.274231678486997e-06, "epoch": 1.7013132011271201, "percentage": 85.11, "elapsed_time": "10:42:00", "remaining_time": "1:52:21"}
+{"current_steps": 2005, "total_steps": 2350, "loss": 1.9562, "learning_rate": 8.156028368794328e-06, "epoch": 1.7055664841299378, "percentage": 85.32, "elapsed_time": "10:45:16", "remaining_time": "1:51:01"}
+{"current_steps": 2010, "total_steps": 2350, "loss": 1.9415, "learning_rate": 8.037825059101656e-06, "epoch": 1.7098197671327555, "percentage": 85.53, "elapsed_time": "10:46:52", "remaining_time": "1:49:25"}
+{"current_steps": 2015, "total_steps": 2350, "loss": 1.9663, "learning_rate": 7.919621749408983e-06, "epoch": 1.7140730501355734, "percentage": 85.74, "elapsed_time": "10:48:29", "remaining_time": "1:47:48"}
+{"current_steps": 2020, "total_steps": 2350, "loss": 1.9647, "learning_rate": 7.801418439716313e-06, "epoch": 1.7183263331383913, "percentage": 85.96, "elapsed_time": "10:50:06", "remaining_time": "1:46:12"}
+{"current_steps": 2025, "total_steps": 2350, "loss": 1.9552, "learning_rate": 7.68321513002364e-06, "epoch": 1.722579616141209, "percentage": 86.17, "elapsed_time": "10:51:42", "remaining_time": "1:44:35"}
+{"current_steps": 2030, "total_steps": 2350, "loss": 1.9452, "learning_rate": 7.565011820330969e-06, "epoch": 1.7268328991440267, "percentage": 86.38, "elapsed_time": "10:53:18", "remaining_time": "1:42:59"}
+{"current_steps": 2035, "total_steps": 2350, "loss": 1.9519, "learning_rate": 7.446808510638298e-06, "epoch": 1.7310861821468446, "percentage": 86.6, "elapsed_time": "10:54:54", "remaining_time": "1:41:22"}
+{"current_steps": 2040, "total_steps": 2350, "loss": 1.9613, "learning_rate": 7.3286052009456266e-06, "epoch": 1.7353394651496625, "percentage": 86.81, "elapsed_time": "10:56:30", "remaining_time": "1:39:45"}
+{"current_steps": 2045, "total_steps": 2350, "loss": 1.9412, "learning_rate": 7.210401891252956e-06, "epoch": 1.7395927481524802, "percentage": 87.02, "elapsed_time": "10:58:05", "remaining_time": "1:38:09"}
+{"current_steps": 2050, "total_steps": 2350, "loss": 1.9559, "learning_rate": 7.092198581560285e-06, "epoch": 1.743846031155298, "percentage": 87.23, "elapsed_time": "10:59:41", "remaining_time": "1:36:32"}
+{"current_steps": 2055, "total_steps": 2350, "loss": 1.9794, "learning_rate": 6.973995271867613e-06, "epoch": 1.7480993141581158, "percentage": 87.45, "elapsed_time": "11:01:18", "remaining_time": "1:34:55"}
+{"current_steps": 2060, "total_steps": 2350, "loss": 1.9788, "learning_rate": 6.855791962174941e-06, "epoch": 1.7523525971609337, "percentage": 87.66, "elapsed_time": "11:02:53", "remaining_time": "1:33:19"}
+{"current_steps": 2065, "total_steps": 2350, "loss": 1.9228, "learning_rate": 6.73758865248227e-06, "epoch": 1.7566058801637514, "percentage": 87.87, "elapsed_time": "11:04:29", "remaining_time": "1:31:42"}
+{"current_steps": 2070, "total_steps": 2350, "loss": 1.9189, "learning_rate": 6.619385342789598e-06, "epoch": 1.760859163166569, "percentage": 88.09, "elapsed_time": "11:06:04", "remaining_time": "1:30:05"}
+{"current_steps": 2075, "total_steps": 2350, "loss": 1.9285, "learning_rate": 6.501182033096927e-06, "epoch": 1.765112446169387, "percentage": 88.3, "elapsed_time": "11:07:40", "remaining_time": "1:28:29"}
+{"current_steps": 2080, "total_steps": 2350, "loss": 1.9816, "learning_rate": 6.3829787234042555e-06, "epoch": 1.769365729172205, "percentage": 88.51, "elapsed_time": "11:09:16", "remaining_time": "1:26:52"}
+{"current_steps": 2085, "total_steps": 2350, "loss": 1.9315, "learning_rate": 6.264775413711583e-06, "epoch": 1.7736190121750226, "percentage": 88.72, "elapsed_time": "11:10:52", "remaining_time": "1:25:15"}
+{"current_steps": 2090, "total_steps": 2350, "loss": 1.9809, "learning_rate": 6.146572104018913e-06, "epoch": 1.7778722951778403, "percentage": 88.94, "elapsed_time": "11:12:27", "remaining_time": "1:23:39"}
+{"current_steps": 2095, "total_steps": 2350, "loss": 1.9732, "learning_rate": 6.028368794326241e-06, "epoch": 1.7821255781806582, "percentage": 89.15, "elapsed_time": "11:14:03", "remaining_time": "1:22:02"}
+{"current_steps": 2100, "total_steps": 2350, "loss": 1.9602, "learning_rate": 5.91016548463357e-06, "epoch": 1.786378861183476, "percentage": 89.36, "elapsed_time": "11:15:38", "remaining_time": "1:20:26"}
+{"current_steps": 2105, "total_steps": 2350, "loss": 1.941, "learning_rate": 5.791962174940899e-06, "epoch": 1.7906321441862938, "percentage": 89.57, "elapsed_time": "11:17:14", "remaining_time": "1:18:49"}
+{"current_steps": 2110, "total_steps": 2350, "loss": 1.9833, "learning_rate": 5.673758865248227e-06, "epoch": 1.7948854271891115, "percentage": 89.79, "elapsed_time": "11:18:49", "remaining_time": "1:17:12"}
+{"current_steps": 2115, "total_steps": 2350, "loss": 1.9538, "learning_rate": 5.555555555555556e-06, "epoch": 1.7991387101919294, "percentage": 90.0, "elapsed_time": "11:20:25", "remaining_time": "1:15:36"}
+{"current_steps": 2120, "total_steps": 2350, "loss": 1.944, "learning_rate": 5.4373522458628844e-06, "epoch": 1.8033919931947473, "percentage": 90.21, "elapsed_time": "11:22:01", "remaining_time": "1:13:59"}
+{"current_steps": 2125, "total_steps": 2350, "loss": 1.9467, "learning_rate": 5.319148936170213e-06, "epoch": 1.807645276197565, "percentage": 90.43, "elapsed_time": "11:23:36", "remaining_time": "1:12:22"}
+{"current_steps": 2130, "total_steps": 2350, "loss": 1.9437, "learning_rate": 5.200945626477542e-06, "epoch": 1.8118985592003827, "percentage": 90.64, "elapsed_time": "11:25:12", "remaining_time": "1:10:46"}
+{"current_steps": 2135, "total_steps": 2350, "loss": 1.9043, "learning_rate": 5.08274231678487e-06, "epoch": 1.8161518422032006, "percentage": 90.85, "elapsed_time": "11:26:47", "remaining_time": "1:09:09"}
+{"current_steps": 2140, "total_steps": 2350, "loss": 1.9349, "learning_rate": 4.964539007092199e-06, "epoch": 1.8204051252060185, "percentage": 91.06, "elapsed_time": "11:28:23", "remaining_time": "1:07:33"}
+{"current_steps": 2145, "total_steps": 2350, "loss": 1.9543, "learning_rate": 4.8463356973995275e-06, "epoch": 1.8246584082088362, "percentage": 91.28, "elapsed_time": "11:29:58", "remaining_time": "1:05:56"}
+{"current_steps": 2150, "total_steps": 2350, "loss": 1.9475, "learning_rate": 4.728132387706856e-06, "epoch": 1.8289116912116539, "percentage": 91.49, "elapsed_time": "11:31:34", "remaining_time": "1:04:19"}
+{"current_steps": 2155, "total_steps": 2350, "loss": 1.8929, "learning_rate": 4.609929078014184e-06, "epoch": 1.8331649742144718, "percentage": 91.7, "elapsed_time": "11:33:09", "remaining_time": "1:02:43"}
+{"current_steps": 2160, "total_steps": 2350, "loss": 1.95, "learning_rate": 4.491725768321513e-06, "epoch": 1.8374182572172897, "percentage": 91.91, "elapsed_time": "11:34:44", "remaining_time": "1:01:06"}
+{"current_steps": 2165, "total_steps": 2350, "loss": 1.9575, "learning_rate": 4.373522458628842e-06, "epoch": 1.8416715402201074, "percentage": 92.13, "elapsed_time": "11:36:20", "remaining_time": "0:59:30"}
+{"current_steps": 2170, "total_steps": 2350, "loss": 1.8995, "learning_rate": 4.255319148936171e-06, "epoch": 1.845924823222925, "percentage": 92.34, "elapsed_time": "11:37:56", "remaining_time": "0:57:53"}
+{"current_steps": 2175, "total_steps": 2350, "loss": 1.9353, "learning_rate": 4.137115839243498e-06, "epoch": 1.850178106225743, "percentage": 92.55, "elapsed_time": "11:39:31", "remaining_time": "0:56:17"}
+{"current_steps": 2180, "total_steps": 2350, "loss": 1.95, "learning_rate": 4.018912529550828e-06, "epoch": 1.8544313892285609, "percentage": 92.77, "elapsed_time": "11:41:07", "remaining_time": "0:54:40"}
+{"current_steps": 2185, "total_steps": 2350, "loss": 1.9362, "learning_rate": 3.9007092198581565e-06, "epoch": 1.8586846722313786, "percentage": 92.98, "elapsed_time": "11:42:42", "remaining_time": "0:53:03"}
+{"current_steps": 2190, "total_steps": 2350, "loss": 1.9492, "learning_rate": 3.7825059101654847e-06, "epoch": 1.8629379552341963, "percentage": 93.19, "elapsed_time": "11:44:21", "remaining_time": "0:51:27"}
+{"current_steps": 2195, "total_steps": 2350, "loss": 1.928, "learning_rate": 3.6643026004728133e-06, "epoch": 1.8671912382370142, "percentage": 93.4, "elapsed_time": "11:45:58", "remaining_time": "0:49:51"}
+{"current_steps": 2200, "total_steps": 2350, "loss": 1.9407, "learning_rate": 3.5460992907801423e-06, "epoch": 1.871444521239832, "percentage": 93.62, "elapsed_time": "11:47:35", "remaining_time": "0:48:14"}
+{"current_steps": 2205, "total_steps": 2350, "loss": 1.9631, "learning_rate": 3.4278959810874705e-06, "epoch": 1.8756978042426498, "percentage": 93.83, "elapsed_time": "11:49:12", "remaining_time": "0:46:38"}
+{"current_steps": 2210, "total_steps": 2350, "loss": 1.9274, "learning_rate": 3.309692671394799e-06, "epoch": 1.8799510872454674, "percentage": 94.04, "elapsed_time": "11:50:49", "remaining_time": "0:45:01"}
+{"current_steps": 2215, "total_steps": 2350, "loss": 1.9329, "learning_rate": 3.1914893617021277e-06, "epoch": 1.8842043702482854, "percentage": 94.26, "elapsed_time": "11:52:25", "remaining_time": "0:43:25"}
+{"current_steps": 2220, "total_steps": 2350, "loss": 1.918, "learning_rate": 3.0732860520094564e-06, "epoch": 1.8884576532511033, "percentage": 94.47, "elapsed_time": "11:54:02", "remaining_time": "0:41:48"}
+{"current_steps": 2225, "total_steps": 2350, "loss": 1.9898, "learning_rate": 2.955082742316785e-06, "epoch": 1.892710936253921, "percentage": 94.68, "elapsed_time": "11:55:38", "remaining_time": "0:40:12"}
+{"current_steps": 2230, "total_steps": 2350, "loss": 1.9357, "learning_rate": 2.8368794326241136e-06, "epoch": 1.8969642192567386, "percentage": 94.89, "elapsed_time": "11:57:17", "remaining_time": "0:38:35"}
+{"current_steps": 2235, "total_steps": 2350, "loss": 1.9508, "learning_rate": 2.7186761229314422e-06, "epoch": 1.9012175022595565, "percentage": 95.11, "elapsed_time": "11:58:54", "remaining_time": "0:36:59"}
+{"current_steps": 2240, "total_steps": 2350, "loss": 1.9221, "learning_rate": 2.600472813238771e-06, "epoch": 1.9054707852623745, "percentage": 95.32, "elapsed_time": "12:00:29", "remaining_time": "0:35:22"}
+{"current_steps": 2245, "total_steps": 2350, "loss": 1.9175, "learning_rate": 2.4822695035460995e-06, "epoch": 1.9097240682651921, "percentage": 95.53, "elapsed_time": "12:02:05", "remaining_time": "0:33:46"}
+{"current_steps": 2250, "total_steps": 2350, "loss": 1.9834, "learning_rate": 2.364066193853428e-06, "epoch": 1.9139773512680098, "percentage": 95.74, "elapsed_time": "12:03:41", "remaining_time": "0:32:09"}
+{"current_steps": 2255, "total_steps": 2350, "loss": 1.9414, "learning_rate": 2.2458628841607567e-06, "epoch": 1.9182306342708277, "percentage": 95.96, "elapsed_time": "12:05:18", "remaining_time": "0:30:33"}
+{"current_steps": 2260, "total_steps": 2350, "loss": 1.9541, "learning_rate": 2.1276595744680853e-06, "epoch": 1.9224839172736456, "percentage": 96.17, "elapsed_time": "12:06:54", "remaining_time": "0:28:56"}
+{"current_steps": 2265, "total_steps": 2350, "loss": 1.9599, "learning_rate": 2.009456264775414e-06, "epoch": 1.9267372002764636, "percentage": 96.38, "elapsed_time": "12:08:30", "remaining_time": "0:27:20"}
+{"current_steps": 2270, "total_steps": 2350, "loss": 1.9898, "learning_rate": 1.8912529550827423e-06, "epoch": 1.9309904832792812, "percentage": 96.6, "elapsed_time": "12:10:06", "remaining_time": "0:25:43"}
+{"current_steps": 2275, "total_steps": 2350, "loss": 1.9505, "learning_rate": 1.7730496453900712e-06, "epoch": 1.935243766282099, "percentage": 96.81, "elapsed_time": "12:11:42", "remaining_time": "0:24:07"}
+{"current_steps": 2280, "total_steps": 2350, "loss": 1.9509, "learning_rate": 1.6548463356973996e-06, "epoch": 1.9394970492849168, "percentage": 97.02, "elapsed_time": "12:13:18", "remaining_time": "0:22:30"}
+{"current_steps": 2285, "total_steps": 2350, "loss": 1.9495, "learning_rate": 1.5366430260047282e-06, "epoch": 1.9437503322877348, "percentage": 97.23, "elapsed_time": "12:14:54", "remaining_time": "0:20:54"}
+{"current_steps": 2290, "total_steps": 2350, "loss": 1.9543, "learning_rate": 1.4184397163120568e-06, "epoch": 1.9480036152905524, "percentage": 97.45, "elapsed_time": "12:16:29", "remaining_time": "0:19:17"}
+{"current_steps": 2295, "total_steps": 2350, "loss": 1.9254, "learning_rate": 1.3002364066193854e-06, "epoch": 1.9522568982933701, "percentage": 97.66, "elapsed_time": "12:18:05", "remaining_time": "0:17:41"}
+{"current_steps": 2300, "total_steps": 2350, "loss": 1.8833, "learning_rate": 1.182033096926714e-06, "epoch": 1.956510181296188, "percentage": 97.87, "elapsed_time": "12:19:41", "remaining_time": "0:16:04"}
+{"current_steps": 2305, "total_steps": 2350, "loss": 1.9878, "learning_rate": 1.0638297872340427e-06, "epoch": 1.960763464299006, "percentage": 98.09, "elapsed_time": "12:21:16", "remaining_time": "0:14:28"}
+{"current_steps": 2310, "total_steps": 2350, "loss": 1.9668, "learning_rate": 9.456264775413712e-07, "epoch": 1.9650167473018236, "percentage": 98.3, "elapsed_time": "12:22:52", "remaining_time": "0:12:51"}
+{"current_steps": 2315, "total_steps": 2350, "loss": 1.9359, "learning_rate": 8.274231678486998e-07, "epoch": 1.9692700303046413, "percentage": 98.51, "elapsed_time": "12:24:27", "remaining_time": "0:11:15"}
+{"current_steps": 2320, "total_steps": 2350, "loss": 1.947, "learning_rate": 7.092198581560284e-07, "epoch": 1.9735233133074592, "percentage": 98.72, "elapsed_time": "12:26:03", "remaining_time": "0:09:38"}
+{"current_steps": 2325, "total_steps": 2350, "loss": 1.9126, "learning_rate": 5.91016548463357e-07, "epoch": 1.9777765963102771, "percentage": 98.94, "elapsed_time": "12:27:39", "remaining_time": "0:08:02"}
+{"current_steps": 2330, "total_steps": 2350, "loss": 1.9494, "learning_rate": 4.728132387706856e-07, "epoch": 1.9820298793130948, "percentage": 99.15, "elapsed_time": "12:29:15", "remaining_time": "0:06:25"}
+{"current_steps": 2335, "total_steps": 2350, "loss": 1.9494, "learning_rate": 3.546099290780142e-07, "epoch": 1.9862831623159125, "percentage": 99.36, "elapsed_time": "12:30:50", "remaining_time": "0:04:49"}
+{"current_steps": 2340, "total_steps": 2350, "loss": 1.9362, "learning_rate": 2.364066193853428e-07, "epoch": 1.9905364453187304, "percentage": 99.57, "elapsed_time": "12:32:25", "remaining_time": "0:03:12"}
+{"current_steps": 2345, "total_steps": 2350, "loss": 1.9238, "learning_rate": 1.182033096926714e-07, "epoch": 1.9947897283215483, "percentage": 99.79, "elapsed_time": "12:34:01", "remaining_time": "0:01:36"}
+{"current_steps": 2350, "total_steps": 2350, "loss": 1.9951, "learning_rate": 0.0, "epoch": 1.999043011324366, "percentage": 100.0, "elapsed_time": "12:35:37", "remaining_time": "0:00:00"}
+{"current_steps": 2350, "total_steps": 2350, "epoch": 1.999043011324366, "percentage": 100.0, "elapsed_time": "12:37:19", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,3332 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.999043011324366,
+  "eval_steps": 500,
+  "global_step": 2350,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0042532830028178,
+      "grad_norm": 360.0,
+      "learning_rate": 1.0638297872340427e-06,
+      "loss": 7.8552,
+      "step": 5
+    },
+    {
+      "epoch": 0.0085065660056356,
+      "grad_norm": 109.0,
+      "learning_rate": 2.1276595744680853e-06,
+      "loss": 6.1537,
+      "step": 10
+    },
+    {
+      "epoch": 0.0127598490084534,
+      "grad_norm": 33.25,
+      "learning_rate": 3.1914893617021277e-06,
+      "loss": 4.3481,
+      "step": 15
+    },
+    {
+      "epoch": 0.0170131320112712,
+      "grad_norm": 8.1875,
+      "learning_rate": 4.255319148936171e-06,
+      "loss": 3.4135,
+      "step": 20
+    },
+    {
+      "epoch": 0.021266415014089,
+      "grad_norm": 7.21875,
+      "learning_rate": 5.319148936170213e-06,
+      "loss": 2.9841,
+      "step": 25
+    },
+    {
+      "epoch": 0.0255196980169068,
+      "grad_norm": 3.328125,
+      "learning_rate": 6.3829787234042555e-06,
+      "loss": 2.8886,
+      "step": 30
+    },
+    {
+      "epoch": 0.0297729810197246,
+      "grad_norm": 2.40625,
+      "learning_rate": 7.446808510638298e-06,
+      "loss": 2.7333,
+      "step": 35
+    },
+    {
+      "epoch": 0.0340262640225424,
+      "grad_norm": 2.171875,
+      "learning_rate": 8.510638297872341e-06,
+      "loss": 2.7314,
+      "step": 40
+    },
+    {
+      "epoch": 0.0382795470253602,
+      "grad_norm": 2.375,
+      "learning_rate": 9.574468085106383e-06,
+      "loss": 2.7024,
+      "step": 45
+    },
+    {
+      "epoch": 0.042532830028178,
+      "grad_norm": 1.9296875,
+      "learning_rate": 1.0638297872340426e-05,
+      "loss": 2.7374,
+      "step": 50
+    },
+    {
+      "epoch": 0.0467861130309958,
+      "grad_norm": 1.9921875,
+      "learning_rate": 1.170212765957447e-05,
+      "loss": 2.6406,
+      "step": 55
+    },
+    {
+      "epoch": 0.0510393960338136,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.2765957446808511e-05,
+      "loss": 2.658,
+      "step": 60
+    },
+    {
+      "epoch": 0.0552926790366314,
+      "grad_norm": 2.265625,
+      "learning_rate": 1.3829787234042554e-05,
+      "loss": 2.6501,
+      "step": 65
+    },
+    {
+      "epoch": 0.0595459620394492,
+      "grad_norm": 1.9453125,
+      "learning_rate": 1.4893617021276596e-05,
+      "loss": 2.6481,
+      "step": 70
+    },
+    {
+      "epoch": 0.063799245042267,
+      "grad_norm": 2.171875,
+      "learning_rate": 1.595744680851064e-05,
+      "loss": 2.6195,
+      "step": 75
+    },
+    {
+      "epoch": 0.0680525280450848,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.7021276595744682e-05,
+      "loss": 2.6704,
+      "step": 80
+    },
+    {
+      "epoch": 0.0723058110479026,
+      "grad_norm": 1.890625,
+      "learning_rate": 1.8085106382978724e-05,
+      "loss": 2.6392,
+      "step": 85
+    },
+    {
+      "epoch": 0.0765590940507204,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.9148936170212766e-05,
+      "loss": 2.6517,
+      "step": 90
+    },
+    {
+      "epoch": 0.0808123770535382,
+      "grad_norm": 2.703125,
+      "learning_rate": 2.0212765957446807e-05,
+      "loss": 2.6277,
+      "step": 95
+    },
+    {
+      "epoch": 0.085065660056356,
+      "grad_norm": 2.125,
+      "learning_rate": 2.1276595744680852e-05,
+      "loss": 2.6178,
+      "step": 100
+    },
+    {
+      "epoch": 0.0893189430591738,
+      "grad_norm": 2.015625,
+      "learning_rate": 2.2340425531914894e-05,
+      "loss": 2.6114,
+      "step": 105
+    },
+    {
+      "epoch": 0.0935722260619916,
+      "grad_norm": 2.328125,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 2.583,
+      "step": 110
+    },
+    {
+      "epoch": 0.0978255090648094,
+      "grad_norm": 3.375,
+      "learning_rate": 2.446808510638298e-05,
+      "loss": 2.585,
+      "step": 115
+    },
+    {
+      "epoch": 0.1020787920676272,
+      "grad_norm": 2.796875,
+      "learning_rate": 2.5531914893617022e-05,
+      "loss": 2.5928,
+      "step": 120
+    },
+    {
+      "epoch": 0.106332075070445,
+      "grad_norm": 2.34375,
+      "learning_rate": 2.6595744680851064e-05,
+      "loss": 2.5695,
+      "step": 125
+    },
+    {
+      "epoch": 0.1105853580732628,
+      "grad_norm": 2.390625,
+      "learning_rate": 2.765957446808511e-05,
+      "loss": 2.5822,
+      "step": 130
+    },
+    {
+      "epoch": 0.1148386410760806,
+      "grad_norm": 4.28125,
+      "learning_rate": 2.8723404255319154e-05,
+      "loss": 2.6068,
+      "step": 135
+    },
+    {
+      "epoch": 0.1190919240788984,
+      "grad_norm": 2.90625,
+      "learning_rate": 2.9787234042553192e-05,
+      "loss": 2.6135,
+      "step": 140
+    },
+    {
+      "epoch": 0.1233452070817162,
+      "grad_norm": 2.109375,
+      "learning_rate": 3.085106382978723e-05,
+      "loss": 2.6274,
+      "step": 145
+    },
+    {
+      "epoch": 0.127598490084534,
+      "grad_norm": 2.125,
+      "learning_rate": 3.191489361702128e-05,
+      "loss": 2.5456,
+      "step": 150
+    },
+    {
+      "epoch": 0.1318517730873518,
+      "grad_norm": 3.71875,
+      "learning_rate": 3.2978723404255317e-05,
+      "loss": 2.5587,
+      "step": 155
+    },
+    {
+      "epoch": 0.1361050560901696,
+      "grad_norm": 3.390625,
+      "learning_rate": 3.4042553191489365e-05,
+      "loss": 2.5825,
+      "step": 160
+    },
+    {
+      "epoch": 0.1403583390929874,
+      "grad_norm": 3.28125,
+      "learning_rate": 3.5106382978723407e-05,
+      "loss": 2.5997,
+      "step": 165
+    },
+    {
+      "epoch": 0.1446116220958052,
+      "grad_norm": 2.421875,
+      "learning_rate": 3.617021276595745e-05,
+      "loss": 2.5429,
+      "step": 170
+    },
+    {
+      "epoch": 0.148864905098623,
+      "grad_norm": 1.9375,
+      "learning_rate": 3.723404255319149e-05,
+      "loss": 2.5906,
+      "step": 175
+    },
+    {
+      "epoch": 0.1531181881014408,
+      "grad_norm": 2.03125,
+      "learning_rate": 3.829787234042553e-05,
+      "loss": 2.5513,
+      "step": 180
+    },
+    {
+      "epoch": 0.1573714711042586,
+      "grad_norm": 3.515625,
+      "learning_rate": 3.936170212765958e-05,
+      "loss": 2.582,
+      "step": 185
+    },
+    {
+      "epoch": 0.1616247541070764,
+      "grad_norm": 4.34375,
+      "learning_rate": 4.0425531914893614e-05,
+      "loss": 2.6428,
+      "step": 190
+    },
+    {
+      "epoch": 0.1658780371098942,
+      "grad_norm": 2.3125,
+      "learning_rate": 4.148936170212766e-05,
+      "loss": 2.5858,
+      "step": 195
+    },
+    {
+      "epoch": 0.170131320112712,
+      "grad_norm": 2.453125,
+      "learning_rate": 4.2553191489361704e-05,
+      "loss": 2.571,
+      "step": 200
+    },
+    {
+      "epoch": 0.1743846031155298,
+      "grad_norm": 2.28125,
+      "learning_rate": 4.3617021276595746e-05,
+      "loss": 2.5948,
+      "step": 205
+    },
+    {
+      "epoch": 0.1786378861183476,
+      "grad_norm": 1.953125,
+      "learning_rate": 4.468085106382979e-05,
+      "loss": 2.5763,
+      "step": 210
+    },
+    {
+      "epoch": 0.1828911691211654,
+      "grad_norm": 2.5625,
+      "learning_rate": 4.574468085106383e-05,
+      "loss": 2.5975,
+      "step": 215
+    },
+    {
+      "epoch": 0.1871444521239832,
+      "grad_norm": 2.4375,
+      "learning_rate": 4.680851063829788e-05,
+      "loss": 2.5627,
+      "step": 220
+    },
+    {
+      "epoch": 0.191397735126801,
+      "grad_norm": 3.28125,
+      "learning_rate": 4.787234042553192e-05,
+      "loss": 2.6303,
+      "step": 225
+    },
+    {
+      "epoch": 0.1956510181296188,
+      "grad_norm": 3.5,
+      "learning_rate": 4.893617021276596e-05,
+      "loss": 2.614,
+      "step": 230
+    },
+    {
+      "epoch": 0.1999043011324366,
+      "grad_norm": 3.171875,
+      "learning_rate": 5e-05,
+      "loss": 2.6077,
+      "step": 235
+    },
+    {
+      "epoch": 0.2041575841352544,
+      "grad_norm": 2.4375,
+      "learning_rate": 4.988179669030733e-05,
+      "loss": 2.5803,
+      "step": 240
+    },
+    {
+      "epoch": 0.2084108671380722,
+      "grad_norm": 2.0625,
+      "learning_rate": 4.976359338061466e-05,
+      "loss": 2.579,
+      "step": 245
+    },
+    {
+      "epoch": 0.21266415014089,
+      "grad_norm": 2.828125,
+      "learning_rate": 4.964539007092199e-05,
+      "loss": 2.5945,
+      "step": 250
+    },
+    {
+      "epoch": 0.2169174331437078,
+      "grad_norm": 2.359375,
+      "learning_rate": 4.9527186761229313e-05,
+      "loss": 2.6439,
+      "step": 255
+    },
+    {
+      "epoch": 0.2211707161465256,
+      "grad_norm": 1.875,
+      "learning_rate": 4.940898345153664e-05,
+      "loss": 2.5977,
+      "step": 260
+    },
+    {
+      "epoch": 0.2254239991493434,
+      "grad_norm": 1.703125,
+      "learning_rate": 4.929078014184397e-05,
+      "loss": 2.6129,
+      "step": 265
+    },
+    {
+      "epoch": 0.2296772821521612,
+      "grad_norm": 1.5234375,
+      "learning_rate": 4.91725768321513e-05,
+      "loss": 2.5694,
+      "step": 270
+    },
+    {
+      "epoch": 0.233930565154979,
+      "grad_norm": 1.7265625,
+      "learning_rate": 4.905437352245863e-05,
+      "loss": 2.5778,
+      "step": 275
+    },
+    {
+      "epoch": 0.2381838481577968,
+      "grad_norm": 2.21875,
+      "learning_rate": 4.893617021276596e-05,
+      "loss": 2.6525,
+      "step": 280
+    },
+    {
+      "epoch": 0.2424371311606146,
+      "grad_norm": 1.578125,
+      "learning_rate": 4.8817966903073283e-05,
+      "loss": 2.5845,
+      "step": 285
+    },
+    {
+      "epoch": 0.2466904141634324,
+      "grad_norm": 1.5859375,
+      "learning_rate": 4.869976359338061e-05,
+      "loss": 2.6101,
+      "step": 290
+    },
+    {
+      "epoch": 0.2509436971662502,
+      "grad_norm": 1.6484375,
+      "learning_rate": 4.858156028368794e-05,
+      "loss": 2.5458,
+      "step": 295
+    },
+    {
+      "epoch": 0.255196980169068,
+      "grad_norm": 1.9140625,
+      "learning_rate": 4.846335697399527e-05,
+      "loss": 2.5742,
+      "step": 300
+    },
+    {
+      "epoch": 0.2594502631718858,
+      "grad_norm": 1.609375,
+      "learning_rate": 4.83451536643026e-05,
+      "loss": 2.5727,
+      "step": 305
+    },
+    {
+      "epoch": 0.2637035461747036,
+      "grad_norm": 1.59375,
+      "learning_rate": 4.822695035460993e-05,
+      "loss": 2.5882,
+      "step": 310
+    },
+    {
+      "epoch": 0.2679568291775214,
+      "grad_norm": 1.5859375,
+      "learning_rate": 4.810874704491726e-05,
+      "loss": 2.5671,
+      "step": 315
+    },
+    {
+      "epoch": 0.2722101121803392,
+      "grad_norm": 1.75,
+      "learning_rate": 4.799054373522459e-05,
+      "loss": 2.622,
+      "step": 320
+    },
+    {
+      "epoch": 0.276463395183157,
+      "grad_norm": 1.578125,
+      "learning_rate": 4.787234042553192e-05,
+      "loss": 2.5931,
+      "step": 325
+    },
+    {
+      "epoch": 0.2807166781859748,
+      "grad_norm": 2.359375,
+      "learning_rate": 4.775413711583925e-05,
+      "loss": 2.5471,
+      "step": 330
+    },
+    {
+      "epoch": 0.2849699611887926,
+      "grad_norm": 1.734375,
+      "learning_rate": 4.763593380614658e-05,
+      "loss": 2.5485,
+      "step": 335
+    },
+    {
+      "epoch": 0.2892232441916104,
+      "grad_norm": 1.5625,
+      "learning_rate": 4.751773049645391e-05,
+      "loss": 2.5652,
+      "step": 340
+    },
+    {
+      "epoch": 0.2934765271944282,
+      "grad_norm": 1.6484375,
+      "learning_rate": 4.739952718676123e-05,
+      "loss": 2.617,
+      "step": 345
+    },
+    {
+      "epoch": 0.297729810197246,
+      "grad_norm": 1.4765625,
+      "learning_rate": 4.728132387706856e-05,
+      "loss": 2.5512,
+      "step": 350
+    },
+    {
+      "epoch": 0.3019830932000638,
+      "grad_norm": 1.3515625,
+      "learning_rate": 4.716312056737589e-05,
+      "loss": 2.5682,
+      "step": 355
+    },
+    {
+      "epoch": 0.3062363762028816,
+      "grad_norm": 1.296875,
+      "learning_rate": 4.704491725768322e-05,
+      "loss": 2.5709,
+      "step": 360
+    },
+    {
+      "epoch": 0.3104896592056994,
+      "grad_norm": 1.4921875,
+      "learning_rate": 4.692671394799055e-05,
+      "loss": 2.5834,
+      "step": 365
+    },
+    {
+      "epoch": 0.3147429422085172,
+      "grad_norm": 1.5859375,
+      "learning_rate": 4.680851063829788e-05,
+      "loss": 2.5736,
+      "step": 370
+    },
+    {
+      "epoch": 0.31899622521133497,
+      "grad_norm": 1.3671875,
+      "learning_rate": 4.669030732860521e-05,
+      "loss": 2.5321,
+      "step": 375
+    },
+    {
+      "epoch": 0.3232495082141528,
+      "grad_norm": 1.5234375,
+      "learning_rate": 4.657210401891253e-05,
+      "loss": 2.5701,
+      "step": 380
+    },
+    {
+      "epoch": 0.3275027912169706,
+      "grad_norm": 1.4453125,
+      "learning_rate": 4.645390070921986e-05,
+      "loss": 2.553,
+      "step": 385
+    },
+    {
+      "epoch": 0.3317560742197884,
+      "grad_norm": 1.5,
+      "learning_rate": 4.633569739952719e-05,
+      "loss": 2.5242,
+      "step": 390
+    },
+    {
+      "epoch": 0.3360093572226062,
+      "grad_norm": 1.2578125,
+      "learning_rate": 4.621749408983452e-05,
+      "loss": 2.5279,
+      "step": 395
+    },
+    {
+      "epoch": 0.340262640225424,
+      "grad_norm": 1.203125,
+      "learning_rate": 4.609929078014185e-05,
+      "loss": 2.607,
+      "step": 400
+    },
+    {
+      "epoch": 0.3445159232282418,
+      "grad_norm": 1.3515625,
+      "learning_rate": 4.598108747044918e-05,
+      "loss": 2.5556,
+      "step": 405
+    },
+    {
+      "epoch": 0.3487692062310596,
+      "grad_norm": 1.4921875,
+      "learning_rate": 4.58628841607565e-05,
+      "loss": 2.5468,
+      "step": 410
+    },
+    {
+      "epoch": 0.3530224892338774,
+      "grad_norm": 1.1875,
+      "learning_rate": 4.574468085106383e-05,
+      "loss": 2.5511,
+      "step": 415
+    },
+    {
+      "epoch": 0.3572757722366952,
+      "grad_norm": 1.28125,
+      "learning_rate": 4.562647754137116e-05,
+      "loss": 2.5438,
+      "step": 420
+    },
+    {
+      "epoch": 0.361529055239513,
+      "grad_norm": 1.2578125,
+      "learning_rate": 4.550827423167849e-05,
+      "loss": 2.5264,
+      "step": 425
+    },
+    {
+      "epoch": 0.3657823382423308,
+      "grad_norm": 1.1796875,
+      "learning_rate": 4.539007092198582e-05,
+      "loss": 2.5365,
+      "step": 430
+    },
+    {
+      "epoch": 0.3700356212451486,
+      "grad_norm": 1.3984375,
+      "learning_rate": 4.527186761229315e-05,
+      "loss": 2.5596,
+      "step": 435
+    },
+    {
+      "epoch": 0.3742889042479664,
+      "grad_norm": 1.1640625,
+      "learning_rate": 4.5153664302600477e-05,
+      "loss": 2.4894,
+      "step": 440
+    },
+    {
+      "epoch": 0.3785421872507842,
+      "grad_norm": 1.5390625,
+      "learning_rate": 4.50354609929078e-05,
+      "loss": 2.5441,
+      "step": 445
+    },
+    {
+      "epoch": 0.382795470253602,
+      "grad_norm": 1.1640625,
+      "learning_rate": 4.491725768321513e-05,
+      "loss": 2.5379,
+      "step": 450
+    },
+    {
+      "epoch": 0.3870487532564198,
+      "grad_norm": 1.28125,
+      "learning_rate": 4.479905437352246e-05,
+      "loss": 2.5194,
+      "step": 455
+    },
+    {
+      "epoch": 0.3913020362592376,
+      "grad_norm": 1.1640625,
+      "learning_rate": 4.468085106382979e-05,
+      "loss": 2.5767,
+      "step": 460
+    },
+    {
+      "epoch": 0.3955553192620554,
+      "grad_norm": 1.0859375,
+      "learning_rate": 4.456264775413712e-05,
+      "loss": 2.5485,
+      "step": 465
+    },
+    {
+      "epoch": 0.3998086022648732,
+      "grad_norm": 1.890625,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 2.547,
+      "step": 470
+    },
+    {
+      "epoch": 0.404061885267691,
+      "grad_norm": 1.53125,
+      "learning_rate": 4.432624113475177e-05,
+      "loss": 2.5278,
+      "step": 475
+    },
+    {
+      "epoch": 0.4083151682705088,
+      "grad_norm": 1.2421875,
+      "learning_rate": 4.4208037825059106e-05,
+      "loss": 2.5245,
+      "step": 480
+    },
+    {
+      "epoch": 0.4125684512733266,
+      "grad_norm": 1.671875,
+      "learning_rate": 4.4089834515366435e-05,
+      "loss": 2.5318,
+      "step": 485
+    },
+    {
+      "epoch": 0.4168217342761444,
+      "grad_norm": 1.125,
+      "learning_rate": 4.3971631205673764e-05,
+      "loss": 2.4747,
+      "step": 490
+    },
+    {
+      "epoch": 0.4210750172789622,
+      "grad_norm": 1.7578125,
+      "learning_rate": 4.3853427895981094e-05,
+      "loss": 2.4824,
+      "step": 495
+    },
+    {
+      "epoch": 0.42532830028178,
+      "grad_norm": 1.2265625,
+      "learning_rate": 4.373522458628842e-05,
+      "loss": 2.5236,
+      "step": 500
+    },
+    {
+      "epoch": 0.4295815832845978,
+      "grad_norm": 1.2421875,
+      "learning_rate": 4.3617021276595746e-05,
+      "loss": 2.5287,
+      "step": 505
+    },
+    {
+      "epoch": 0.4338348662874156,
+      "grad_norm": 1.15625,
+      "learning_rate": 4.3498817966903076e-05,
+      "loss": 2.4732,
+      "step": 510
+    },
+    {
+      "epoch": 0.4380881492902334,
+      "grad_norm": 1.296875,
+      "learning_rate": 4.3380614657210405e-05,
+      "loss": 2.5246,
+      "step": 515
+    },
+    {
+      "epoch": 0.4423414322930512,
+      "grad_norm": 1.1953125,
+      "learning_rate": 4.3262411347517734e-05,
+      "loss": 2.4755,
+      "step": 520
+    },
+    {
+      "epoch": 0.44659471529586897,
+      "grad_norm": 1.0234375,
+      "learning_rate": 4.3144208037825064e-05,
+      "loss": 2.5074,
+      "step": 525
+    },
+    {
+      "epoch": 0.4508479982986868,
+      "grad_norm": 1.859375,
+      "learning_rate": 4.302600472813239e-05,
+      "loss": 2.5037,
+      "step": 530
+    },
+    {
+      "epoch": 0.4551012813015046,
+      "grad_norm": 1.59375,
+      "learning_rate": 4.2907801418439716e-05,
+      "loss": 2.5041,
+      "step": 535
+    },
+    {
+      "epoch": 0.4593545643043224,
+      "grad_norm": 1.234375,
+      "learning_rate": 4.2789598108747046e-05,
+      "loss": 2.5235,
+      "step": 540
+    },
+    {
+      "epoch": 0.4636078473071402,
+      "grad_norm": 1.1171875,
+      "learning_rate": 4.2671394799054375e-05,
+      "loss": 2.5049,
+      "step": 545
+    },
+    {
+      "epoch": 0.467861130309958,
+      "grad_norm": 1.1015625,
+      "learning_rate": 4.2553191489361704e-05,
+      "loss": 2.5086,
+      "step": 550
+    },
+    {
+      "epoch": 0.4721144133127758,
+      "grad_norm": 1.09375,
+      "learning_rate": 4.2434988179669034e-05,
+      "loss": 2.5231,
+      "step": 555
+    },
+    {
+      "epoch": 0.4763676963155936,
+      "grad_norm": 0.96484375,
+      "learning_rate": 4.231678486997636e-05,
+      "loss": 2.4703,
+      "step": 560
+    },
+    {
+      "epoch": 0.4806209793184114,
+      "grad_norm": 1.0703125,
+      "learning_rate": 4.219858156028369e-05,
+      "loss": 2.4774,
+      "step": 565
+    },
+    {
+      "epoch": 0.4848742623212292,
+      "grad_norm": 1.1328125,
+      "learning_rate": 4.2080378250591016e-05,
+      "loss": 2.5054,
+      "step": 570
+    },
+    {
+      "epoch": 0.489127545324047,
+      "grad_norm": 1.0859375,
+      "learning_rate": 4.1962174940898345e-05,
+      "loss": 2.4538,
+      "step": 575
+    },
+    {
+      "epoch": 0.4933808283268648,
+      "grad_norm": 1.0390625,
+      "learning_rate": 4.1843971631205674e-05,
+      "loss": 2.5135,
+      "step": 580
+    },
+    {
+      "epoch": 0.4976341113296826,
+      "grad_norm": 1.046875,
+      "learning_rate": 4.1725768321513004e-05,
+      "loss": 2.4991,
+      "step": 585
+    },
+    {
+      "epoch": 0.5018873943325004,
+      "grad_norm": 1.2421875,
+      "learning_rate": 4.1607565011820333e-05,
+      "loss": 2.522,
+      "step": 590
+    },
+    {
+      "epoch": 0.5061406773353182,
+      "grad_norm": 1.0390625,
+      "learning_rate": 4.148936170212766e-05,
+      "loss": 2.4663,
+      "step": 595
+    },
+    {
+      "epoch": 0.510393960338136,
+      "grad_norm": 1.203125,
+      "learning_rate": 4.1371158392434986e-05,
+      "loss": 2.4471,
+      "step": 600
+    },
+    {
+      "epoch": 0.5146472433409538,
+      "grad_norm": 1.0234375,
+      "learning_rate": 4.1252955082742315e-05,
+      "loss": 2.5032,
+      "step": 605
+    },
+    {
+      "epoch": 0.5189005263437716,
+      "grad_norm": 0.98828125,
+      "learning_rate": 4.1134751773049644e-05,
+      "loss": 2.4995,
+      "step": 610
+    },
+    {
+      "epoch": 0.5231538093465894,
+      "grad_norm": 1.140625,
+      "learning_rate": 4.1016548463356974e-05,
+      "loss": 2.4837,
+      "step": 615
+    },
+    {
+      "epoch": 0.5274070923494072,
+      "grad_norm": 1.046875,
+      "learning_rate": 4.0898345153664303e-05,
+      "loss": 2.484,
+      "step": 620
+    },
+    {
+      "epoch": 0.531660375352225,
+      "grad_norm": 1.09375,
+      "learning_rate": 4.078014184397163e-05,
+      "loss": 2.4921,
+      "step": 625
+    },
+    {
+      "epoch": 0.5359136583550428,
+      "grad_norm": 1.0390625,
+      "learning_rate": 4.066193853427896e-05,
+      "loss": 2.5048,
+      "step": 630
+    },
+    {
+      "epoch": 0.5401669413578606,
+      "grad_norm": 1.0546875,
+      "learning_rate": 4.0543735224586285e-05,
+      "loss": 2.4843,
+      "step": 635
+    },
+    {
+      "epoch": 0.5444202243606784,
+      "grad_norm": 1.0234375,
+      "learning_rate": 4.0425531914893614e-05,
+      "loss": 2.4262,
+      "step": 640
+    },
+    {
+      "epoch": 0.5486735073634962,
+      "grad_norm": 1.0859375,
+      "learning_rate": 4.030732860520095e-05,
+      "loss": 2.4656,
+      "step": 645
+    },
+    {
+      "epoch": 0.552926790366314,
+      "grad_norm": 0.9609375,
+      "learning_rate": 4.018912529550828e-05,
+      "loss": 2.4967,
+      "step": 650
+    },
+    {
+      "epoch": 0.5571800733691318,
+      "grad_norm": 1.0078125,
+      "learning_rate": 4.007092198581561e-05,
+      "loss": 2.4475,
+      "step": 655
+    },
+    {
+      "epoch": 0.5614333563719496,
+      "grad_norm": 1.109375,
+      "learning_rate": 3.995271867612293e-05,
+      "loss": 2.4688,
+      "step": 660
+    },
+    {
+      "epoch": 0.5656866393747674,
+      "grad_norm": 1.0625,
+      "learning_rate": 3.983451536643026e-05,
+      "loss": 2.492,
+      "step": 665
+    },
+    {
+      "epoch": 0.5699399223775852,
+      "grad_norm": 1.09375,
+      "learning_rate": 3.971631205673759e-05,
+      "loss": 2.4977,
+      "step": 670
+    },
+    {
+      "epoch": 0.574193205380403,
+      "grad_norm": 1.125,
+      "learning_rate": 3.959810874704492e-05,
+      "loss": 2.4755,
+      "step": 675
+    },
+    {
+      "epoch": 0.5784464883832208,
+      "grad_norm": 1.0078125,
+      "learning_rate": 3.947990543735225e-05,
+      "loss": 2.513,
+      "step": 680
+    },
+    {
+      "epoch": 0.5826997713860386,
+      "grad_norm": 1.0078125,
+      "learning_rate": 3.936170212765958e-05,
+      "loss": 2.4759,
+      "step": 685
+    },
+    {
+      "epoch": 0.5869530543888564,
+      "grad_norm": 1.0078125,
+      "learning_rate": 3.924349881796691e-05,
+      "loss": 2.4487,
+      "step": 690
+    },
+    {
+      "epoch": 0.5912063373916742,
+      "grad_norm": 1.0,
+      "learning_rate": 3.912529550827423e-05,
+      "loss": 2.4394,
+      "step": 695
+    },
+    {
+      "epoch": 0.595459620394492,
+      "grad_norm": 0.953125,
+      "learning_rate": 3.900709219858156e-05,
+      "loss": 2.4777,
+      "step": 700
+    },
+    {
+      "epoch": 0.5997129033973098,
+      "grad_norm": 1.046875,
+      "learning_rate": 3.888888888888889e-05,
+      "loss": 2.4709,
+      "step": 705
+    },
+    {
+      "epoch": 0.6039661864001276,
+      "grad_norm": 1.0859375,
+      "learning_rate": 3.877068557919622e-05,
+      "loss": 2.4775,
+      "step": 710
+    },
+    {
+      "epoch": 0.6082194694029454,
+      "grad_norm": 0.93359375,
+      "learning_rate": 3.865248226950355e-05,
+      "loss": 2.4405,
+      "step": 715
+    },
+    {
+      "epoch": 0.6124727524057632,
+      "grad_norm": 0.95703125,
+      "learning_rate": 3.853427895981088e-05,
+      "loss": 2.3798,
+      "step": 720
+    },
+    {
+      "epoch": 0.616726035408581,
+      "grad_norm": 0.9140625,
+      "learning_rate": 3.84160756501182e-05,
+      "loss": 2.4553,
+      "step": 725
+    },
+    {
+      "epoch": 0.6209793184113988,
+      "grad_norm": 0.8828125,
+      "learning_rate": 3.829787234042553e-05,
+      "loss": 2.4596,
+      "step": 730
+    },
+    {
+      "epoch": 0.6252326014142165,
+      "grad_norm": 1.015625,
+      "learning_rate": 3.817966903073286e-05,
+      "loss": 2.494,
+      "step": 735
+    },
+    {
+      "epoch": 0.6294858844170343,
+      "grad_norm": 1.015625,
+      "learning_rate": 3.806146572104019e-05,
+      "loss": 2.3969,
+      "step": 740
+    },
+    {
+      "epoch": 0.6337391674198521,
+      "grad_norm": 0.94140625,
+      "learning_rate": 3.794326241134752e-05,
+      "loss": 2.4438,
+      "step": 745
+    },
+    {
+      "epoch": 0.6379924504226699,
+      "grad_norm": 0.9921875,
+      "learning_rate": 3.782505910165485e-05,
+      "loss": 2.4191,
+      "step": 750
+    },
+    {
+      "epoch": 0.6422457334254879,
+      "grad_norm": 0.97265625,
+      "learning_rate": 3.770685579196218e-05,
+      "loss": 2.4433,
+      "step": 755
+    },
+    {
+      "epoch": 0.6464990164283057,
+      "grad_norm": 1.171875,
+      "learning_rate": 3.75886524822695e-05,
+      "loss": 2.4597,
+      "step": 760
+    },
+    {
+      "epoch": 0.6507522994311234,
+      "grad_norm": 1.0546875,
+      "learning_rate": 3.747044917257683e-05,
+      "loss": 2.4633,
+      "step": 765
+    },
+    {
+      "epoch": 0.6550055824339412,
+      "grad_norm": 0.9296875,
+      "learning_rate": 3.735224586288416e-05,
+      "loss": 2.4522,
+      "step": 770
+    },
+    {
+      "epoch": 0.659258865436759,
+      "grad_norm": 0.91015625,
+      "learning_rate": 3.723404255319149e-05,
+      "loss": 2.4274,
+      "step": 775
+    },
+    {
+      "epoch": 0.6635121484395768,
+      "grad_norm": 1.0546875,
+      "learning_rate": 3.711583924349882e-05,
+      "loss": 2.4793,
+      "step": 780
+    },
+    {
+      "epoch": 0.6677654314423946,
+      "grad_norm": 0.9453125,
+      "learning_rate": 3.699763593380615e-05,
+      "loss": 2.4104,
+      "step": 785
+    },
+    {
+      "epoch": 0.6720187144452124,
+      "grad_norm": 1.2421875,
+      "learning_rate": 3.687943262411347e-05,
+      "loss": 2.4158,
+      "step": 790
+    },
+    {
+      "epoch": 0.6762719974480302,
+      "grad_norm": 1.0390625,
+      "learning_rate": 3.67612293144208e-05,
+      "loss": 2.4096,
+      "step": 795
+    },
+    {
+      "epoch": 0.680525280450848,
+      "grad_norm": 0.91015625,
+      "learning_rate": 3.664302600472813e-05,
+      "loss": 2.4041,
+      "step": 800
+    },
+    {
+      "epoch": 0.6847785634536658,
+      "grad_norm": 1.0703125,
+      "learning_rate": 3.6524822695035466e-05,
+      "loss": 2.4548,
+      "step": 805
+    },
+    {
+      "epoch": 0.6890318464564836,
+      "grad_norm": 1.015625,
+      "learning_rate": 3.6406619385342796e-05,
+      "loss": 2.395,
+      "step": 810
+    },
+    {
+      "epoch": 0.6932851294593014,
+      "grad_norm": 1.0703125,
+      "learning_rate": 3.6288416075650125e-05,
+      "loss": 2.4343,
+      "step": 815
+    },
+    {
+      "epoch": 0.6975384124621192,
+      "grad_norm": 0.9375,
+      "learning_rate": 3.617021276595745e-05,
+      "loss": 2.4177,
+      "step": 820
+    },
+    {
+      "epoch": 0.701791695464937,
+      "grad_norm": 0.90234375,
+      "learning_rate": 3.605200945626478e-05,
+      "loss": 2.3848,
+      "step": 825
+    },
+    {
+      "epoch": 0.7060449784677548,
+      "grad_norm": 0.88671875,
+      "learning_rate": 3.593380614657211e-05,
+      "loss": 2.4164,
+      "step": 830
+    },
+    {
+      "epoch": 0.7102982614705726,
+      "grad_norm": 0.8828125,
+      "learning_rate": 3.5815602836879437e-05,
+      "loss": 2.4032,
+      "step": 835
+    },
+    {
+      "epoch": 0.7145515444733904,
+      "grad_norm": 0.96484375,
+      "learning_rate": 3.5697399527186766e-05,
+      "loss": 2.3989,
+      "step": 840
+    },
+    {
+      "epoch": 0.7188048274762082,
+      "grad_norm": 1.03125,
+      "learning_rate": 3.5579196217494095e-05,
+      "loss": 2.3684,
+      "step": 845
+    },
+    {
+      "epoch": 0.723058110479026,
+      "grad_norm": 0.94140625,
+      "learning_rate": 3.546099290780142e-05,
+      "loss": 2.4081,
+      "step": 850
+    },
+    {
+      "epoch": 0.7273113934818438,
+      "grad_norm": 0.90234375,
+      "learning_rate": 3.534278959810875e-05,
+      "loss": 2.3928,
+      "step": 855
+    },
+    {
+      "epoch": 0.7315646764846616,
+      "grad_norm": 1.0703125,
+      "learning_rate": 3.522458628841608e-05,
+      "loss": 2.4376,
+      "step": 860
+    },
+    {
+      "epoch": 0.7358179594874794,
+      "grad_norm": 1.1171875,
+      "learning_rate": 3.5106382978723407e-05,
+      "loss": 2.4199,
+      "step": 865
+    },
+    {
+      "epoch": 0.7400712424902972,
+      "grad_norm": 0.94140625,
+      "learning_rate": 3.4988179669030736e-05,
+      "loss": 2.4299,
+      "step": 870
+    },
+    {
+      "epoch": 0.744324525493115,
+      "grad_norm": 0.98046875,
+      "learning_rate": 3.4869976359338065e-05,
+      "loss": 2.3908,
+      "step": 875
+    },
+    {
+      "epoch": 0.7485778084959328,
+      "grad_norm": 0.953125,
+      "learning_rate": 3.4751773049645395e-05,
+      "loss": 2.3904,
+      "step": 880
+    },
+    {
+      "epoch": 0.7528310914987506,
+      "grad_norm": 0.9375,
+      "learning_rate": 3.463356973995272e-05,
+      "loss": 2.3385,
+      "step": 885
+    },
+    {
+      "epoch": 0.7570843745015684,
+      "grad_norm": 1.0,
+      "learning_rate": 3.451536643026005e-05,
+      "loss": 2.3727,
+      "step": 890
+    },
+    {
+      "epoch": 0.7613376575043862,
+      "grad_norm": 0.91796875,
+      "learning_rate": 3.4397163120567377e-05,
+      "loss": 2.3772,
+      "step": 895
+    },
+    {
+      "epoch": 0.765590940507204,
+      "grad_norm": 0.92578125,
+      "learning_rate": 3.4278959810874706e-05,
+      "loss": 2.4052,
+      "step": 900
+    },
+    {
+      "epoch": 0.7698442235100218,
+      "grad_norm": 0.89453125,
+      "learning_rate": 3.4160756501182035e-05,
+      "loss": 2.3386,
+      "step": 905
+    },
+    {
+      "epoch": 0.7740975065128396,
+      "grad_norm": 0.87890625,
+      "learning_rate": 3.4042553191489365e-05,
+      "loss": 2.3723,
+      "step": 910
+    },
+    {
+      "epoch": 0.7783507895156574,
+      "grad_norm": 1.0390625,
+      "learning_rate": 3.392434988179669e-05,
+      "loss": 2.3892,
+      "step": 915
+    },
+    {
+      "epoch": 0.7826040725184752,
+      "grad_norm": 1.171875,
+      "learning_rate": 3.380614657210402e-05,
+      "loss": 2.4115,
+      "step": 920
+    },
+    {
+      "epoch": 0.786857355521293,
+      "grad_norm": 1.1015625,
+      "learning_rate": 3.3687943262411347e-05,
+      "loss": 2.377,
+      "step": 925
+    },
+    {
+      "epoch": 0.7911106385241108,
+      "grad_norm": 0.953125,
+      "learning_rate": 3.3569739952718676e-05,
+      "loss": 2.3725,
+      "step": 930
+    },
+    {
+      "epoch": 0.7953639215269286,
+      "grad_norm": 0.8515625,
+      "learning_rate": 3.3451536643026005e-05,
+      "loss": 2.3879,
+      "step": 935
+    },
+    {
+      "epoch": 0.7996172045297464,
+      "grad_norm": 0.88671875,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 2.3657,
+      "step": 940
+    },
+    {
+      "epoch": 0.8038704875325642,
+      "grad_norm": 0.90625,
+      "learning_rate": 3.3215130023640664e-05,
+      "loss": 2.3563,
+      "step": 945
+    },
+    {
+      "epoch": 0.808123770535382,
+      "grad_norm": 0.84375,
+      "learning_rate": 3.309692671394799e-05,
+      "loss": 2.3464,
+      "step": 950
+    },
+    {
+      "epoch": 0.8123770535381998,
+      "grad_norm": 0.8828125,
+      "learning_rate": 3.2978723404255317e-05,
+      "loss": 2.4217,
+      "step": 955
+    },
+    {
+      "epoch": 0.8166303365410176,
+      "grad_norm": 0.91015625,
+      "learning_rate": 3.2860520094562646e-05,
+      "loss": 2.3891,
+      "step": 960
+    },
+    {
+      "epoch": 0.8208836195438354,
+      "grad_norm": 0.875,
+      "learning_rate": 3.2742316784869975e-05,
+      "loss": 2.3703,
+      "step": 965
+    },
+    {
+      "epoch": 0.8251369025466532,
+      "grad_norm": 0.89453125,
+      "learning_rate": 3.262411347517731e-05,
+      "loss": 2.3914,
+      "step": 970
+    },
+    {
+      "epoch": 0.829390185549471,
+      "grad_norm": 0.9453125,
+      "learning_rate": 3.2505910165484634e-05,
+      "loss": 2.3948,
+      "step": 975
+    },
+    {
+      "epoch": 0.8336434685522888,
+      "grad_norm": 0.9140625,
+      "learning_rate": 3.2387706855791964e-05,
+      "loss": 2.397,
+      "step": 980
+    },
+    {
+      "epoch": 0.8378967515551066,
+      "grad_norm": 0.91796875,
+      "learning_rate": 3.226950354609929e-05,
+      "loss": 2.379,
+      "step": 985
+    },
+    {
+      "epoch": 0.8421500345579244,
+      "grad_norm": 0.8984375,
+      "learning_rate": 3.215130023640662e-05,
+      "loss": 2.366,
+      "step": 990
+    },
+    {
+      "epoch": 0.8464033175607422,
+      "grad_norm": 0.875,
+      "learning_rate": 3.203309692671395e-05,
+      "loss": 2.3653,
+      "step": 995
+    },
+    {
+      "epoch": 0.85065660056356,
+      "grad_norm": 0.86328125,
+      "learning_rate": 3.191489361702128e-05,
+      "loss": 2.3757,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8549098835663778,
+      "grad_norm": 0.88671875,
+      "learning_rate": 3.179669030732861e-05,
+      "loss": 2.3504,
+      "step": 1005
+    },
+    {
+      "epoch": 0.8591631665691956,
+      "grad_norm": 0.88671875,
+      "learning_rate": 3.1678486997635934e-05,
+      "loss": 2.367,
+      "step": 1010
+    },
+    {
+      "epoch": 0.8634164495720134,
+      "grad_norm": 1.34375,
+      "learning_rate": 3.156028368794326e-05,
+      "loss": 2.3557,
+      "step": 1015
+    },
+    {
+      "epoch": 0.8676697325748312,
+      "grad_norm": 1.03125,
+      "learning_rate": 3.144208037825059e-05,
+      "loss": 2.3673,
+      "step": 1020
+    },
+    {
+      "epoch": 0.871923015577649,
+      "grad_norm": 0.875,
+      "learning_rate": 3.132387706855792e-05,
+      "loss": 2.3407,
+      "step": 1025
+    },
+    {
+      "epoch": 0.8761762985804668,
+      "grad_norm": 0.8828125,
+      "learning_rate": 3.120567375886525e-05,
+      "loss": 2.3363,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8804295815832845,
+      "grad_norm": 0.92578125,
+      "learning_rate": 3.108747044917258e-05,
+      "loss": 2.3894,
+      "step": 1035
+    },
+    {
+      "epoch": 0.8846828645861023,
+      "grad_norm": 1.203125,
+      "learning_rate": 3.0969267139479904e-05,
+      "loss": 2.3537,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8889361475889201,
+      "grad_norm": 0.91015625,
+      "learning_rate": 3.085106382978723e-05,
+      "loss": 2.3588,
+      "step": 1045
+    },
+    {
+      "epoch": 0.8931894305917379,
+      "grad_norm": 0.87109375,
+      "learning_rate": 3.073286052009456e-05,
+      "loss": 2.3798,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8974427135945559,
+      "grad_norm": 0.8984375,
+      "learning_rate": 3.061465721040189e-05,
+      "loss": 2.364,
+      "step": 1055
+    },
+    {
+      "epoch": 0.9016959965973737,
+      "grad_norm": 0.828125,
+      "learning_rate": 3.0496453900709222e-05,
+      "loss": 2.3922,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9059492796001914,
+      "grad_norm": 0.86328125,
+      "learning_rate": 3.0378250591016548e-05,
+      "loss": 2.3189,
+      "step": 1065
+    },
+    {
+      "epoch": 0.9102025626030092,
+      "grad_norm": 0.83203125,
+      "learning_rate": 3.0260047281323877e-05,
+      "loss": 2.3871,
+      "step": 1070
+    },
+    {
+      "epoch": 0.914455845605827,
+      "grad_norm": 0.88671875,
+      "learning_rate": 3.0141843971631207e-05,
+      "loss": 2.3712,
+      "step": 1075
+    },
+    {
+      "epoch": 0.9187091286086448,
+      "grad_norm": 0.89453125,
+      "learning_rate": 3.0023640661938536e-05,
+      "loss": 2.3451,
+      "step": 1080
+    },
+    {
+      "epoch": 0.9229624116114626,
+      "grad_norm": 0.86328125,
+      "learning_rate": 2.9905437352245862e-05,
+      "loss": 2.2947,
+      "step": 1085
+    },
+    {
+      "epoch": 0.9272156946142804,
+      "grad_norm": 0.90625,
+      "learning_rate": 2.9787234042553192e-05,
+      "loss": 2.3742,
+      "step": 1090
+    },
+    {
+      "epoch": 0.9314689776170982,
+      "grad_norm": 1.03125,
+      "learning_rate": 2.966903073286052e-05,
+      "loss": 2.3482,
+      "step": 1095
+    },
+    {
+      "epoch": 0.935722260619916,
+      "grad_norm": 0.9375,
+      "learning_rate": 2.9550827423167847e-05,
+      "loss": 2.3385,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9399755436227338,
+      "grad_norm": 0.9453125,
+      "learning_rate": 2.9432624113475177e-05,
+      "loss": 2.3676,
+      "step": 1105
+    },
+    {
+      "epoch": 0.9442288266255516,
+      "grad_norm": 0.87109375,
+      "learning_rate": 2.9314420803782506e-05,
+      "loss": 2.3245,
+      "step": 1110
+    },
+    {
+      "epoch": 0.9484821096283694,
+      "grad_norm": 0.859375,
+      "learning_rate": 2.9196217494089832e-05,
+      "loss": 2.346,
+      "step": 1115
+    },
+    {
+      "epoch": 0.9527353926311872,
+      "grad_norm": 0.8359375,
+      "learning_rate": 2.9078014184397162e-05,
+      "loss": 2.3601,
+      "step": 1120
+    },
+    {
+      "epoch": 0.956988675634005,
+      "grad_norm": 0.8125,
+      "learning_rate": 2.895981087470449e-05,
+      "loss": 2.3,
+      "step": 1125
+    },
+    {
+      "epoch": 0.9612419586368228,
+      "grad_norm": 0.82421875,
+      "learning_rate": 2.8841607565011824e-05,
+      "loss": 2.3347,
+      "step": 1130
+    },
+    {
+      "epoch": 0.9654952416396406,
+      "grad_norm": 0.92578125,
+      "learning_rate": 2.8723404255319154e-05,
+      "loss": 2.3142,
+      "step": 1135
+    },
+    {
+      "epoch": 0.9697485246424584,
+      "grad_norm": 0.81640625,
+      "learning_rate": 2.8605200945626483e-05,
+      "loss": 2.3444,
+      "step": 1140
+    },
+    {
+      "epoch": 0.9740018076452762,
+      "grad_norm": 0.8671875,
+      "learning_rate": 2.848699763593381e-05,
+      "loss": 2.332,
+      "step": 1145
+    },
+    {
+      "epoch": 0.978255090648094,
+      "grad_norm": 0.94921875,
+      "learning_rate": 2.836879432624114e-05,
+      "loss": 2.3412,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9825083736509118,
+      "grad_norm": 0.8828125,
+      "learning_rate": 2.8250591016548468e-05,
+      "loss": 2.3245,
+      "step": 1155
+    },
+    {
+      "epoch": 0.9867616566537296,
+      "grad_norm": 0.8671875,
+      "learning_rate": 2.8132387706855794e-05,
+      "loss": 2.3507,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9910149396565474,
+      "grad_norm": 0.83203125,
+      "learning_rate": 2.8014184397163124e-05,
+      "loss": 2.3284,
+      "step": 1165
+    },
+    {
+      "epoch": 0.9952682226593652,
+      "grad_norm": 0.87890625,
+      "learning_rate": 2.7895981087470453e-05,
+      "loss": 2.3622,
+      "step": 1170
+    },
+    {
+      "epoch": 0.999521505662183,
+      "grad_norm": 0.87109375,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 2.3499,
+      "step": 1175
+    },
+    {
+      "epoch": 1.0037747886650008,
+      "grad_norm": 1.828125,
+      "learning_rate": 2.765957446808511e-05,
+      "loss": 2.0719,
+      "step": 1180
+    },
+    {
+      "epoch": 1.0080280716678185,
+      "grad_norm": 1.265625,
+      "learning_rate": 2.7541371158392438e-05,
+      "loss": 2.0455,
+      "step": 1185
+    },
+    {
+      "epoch": 1.0122813546706364,
+      "grad_norm": 0.9609375,
+      "learning_rate": 2.7423167848699764e-05,
+      "loss": 2.0007,
+      "step": 1190
+    },
+    {
+      "epoch": 1.016534637673454,
+      "grad_norm": 1.1796875,
+      "learning_rate": 2.7304964539007094e-05,
+      "loss": 2.0265,
+      "step": 1195
+    },
+    {
+      "epoch": 1.020787920676272,
+      "grad_norm": 1.1953125,
+      "learning_rate": 2.7186761229314423e-05,
+      "loss": 1.9858,
+      "step": 1200
+    },
+    {
+      "epoch": 1.02504120367909,
+      "grad_norm": 1.5390625,
+      "learning_rate": 2.7068557919621753e-05,
+      "loss": 2.0112,
+      "step": 1205
+    },
+    {
+      "epoch": 1.0292944866819076,
+      "grad_norm": 0.92578125,
+      "learning_rate": 2.695035460992908e-05,
+      "loss": 2.0078,
+      "step": 1210
+    },
+    {
+      "epoch": 1.0335477696847255,
+      "grad_norm": 1.1328125,
+      "learning_rate": 2.6832151300236408e-05,
+      "loss": 2.0072,
+      "step": 1215
+    },
+    {
+      "epoch": 1.0378010526875432,
+      "grad_norm": 1.015625,
+      "learning_rate": 2.6713947990543738e-05,
+      "loss": 1.9904,
+      "step": 1220
+    },
+    {
+      "epoch": 1.042054335690361,
+      "grad_norm": 0.93359375,
+      "learning_rate": 2.6595744680851064e-05,
+      "loss": 2.0318,
+      "step": 1225
+    },
+    {
+      "epoch": 1.0463076186931788,
+      "grad_norm": 0.91796875,
+      "learning_rate": 2.6477541371158393e-05,
+      "loss": 2.0214,
+      "step": 1230
+    },
+    {
+      "epoch": 1.0505609016959967,
+      "grad_norm": 0.9453125,
+      "learning_rate": 2.6359338061465723e-05,
+      "loss": 2.0233,
+      "step": 1235
+    },
+    {
+      "epoch": 1.0548141846988144,
+      "grad_norm": 1.0078125,
+      "learning_rate": 2.624113475177305e-05,
+      "loss": 1.9697,
+      "step": 1240
+    },
+    {
+      "epoch": 1.0590674677016323,
+      "grad_norm": 0.9453125,
+      "learning_rate": 2.6122931442080378e-05,
+      "loss": 1.9899,
+      "step": 1245
+    },
+    {
+      "epoch": 1.06332075070445,
+      "grad_norm": 0.92578125,
+      "learning_rate": 2.6004728132387708e-05,
+      "loss": 2.0306,
+      "step": 1250
+    },
+    {
+      "epoch": 1.0675740337072679,
+      "grad_norm": 0.97265625,
+      "learning_rate": 2.5886524822695034e-05,
+      "loss": 1.9851,
+      "step": 1255
+    },
+    {
+      "epoch": 1.0718273167100856,
+      "grad_norm": 1.1953125,
+      "learning_rate": 2.5768321513002363e-05,
+      "loss": 2.0191,
+      "step": 1260
+    },
+    {
+      "epoch": 1.0760805997129035,
+      "grad_norm": 1.1640625,
+      "learning_rate": 2.5650118203309693e-05,
+      "loss": 2.0307,
+      "step": 1265
+    },
+    {
+      "epoch": 1.0803338827157212,
+      "grad_norm": 1.0703125,
+      "learning_rate": 2.5531914893617022e-05,
+      "loss": 2.0119,
+      "step": 1270
+    },
+    {
+      "epoch": 1.084587165718539,
+      "grad_norm": 0.9375,
+      "learning_rate": 2.5413711583924348e-05,
+      "loss": 1.9794,
+      "step": 1275
+    },
+    {
+      "epoch": 1.0888404487213568,
+      "grad_norm": 1.15625,
+      "learning_rate": 2.5295508274231678e-05,
+      "loss": 2.0091,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0930937317241747,
+      "grad_norm": 1.0546875,
+      "learning_rate": 2.5177304964539007e-05,
+      "loss": 1.9977,
+      "step": 1285
+    },
+    {
+      "epoch": 1.0973470147269924,
+      "grad_norm": 1.0390625,
+      "learning_rate": 2.5059101654846333e-05,
+      "loss": 2.0054,
+      "step": 1290
+    },
+    {
+      "epoch": 1.1016002977298103,
+      "grad_norm": 1.09375,
+      "learning_rate": 2.4940898345153666e-05,
+      "loss": 2.0273,
+      "step": 1295
+    },
+    {
+      "epoch": 1.105853580732628,
+      "grad_norm": 0.97265625,
+      "learning_rate": 2.4822695035460995e-05,
+      "loss": 1.9994,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1101068637354459,
+      "grad_norm": 0.92578125,
+      "learning_rate": 2.470449172576832e-05,
+      "loss": 2.0084,
+      "step": 1305
+    },
+    {
+      "epoch": 1.1143601467382636,
+      "grad_norm": 1.015625,
+      "learning_rate": 2.458628841607565e-05,
+      "loss": 2.0087,
+      "step": 1310
+    },
+    {
+      "epoch": 1.1186134297410815,
+      "grad_norm": 0.9296875,
+      "learning_rate": 2.446808510638298e-05,
+      "loss": 1.9766,
+      "step": 1315
+    },
+    {
+      "epoch": 1.1228667127438992,
+      "grad_norm": 1.0,
+      "learning_rate": 2.4349881796690306e-05,
+      "loss": 2.0408,
+      "step": 1320
+    },
+    {
+      "epoch": 1.127119995746717,
+      "grad_norm": 1.03125,
+      "learning_rate": 2.4231678486997636e-05,
+      "loss": 1.9739,
+      "step": 1325
+    },
+    {
+      "epoch": 1.1313732787495347,
+      "grad_norm": 1.0234375,
+      "learning_rate": 2.4113475177304965e-05,
+      "loss": 2.005,
+      "step": 1330
+    },
+    {
+      "epoch": 1.1356265617523527,
+      "grad_norm": 1.0859375,
+      "learning_rate": 2.3995271867612295e-05,
+      "loss": 2.0161,
+      "step": 1335
+    },
+    {
+      "epoch": 1.1398798447551703,
+      "grad_norm": 1.125,
+      "learning_rate": 2.3877068557919624e-05,
+      "loss": 1.995,
+      "step": 1340
+    },
+    {
+      "epoch": 1.1441331277579883,
+      "grad_norm": 0.92578125,
+      "learning_rate": 2.3758865248226954e-05,
+      "loss": 2.0458,
+      "step": 1345
+    },
+    {
+      "epoch": 1.148386410760806,
+      "grad_norm": 1.0859375,
+      "learning_rate": 2.364066193853428e-05,
+      "loss": 1.9776,
+      "step": 1350
+    },
+    {
+      "epoch": 1.1526396937636239,
+      "grad_norm": 0.99609375,
+      "learning_rate": 2.352245862884161e-05,
+      "loss": 1.9791,
+      "step": 1355
+    },
+    {
+      "epoch": 1.1568929767664415,
+      "grad_norm": 0.96484375,
+      "learning_rate": 2.340425531914894e-05,
+      "loss": 1.9759,
+      "step": 1360
+    },
+    {
+      "epoch": 1.1611462597692594,
+      "grad_norm": 1.0234375,
+      "learning_rate": 2.3286052009456265e-05,
+      "loss": 1.9886,
+      "step": 1365
+    },
+    {
+      "epoch": 1.1653995427720771,
+      "grad_norm": 1.1953125,
+      "learning_rate": 2.3167848699763594e-05,
+      "loss": 2.0073,
+      "step": 1370
+    },
+    {
+      "epoch": 1.169652825774895,
+      "grad_norm": 1.0234375,
+      "learning_rate": 2.3049645390070924e-05,
+      "loss": 1.9885,
+      "step": 1375
+    },
+    {
+      "epoch": 1.1739061087777127,
+      "grad_norm": 1.2734375,
+      "learning_rate": 2.293144208037825e-05,
+      "loss": 1.9871,
+      "step": 1380
+    },
+    {
+      "epoch": 1.1781593917805306,
+      "grad_norm": 1.03125,
+      "learning_rate": 2.281323877068558e-05,
+      "loss": 1.9838,
+      "step": 1385
+    },
+    {
+      "epoch": 1.1824126747833483,
+      "grad_norm": 0.9609375,
+      "learning_rate": 2.269503546099291e-05,
+      "loss": 2.0428,
+      "step": 1390
+    },
+    {
+      "epoch": 1.1866659577861662,
+      "grad_norm": 1.21875,
+      "learning_rate": 2.2576832151300238e-05,
+      "loss": 2.0274,
+      "step": 1395
+    },
+    {
+      "epoch": 1.190919240788984,
+      "grad_norm": 1.0703125,
+      "learning_rate": 2.2458628841607564e-05,
+      "loss": 2.0031,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1951725237918018,
+      "grad_norm": 1.046875,
+      "learning_rate": 2.2340425531914894e-05,
+      "loss": 2.0222,
+      "step": 1405
+    },
+    {
+      "epoch": 1.1994258067946195,
+      "grad_norm": 0.90625,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 2.0086,
+      "step": 1410
+    },
+    {
+      "epoch": 1.2036790897974374,
+      "grad_norm": 1.0390625,
+      "learning_rate": 2.2104018912529553e-05,
+      "loss": 1.9803,
+      "step": 1415
+    },
+    {
+      "epoch": 1.2079323728002551,
+      "grad_norm": 0.9140625,
+      "learning_rate": 2.1985815602836882e-05,
+      "loss": 1.986,
+      "step": 1420
+    },
+    {
+      "epoch": 1.212185655803073,
+      "grad_norm": 1.015625,
+      "learning_rate": 2.186761229314421e-05,
+      "loss": 2.0073,
+      "step": 1425
+    },
+    {
+      "epoch": 1.2164389388058907,
+      "grad_norm": 0.93359375,
+      "learning_rate": 2.1749408983451538e-05,
+      "loss": 1.9992,
+      "step": 1430
+    },
+    {
+      "epoch": 1.2206922218087086,
+      "grad_norm": 1.1328125,
+      "learning_rate": 2.1631205673758867e-05,
+      "loss": 2.004,
+      "step": 1435
+    },
+    {
+      "epoch": 1.2249455048115263,
+      "grad_norm": 0.91015625,
+      "learning_rate": 2.1513002364066197e-05,
+      "loss": 1.9966,
+      "step": 1440
+    },
+    {
+      "epoch": 1.2291987878143442,
+      "grad_norm": 0.98046875,
+      "learning_rate": 2.1394799054373523e-05,
+      "loss": 2.0202,
+      "step": 1445
+    },
+    {
+      "epoch": 1.233452070817162,
+      "grad_norm": 0.98046875,
+      "learning_rate": 2.1276595744680852e-05,
+      "loss": 2.0352,
+      "step": 1450
+    },
+    {
+      "epoch": 1.2377053538199798,
+      "grad_norm": 0.9609375,
+      "learning_rate": 2.115839243498818e-05,
+      "loss": 1.9897,
+      "step": 1455
+    },
+    {
+      "epoch": 1.2419586368227975,
+      "grad_norm": 1.0859375,
+      "learning_rate": 2.1040189125295508e-05,
+      "loss": 1.9795,
+      "step": 1460
+    },
+    {
+      "epoch": 1.2462119198256154,
+      "grad_norm": 0.91796875,
+      "learning_rate": 2.0921985815602837e-05,
+      "loss": 1.9816,
+      "step": 1465
+    },
+    {
+      "epoch": 1.2504652028284333,
+      "grad_norm": 0.98828125,
+      "learning_rate": 2.0803782505910167e-05,
+      "loss": 2.011,
+      "step": 1470
+    },
+    {
+      "epoch": 1.254718485831251,
+      "grad_norm": 0.9375,
+      "learning_rate": 2.0685579196217493e-05,
+      "loss": 1.9725,
+      "step": 1475
+    },
+    {
+      "epoch": 1.2589717688340687,
+      "grad_norm": 0.92578125,
+      "learning_rate": 2.0567375886524822e-05,
+      "loss": 2.0059,
+      "step": 1480
+    },
+    {
+      "epoch": 1.2632250518368866,
+      "grad_norm": 0.921875,
+      "learning_rate": 2.0449172576832152e-05,
+      "loss": 1.9543,
+      "step": 1485
+    },
+    {
+      "epoch": 1.2674783348397045,
+      "grad_norm": 0.96875,
+      "learning_rate": 2.033096926713948e-05,
+      "loss": 2.0051,
+      "step": 1490
+    },
+    {
+      "epoch": 1.2717316178425222,
+      "grad_norm": 1.1328125,
+      "learning_rate": 2.0212765957446807e-05,
+      "loss": 1.9752,
+      "step": 1495
+    },
+    {
+      "epoch": 1.2759849008453399,
+      "grad_norm": 0.953125,
+      "learning_rate": 2.009456264775414e-05,
+      "loss": 1.9795,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2802381838481578,
+      "grad_norm": 0.91796875,
+      "learning_rate": 1.9976359338061466e-05,
+      "loss": 2.0009,
+      "step": 1505
+    },
+    {
+      "epoch": 1.2844914668509757,
+      "grad_norm": 0.984375,
+      "learning_rate": 1.9858156028368796e-05,
+      "loss": 1.994,
+      "step": 1510
+    },
+    {
+      "epoch": 1.2887447498537934,
+      "grad_norm": 0.9921875,
+      "learning_rate": 1.9739952718676125e-05,
+      "loss": 1.9667,
+      "step": 1515
+    },
+    {
+      "epoch": 1.292998032856611,
+      "grad_norm": 0.9375,
+      "learning_rate": 1.9621749408983455e-05,
+      "loss": 1.974,
+      "step": 1520
+    },
+    {
+      "epoch": 1.297251315859429,
+      "grad_norm": 0.91015625,
+      "learning_rate": 1.950354609929078e-05,
+      "loss": 2.0098,
+      "step": 1525
+    },
+    {
+      "epoch": 1.301504598862247,
+      "grad_norm": 0.98828125,
+      "learning_rate": 1.938534278959811e-05,
+      "loss": 1.9415,
+      "step": 1530
+    },
+    {
+      "epoch": 1.3057578818650646,
+      "grad_norm": 1.015625,
+      "learning_rate": 1.926713947990544e-05,
+      "loss": 1.9867,
+      "step": 1535
+    },
+    {
+      "epoch": 1.3100111648678823,
+      "grad_norm": 1.1953125,
+      "learning_rate": 1.9148936170212766e-05,
+      "loss": 2.0018,
+      "step": 1540
+    },
+    {
+      "epoch": 1.3142644478707002,
+      "grad_norm": 0.91796875,
+      "learning_rate": 1.9030732860520095e-05,
+      "loss": 1.9627,
+      "step": 1545
+    },
+    {
+      "epoch": 1.318517730873518,
+      "grad_norm": 1.03125,
+      "learning_rate": 1.8912529550827425e-05,
+      "loss": 2.0021,
+      "step": 1550
+    },
+    {
+      "epoch": 1.3227710138763358,
+      "grad_norm": 0.91015625,
+      "learning_rate": 1.879432624113475e-05,
+      "loss": 1.9893,
+      "step": 1555
+    },
+    {
+      "epoch": 1.3270242968791535,
+      "grad_norm": 0.9140625,
+      "learning_rate": 1.867612293144208e-05,
+      "loss": 1.9839,
+      "step": 1560
+    },
+    {
+      "epoch": 1.3312775798819714,
+      "grad_norm": 0.984375,
+      "learning_rate": 1.855791962174941e-05,
+      "loss": 1.9724,
+      "step": 1565
+    },
+    {
+      "epoch": 1.3355308628847893,
+      "grad_norm": 0.9296875,
+      "learning_rate": 1.8439716312056736e-05,
+      "loss": 1.9775,
+      "step": 1570
+    },
+    {
+      "epoch": 1.339784145887607,
+      "grad_norm": 1.0078125,
+      "learning_rate": 1.8321513002364065e-05,
+      "loss": 1.991,
+      "step": 1575
+    },
+    {
+      "epoch": 1.3440374288904249,
+      "grad_norm": 0.9765625,
+      "learning_rate": 1.8203309692671398e-05,
+      "loss": 1.9752,
+      "step": 1580
+    },
+    {
+      "epoch": 1.3482907118932426,
+      "grad_norm": 1.0390625,
+      "learning_rate": 1.8085106382978724e-05,
+      "loss": 2.0044,
+      "step": 1585
+    },
+    {
+      "epoch": 1.3525439948960605,
+      "grad_norm": 0.97265625,
+      "learning_rate": 1.7966903073286054e-05,
+      "loss": 1.9499,
+      "step": 1590
+    },
+    {
+      "epoch": 1.3567972778988782,
+      "grad_norm": 0.94921875,
+      "learning_rate": 1.7848699763593383e-05,
+      "loss": 2.0067,
+      "step": 1595
+    },
+    {
+      "epoch": 1.361050560901696,
+      "grad_norm": 1.015625,
+      "learning_rate": 1.773049645390071e-05,
+      "loss": 1.9673,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3653038439045138,
+      "grad_norm": 0.984375,
+      "learning_rate": 1.761229314420804e-05,
+      "loss": 2.021,
+      "step": 1605
+    },
+    {
+      "epoch": 1.3695571269073317,
+      "grad_norm": 0.94921875,
+      "learning_rate": 1.7494089834515368e-05,
+      "loss": 1.952,
+      "step": 1610
+    },
+    {
+      "epoch": 1.3738104099101494,
+      "grad_norm": 0.93359375,
+      "learning_rate": 1.7375886524822697e-05,
+      "loss": 2.0264,
+      "step": 1615
+    },
+    {
+      "epoch": 1.3780636929129673,
+      "grad_norm": 1.03125,
+      "learning_rate": 1.7257683215130024e-05,
+      "loss": 1.949,
+      "step": 1620
+    },
+    {
+      "epoch": 1.382316975915785,
+      "grad_norm": 0.9296875,
+      "learning_rate": 1.7139479905437353e-05,
+      "loss": 2.0119,
+      "step": 1625
+    },
+    {
+      "epoch": 1.3865702589186029,
+      "grad_norm": 0.9609375,
+      "learning_rate": 1.7021276595744682e-05,
+      "loss": 1.9519,
+      "step": 1630
+    },
+    {
+      "epoch": 1.3908235419214205,
+      "grad_norm": 0.91015625,
+      "learning_rate": 1.690307328605201e-05,
+      "loss": 1.9907,
+      "step": 1635
+    },
+    {
+      "epoch": 1.3950768249242385,
+      "grad_norm": 1.234375,
+      "learning_rate": 1.6784869976359338e-05,
+      "loss": 1.9697,
+      "step": 1640
+    },
+    {
+      "epoch": 1.3993301079270561,
+      "grad_norm": 0.90625,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 2.0108,
+      "step": 1645
+    },
+    {
+      "epoch": 1.403583390929874,
+      "grad_norm": 0.921875,
+      "learning_rate": 1.6548463356973994e-05,
+      "loss": 1.9791,
+      "step": 1650
+    },
+    {
+      "epoch": 1.4078366739326917,
+      "grad_norm": 0.953125,
+      "learning_rate": 1.6430260047281323e-05,
+      "loss": 1.943,
+      "step": 1655
+    },
+    {
+      "epoch": 1.4120899569355096,
+      "grad_norm": 1.03125,
+      "learning_rate": 1.6312056737588656e-05,
+      "loss": 1.9913,
+      "step": 1660
+    },
+    {
+      "epoch": 1.4163432399383273,
+      "grad_norm": 0.9453125,
+      "learning_rate": 1.6193853427895982e-05,
+      "loss": 1.9857,
+      "step": 1665
+    },
+    {
+      "epoch": 1.4205965229411452,
+      "grad_norm": 0.96484375,
+      "learning_rate": 1.607565011820331e-05,
+      "loss": 1.9915,
+      "step": 1670
+    },
+    {
+      "epoch": 1.424849805943963,
+      "grad_norm": 0.984375,
+      "learning_rate": 1.595744680851064e-05,
+      "loss": 1.981,
+      "step": 1675
+    },
+    {
+      "epoch": 1.4291030889467808,
+      "grad_norm": 0.9375,
+      "learning_rate": 1.5839243498817967e-05,
+      "loss": 1.9546,
+      "step": 1680
+    },
+    {
+      "epoch": 1.4333563719495985,
+      "grad_norm": 1.0625,
+      "learning_rate": 1.5721040189125296e-05,
+      "loss": 1.9395,
+      "step": 1685
+    },
+    {
+      "epoch": 1.4376096549524164,
+      "grad_norm": 0.91015625,
+      "learning_rate": 1.5602836879432626e-05,
+      "loss": 1.9626,
+      "step": 1690
+    },
+    {
+      "epoch": 1.4418629379552341,
+      "grad_norm": 1.0234375,
+      "learning_rate": 1.5484633569739952e-05,
+      "loss": 1.9989,
+      "step": 1695
+    },
+    {
+      "epoch": 1.446116220958052,
+      "grad_norm": 0.9375,
+      "learning_rate": 1.536643026004728e-05,
+      "loss": 1.9689,
+      "step": 1700
+    },
+    {
+      "epoch": 1.4503695039608697,
+      "grad_norm": 0.95703125,
+      "learning_rate": 1.5248226950354611e-05,
+      "loss": 1.9641,
+      "step": 1705
+    },
+    {
+      "epoch": 1.4546227869636876,
+      "grad_norm": 0.9921875,
+      "learning_rate": 1.5130023640661939e-05,
+      "loss": 1.951,
+      "step": 1710
+    },
+    {
+      "epoch": 1.4588760699665053,
+      "grad_norm": 1.078125,
+      "learning_rate": 1.5011820330969268e-05,
+      "loss": 1.9763,
+      "step": 1715
+    },
+    {
+      "epoch": 1.4631293529693232,
+      "grad_norm": 0.98828125,
+      "learning_rate": 1.4893617021276596e-05,
+      "loss": 1.9631,
+      "step": 1720
+    },
+    {
+      "epoch": 1.467382635972141,
+      "grad_norm": 0.9765625,
+      "learning_rate": 1.4775413711583924e-05,
+      "loss": 2.0039,
+      "step": 1725
+    },
+    {
+      "epoch": 1.4716359189749588,
+      "grad_norm": 0.96875,
+      "learning_rate": 1.4657210401891253e-05,
+      "loss": 1.9483,
+      "step": 1730
+    },
+    {
+      "epoch": 1.4758892019777767,
+      "grad_norm": 0.93359375,
+      "learning_rate": 1.4539007092198581e-05,
+      "loss": 1.9422,
+      "step": 1735
+    },
+    {
+      "epoch": 1.4801424849805944,
+      "grad_norm": 1.109375,
+      "learning_rate": 1.4420803782505912e-05,
+      "loss": 1.9517,
+      "step": 1740
+    },
+    {
+      "epoch": 1.484395767983412,
+      "grad_norm": 0.9140625,
+      "learning_rate": 1.4302600472813242e-05,
+      "loss": 1.9792,
+      "step": 1745
+    },
+    {
+      "epoch": 1.48864905098623,
+      "grad_norm": 0.93359375,
+      "learning_rate": 1.418439716312057e-05,
+      "loss": 2.0064,
+      "step": 1750
+    },
+    {
+      "epoch": 1.492902333989048,
+      "grad_norm": 0.93359375,
+      "learning_rate": 1.4066193853427897e-05,
+      "loss": 1.9686,
+      "step": 1755
+    },
+    {
+      "epoch": 1.4971556169918656,
+      "grad_norm": 0.89453125,
+      "learning_rate": 1.3947990543735227e-05,
+      "loss": 1.9391,
+      "step": 1760
+    },
+    {
+      "epoch": 1.5014088999946833,
+      "grad_norm": 0.9140625,
+      "learning_rate": 1.3829787234042554e-05,
+      "loss": 1.9603,
+      "step": 1765
+    },
+    {
+      "epoch": 1.5056621829975012,
+      "grad_norm": 0.92578125,
+      "learning_rate": 1.3711583924349882e-05,
+      "loss": 1.9378,
+      "step": 1770
+    },
+    {
+      "epoch": 1.5099154660003191,
+      "grad_norm": 0.97265625,
+      "learning_rate": 1.3593380614657212e-05,
+      "loss": 1.9399,
+      "step": 1775
+    },
+    {
+      "epoch": 1.5141687490031368,
+      "grad_norm": 0.93359375,
+      "learning_rate": 1.347517730496454e-05,
+      "loss": 1.9586,
+      "step": 1780
+    },
+    {
+      "epoch": 1.5184220320059545,
+      "grad_norm": 0.921875,
+      "learning_rate": 1.3356973995271869e-05,
+      "loss": 1.9742,
+      "step": 1785
+    },
+    {
+      "epoch": 1.5226753150087724,
+      "grad_norm": 0.9375,
+      "learning_rate": 1.3238770685579197e-05,
+      "loss": 1.9867,
+      "step": 1790
+    },
+    {
+      "epoch": 1.5269285980115903,
+      "grad_norm": 1.0859375,
+      "learning_rate": 1.3120567375886524e-05,
+      "loss": 1.9653,
+      "step": 1795
+    },
+    {
+      "epoch": 1.531181881014408,
+      "grad_norm": 0.97265625,
+      "learning_rate": 1.3002364066193854e-05,
+      "loss": 1.9553,
+      "step": 1800
+    },
+    {
+      "epoch": 1.5354351640172257,
+      "grad_norm": 1.0234375,
+      "learning_rate": 1.2884160756501182e-05,
+      "loss": 1.9441,
+      "step": 1805
+    },
+    {
+      "epoch": 1.5396884470200436,
+      "grad_norm": 0.9453125,
+      "learning_rate": 1.2765957446808511e-05,
+      "loss": 1.9559,
+      "step": 1810
+    },
+    {
+      "epoch": 1.5439417300228615,
+      "grad_norm": 0.94140625,
+      "learning_rate": 1.2647754137115839e-05,
+      "loss": 1.9375,
+      "step": 1815
+    },
+    {
+      "epoch": 1.5481950130256792,
+      "grad_norm": 0.95703125,
+      "learning_rate": 1.2529550827423167e-05,
+      "loss": 1.9919,
+      "step": 1820
+    },
+    {
+      "epoch": 1.5524482960284969,
+      "grad_norm": 0.9296875,
+      "learning_rate": 1.2411347517730498e-05,
+      "loss": 1.9703,
+      "step": 1825
+    },
+    {
+      "epoch": 1.5567015790313148,
+      "grad_norm": 0.91796875,
+      "learning_rate": 1.2293144208037825e-05,
+      "loss": 1.9232,
+      "step": 1830
+    },
+    {
+      "epoch": 1.5609548620341327,
+      "grad_norm": 0.94921875,
+      "learning_rate": 1.2174940898345153e-05,
+      "loss": 1.9505,
+      "step": 1835
+    },
+    {
+      "epoch": 1.5652081450369504,
+      "grad_norm": 0.91796875,
+      "learning_rate": 1.2056737588652483e-05,
+      "loss": 1.9262,
+      "step": 1840
+    },
+    {
+      "epoch": 1.569461428039768,
+      "grad_norm": 0.98828125,
+      "learning_rate": 1.1938534278959812e-05,
+      "loss": 1.9686,
+      "step": 1845
+    },
+    {
+      "epoch": 1.573714711042586,
+      "grad_norm": 0.94140625,
+      "learning_rate": 1.182033096926714e-05,
+      "loss": 1.9462,
+      "step": 1850
+    },
+    {
+      "epoch": 1.5779679940454039,
+      "grad_norm": 0.98046875,
+      "learning_rate": 1.170212765957447e-05,
+      "loss": 1.973,
+      "step": 1855
+    },
+    {
+      "epoch": 1.5822212770482216,
+      "grad_norm": 0.89453125,
+      "learning_rate": 1.1583924349881797e-05,
+      "loss": 1.9319,
+      "step": 1860
+    },
+    {
+      "epoch": 1.5864745600510393,
+      "grad_norm": 0.9296875,
+      "learning_rate": 1.1465721040189125e-05,
+      "loss": 1.9633,
+      "step": 1865
+    },
+    {
+      "epoch": 1.5907278430538572,
+      "grad_norm": 0.9453125,
+      "learning_rate": 1.1347517730496454e-05,
+      "loss": 1.9293,
+      "step": 1870
+    },
+    {
+      "epoch": 1.594981126056675,
+      "grad_norm": 0.93359375,
+      "learning_rate": 1.1229314420803782e-05,
+      "loss": 1.9791,
+      "step": 1875
+    },
+    {
+      "epoch": 1.5992344090594928,
+      "grad_norm": 0.94140625,
+      "learning_rate": 1.1111111111111112e-05,
+      "loss": 1.9372,
+      "step": 1880
+    },
+    {
+      "epoch": 1.6034876920623105,
+      "grad_norm": 0.98046875,
+      "learning_rate": 1.0992907801418441e-05,
+      "loss": 1.9623,
+      "step": 1885
+    },
+    {
+      "epoch": 1.6077409750651284,
+      "grad_norm": 0.90234375,
+      "learning_rate": 1.0874704491725769e-05,
+      "loss": 1.9176,
+      "step": 1890
+    },
+    {
+      "epoch": 1.6119942580679463,
+      "grad_norm": 0.94140625,
+      "learning_rate": 1.0756501182033098e-05,
+      "loss": 1.9251,
+      "step": 1895
+    },
+    {
+      "epoch": 1.616247541070764,
+      "grad_norm": 0.890625,
+      "learning_rate": 1.0638297872340426e-05,
+      "loss": 1.9609,
+      "step": 1900
+    },
+    {
+      "epoch": 1.6205008240735816,
+      "grad_norm": 0.94140625,
+      "learning_rate": 1.0520094562647754e-05,
+      "loss": 1.9602,
+      "step": 1905
+    },
+    {
+      "epoch": 1.6247541070763996,
+      "grad_norm": 0.9140625,
+      "learning_rate": 1.0401891252955083e-05,
+      "loss": 1.9371,
+      "step": 1910
+    },
+    {
+      "epoch": 1.6290073900792175,
+      "grad_norm": 0.94921875,
+      "learning_rate": 1.0283687943262411e-05,
+      "loss": 1.9796,
+      "step": 1915
+    },
+    {
+      "epoch": 1.6332606730820352,
+      "grad_norm": 0.93359375,
+      "learning_rate": 1.016548463356974e-05,
+      "loss": 1.943,
+      "step": 1920
+    },
+    {
+      "epoch": 1.6375139560848528,
+      "grad_norm": 1.125,
+      "learning_rate": 1.004728132387707e-05,
+      "loss": 1.933,
+      "step": 1925
+    },
+    {
+      "epoch": 1.6417672390876707,
+      "grad_norm": 0.984375,
+      "learning_rate": 9.929078014184398e-06,
+      "loss": 1.9473,
+      "step": 1930
+    },
+    {
+      "epoch": 1.6460205220904887,
+      "grad_norm": 0.91796875,
+      "learning_rate": 9.810874704491727e-06,
+      "loss": 1.9453,
+      "step": 1935
+    },
+    {
+      "epoch": 1.6502738050933063,
+      "grad_norm": 0.921875,
+      "learning_rate": 9.692671394799055e-06,
+      "loss": 1.9359,
+      "step": 1940
+    },
+    {
+      "epoch": 1.654527088096124,
+      "grad_norm": 0.98828125,
+      "learning_rate": 9.574468085106383e-06,
+      "loss": 1.9362,
+      "step": 1945
+    },
+    {
+      "epoch": 1.658780371098942,
+      "grad_norm": 1.03125,
+      "learning_rate": 9.456264775413712e-06,
+      "loss": 1.9583,
+      "step": 1950
+    },
+    {
+      "epoch": 1.6630336541017599,
+      "grad_norm": 0.96484375,
+      "learning_rate": 9.33806146572104e-06,
+      "loss": 1.9398,
+      "step": 1955
+    },
+    {
+      "epoch": 1.6672869371045778,
+      "grad_norm": 0.98046875,
+      "learning_rate": 9.219858156028368e-06,
+      "loss": 1.9631,
+      "step": 1960
+    },
+    {
+      "epoch": 1.6715402201073954,
+      "grad_norm": 1.015625,
+      "learning_rate": 9.101654846335699e-06,
+      "loss": 1.9779,
+      "step": 1965
+    },
+    {
+      "epoch": 1.6757935031102131,
+      "grad_norm": 0.96875,
+      "learning_rate": 8.983451536643027e-06,
+      "loss": 1.9397,
+      "step": 1970
+    },
+    {
+      "epoch": 1.680046786113031,
+      "grad_norm": 0.98046875,
+      "learning_rate": 8.865248226950355e-06,
+      "loss": 1.9447,
+      "step": 1975
+    },
+    {
+      "epoch": 1.684300069115849,
+      "grad_norm": 0.9296875,
+      "learning_rate": 8.747044917257684e-06,
+      "loss": 1.9395,
+      "step": 1980
+    },
+    {
+      "epoch": 1.6885533521186666,
+      "grad_norm": 0.91015625,
+      "learning_rate": 8.628841607565012e-06,
+      "loss": 1.9629,
+      "step": 1985
+    },
+    {
+      "epoch": 1.6928066351214843,
+      "grad_norm": 1.0625,
+      "learning_rate": 8.510638297872341e-06,
+      "loss": 1.9608,
+      "step": 1990
+    },
+    {
+      "epoch": 1.6970599181243022,
+      "grad_norm": 0.94140625,
+      "learning_rate": 8.392434988179669e-06,
+      "loss": 1.9572,
+      "step": 1995
+    },
+    {
+      "epoch": 1.7013132011271201,
+      "grad_norm": 0.953125,
+      "learning_rate": 8.274231678486997e-06,
+      "loss": 1.9234,
+      "step": 2000
+    },
+    {
+      "epoch": 1.7055664841299378,
+      "grad_norm": 0.9609375,
+      "learning_rate": 8.156028368794328e-06,
+      "loss": 1.9562,
+      "step": 2005
+    },
+    {
+      "epoch": 1.7098197671327555,
+      "grad_norm": 0.8984375,
+      "learning_rate": 8.037825059101656e-06,
+      "loss": 1.9415,
+      "step": 2010
+    },
+    {
+      "epoch": 1.7140730501355734,
+      "grad_norm": 0.90234375,
+      "learning_rate": 7.919621749408983e-06,
+      "loss": 1.9663,
+      "step": 2015
+    },
+    {
+      "epoch": 1.7183263331383913,
+      "grad_norm": 0.9296875,
+      "learning_rate": 7.801418439716313e-06,
+      "loss": 1.9647,
+      "step": 2020
+    },
+    {
+      "epoch": 1.722579616141209,
+      "grad_norm": 0.92578125,
+      "learning_rate": 7.68321513002364e-06,
+      "loss": 1.9552,
+      "step": 2025
+    },
+    {
+      "epoch": 1.7268328991440267,
+      "grad_norm": 0.9140625,
+      "learning_rate": 7.565011820330969e-06,
+      "loss": 1.9452,
+      "step": 2030
+    },
+    {
+      "epoch": 1.7310861821468446,
+      "grad_norm": 0.93359375,
+      "learning_rate": 7.446808510638298e-06,
+      "loss": 1.9519,
+      "step": 2035
+    },
+    {
+      "epoch": 1.7353394651496625,
+      "grad_norm": 0.87890625,
+      "learning_rate": 7.3286052009456266e-06,
+      "loss": 1.9613,
+      "step": 2040
+    },
+    {
+      "epoch": 1.7395927481524802,
+      "grad_norm": 0.90625,
+      "learning_rate": 7.210401891252956e-06,
+      "loss": 1.9412,
+      "step": 2045
+    },
+    {
+      "epoch": 1.743846031155298,
+      "grad_norm": 0.88671875,
+      "learning_rate": 7.092198581560285e-06,
+      "loss": 1.9559,
+      "step": 2050
+    },
+    {
+      "epoch": 1.7480993141581158,
+      "grad_norm": 0.953125,
+      "learning_rate": 6.973995271867613e-06,
+      "loss": 1.9794,
+      "step": 2055
+    },
+    {
+      "epoch": 1.7523525971609337,
+      "grad_norm": 1.03125,
+      "learning_rate": 6.855791962174941e-06,
+      "loss": 1.9788,
+      "step": 2060
+    },
+    {
+      "epoch": 1.7566058801637514,
+      "grad_norm": 0.94140625,
+      "learning_rate": 6.73758865248227e-06,
+      "loss": 1.9228,
+      "step": 2065
+    },
+    {
+      "epoch": 1.760859163166569,
+      "grad_norm": 0.90625,
+      "learning_rate": 6.619385342789598e-06,
+      "loss": 1.9189,
+      "step": 2070
+    },
+    {
+      "epoch": 1.765112446169387,
+      "grad_norm": 0.921875,
+      "learning_rate": 6.501182033096927e-06,
+      "loss": 1.9285,
+      "step": 2075
+    },
+    {
+      "epoch": 1.769365729172205,
+      "grad_norm": 0.91796875,
+      "learning_rate": 6.3829787234042555e-06,
+      "loss": 1.9816,
+      "step": 2080
+    },
+    {
+      "epoch": 1.7736190121750226,
+      "grad_norm": 0.9296875,
+      "learning_rate": 6.264775413711583e-06,
+      "loss": 1.9315,
+      "step": 2085
+    },
+    {
+      "epoch": 1.7778722951778403,
+      "grad_norm": 0.90234375,
+      "learning_rate": 6.146572104018913e-06,
+      "loss": 1.9809,
+      "step": 2090
+    },
+    {
+      "epoch": 1.7821255781806582,
+      "grad_norm": 0.91796875,
+      "learning_rate": 6.028368794326241e-06,
+      "loss": 1.9732,
+      "step": 2095
+    },
+    {
+      "epoch": 1.786378861183476,
+      "grad_norm": 0.91796875,
+      "learning_rate": 5.91016548463357e-06,
+      "loss": 1.9602,
+      "step": 2100
+    },
+    {
+      "epoch": 1.7906321441862938,
+      "grad_norm": 0.9609375,
+      "learning_rate": 5.791962174940899e-06,
+      "loss": 1.941,
+      "step": 2105
+    },
+    {
+      "epoch": 1.7948854271891115,
+      "grad_norm": 1.015625,
+      "learning_rate": 5.673758865248227e-06,
+      "loss": 1.9833,
+      "step": 2110
+    },
+    {
+      "epoch": 1.7991387101919294,
+      "grad_norm": 0.9375,
+      "learning_rate": 5.555555555555556e-06,
+      "loss": 1.9538,
+      "step": 2115
+    },
+    {
+      "epoch": 1.8033919931947473,
+      "grad_norm": 0.94140625,
+      "learning_rate": 5.4373522458628844e-06,
+      "loss": 1.944,
+      "step": 2120
+    },
+    {
+      "epoch": 1.807645276197565,
+      "grad_norm": 0.9140625,
+      "learning_rate": 5.319148936170213e-06,
+      "loss": 1.9467,
+      "step": 2125
+    },
+    {
+      "epoch": 1.8118985592003827,
+      "grad_norm": 0.9140625,
+      "learning_rate": 5.200945626477542e-06,
+      "loss": 1.9437,
+      "step": 2130
+    },
+    {
+      "epoch": 1.8161518422032006,
+      "grad_norm": 0.921875,
+      "learning_rate": 5.08274231678487e-06,
+      "loss": 1.9043,
+      "step": 2135
+    },
+    {
+      "epoch": 1.8204051252060185,
+      "grad_norm": 0.95703125,
+      "learning_rate": 4.964539007092199e-06,
+      "loss": 1.9349,
+      "step": 2140
+    },
+    {
+      "epoch": 1.8246584082088362,
+      "grad_norm": 0.9296875,
+      "learning_rate": 4.8463356973995275e-06,
+      "loss": 1.9543,
+      "step": 2145
+    },
+    {
+      "epoch": 1.8289116912116539,
+      "grad_norm": 0.97265625,
+      "learning_rate": 4.728132387706856e-06,
+      "loss": 1.9475,
+      "step": 2150
+    },
+    {
+      "epoch": 1.8331649742144718,
+      "grad_norm": 0.953125,
+      "learning_rate": 4.609929078014184e-06,
+      "loss": 1.8929,
+      "step": 2155
+    },
+    {
+      "epoch": 1.8374182572172897,
+      "grad_norm": 0.91796875,
+      "learning_rate": 4.491725768321513e-06,
+      "loss": 1.95,
+      "step": 2160
+    },
+    {
+      "epoch": 1.8416715402201074,
+      "grad_norm": 0.9375,
+      "learning_rate": 4.373522458628842e-06,
+      "loss": 1.9575,
+      "step": 2165
+    },
+    {
+      "epoch": 1.845924823222925,
+      "grad_norm": 0.90234375,
+      "learning_rate": 4.255319148936171e-06,
+      "loss": 1.8995,
+      "step": 2170
+    },
+    {
+      "epoch": 1.850178106225743,
+      "grad_norm": 0.91015625,
+      "learning_rate": 4.137115839243498e-06,
+      "loss": 1.9353,
+      "step": 2175
+    },
+    {
+      "epoch": 1.8544313892285609,
+      "grad_norm": 0.9140625,
+      "learning_rate": 4.018912529550828e-06,
+      "loss": 1.95,
+      "step": 2180
+    },
+    {
+      "epoch": 1.8586846722313786,
+      "grad_norm": 0.9140625,
+      "learning_rate": 3.9007092198581565e-06,
+      "loss": 1.9362,
+      "step": 2185
+    },
+    {
+      "epoch": 1.8629379552341963,
+      "grad_norm": 0.92578125,
+      "learning_rate": 3.7825059101654847e-06,
+      "loss": 1.9492,
+      "step": 2190
+    },
+    {
+      "epoch": 1.8671912382370142,
+      "grad_norm": 0.91796875,
+      "learning_rate": 3.6643026004728133e-06,
+      "loss": 1.928,
+      "step": 2195
+    },
+    {
+      "epoch": 1.871444521239832,
+      "grad_norm": 0.94140625,
+      "learning_rate": 3.5460992907801423e-06,
+      "loss": 1.9407,
+      "step": 2200
+    },
+    {
+      "epoch": 1.8756978042426498,
+      "grad_norm": 0.96875,
+      "learning_rate": 3.4278959810874705e-06,
+      "loss": 1.9631,
+      "step": 2205
+    },
+    {
+      "epoch": 1.8799510872454674,
+      "grad_norm": 0.9140625,
+      "learning_rate": 3.309692671394799e-06,
+      "loss": 1.9274,
+      "step": 2210
+    },
+    {
+      "epoch": 1.8842043702482854,
+      "grad_norm": 0.91796875,
+      "learning_rate": 3.1914893617021277e-06,
+      "loss": 1.9329,
+      "step": 2215
+    },
+    {
+      "epoch": 1.8884576532511033,
+      "grad_norm": 0.91796875,
+      "learning_rate": 3.0732860520094564e-06,
+      "loss": 1.918,
+      "step": 2220
+    },
+    {
+      "epoch": 1.892710936253921,
+      "grad_norm": 0.96484375,
+      "learning_rate": 2.955082742316785e-06,
+      "loss": 1.9898,
+      "step": 2225
+    },
+    {
+      "epoch": 1.8969642192567386,
+      "grad_norm": 0.9453125,
+      "learning_rate": 2.8368794326241136e-06,
+      "loss": 1.9357,
+      "step": 2230
+    },
+    {
+      "epoch": 1.9012175022595565,
+      "grad_norm": 0.91015625,
+      "learning_rate": 2.7186761229314422e-06,
+      "loss": 1.9508,
+      "step": 2235
+    },
+    {
+      "epoch": 1.9054707852623745,
+      "grad_norm": 0.91015625,
+      "learning_rate": 2.600472813238771e-06,
+      "loss": 1.9221,
+      "step": 2240
+    },
+    {
+      "epoch": 1.9097240682651921,
+      "grad_norm": 0.91796875,
+      "learning_rate": 2.4822695035460995e-06,
+      "loss": 1.9175,
+      "step": 2245
+    },
+    {
+      "epoch": 1.9139773512680098,
+      "grad_norm": 0.94921875,
+      "learning_rate": 2.364066193853428e-06,
+      "loss": 1.9834,
+      "step": 2250
+    },
+    {
+      "epoch": 1.9182306342708277,
+      "grad_norm": 0.91796875,
+      "learning_rate": 2.2458628841607567e-06,
+      "loss": 1.9414,
+      "step": 2255
+    },
+    {
+      "epoch": 1.9224839172736456,
+      "grad_norm": 0.9453125,
+      "learning_rate": 2.1276595744680853e-06,
+      "loss": 1.9541,
+      "step": 2260
+    },
+    {
+      "epoch": 1.9267372002764636,
+      "grad_norm": 0.89453125,
+      "learning_rate": 2.009456264775414e-06,
+      "loss": 1.9599,
+      "step": 2265
+    },
+    {
+      "epoch": 1.9309904832792812,
+      "grad_norm": 0.9375,
+      "learning_rate": 1.8912529550827423e-06,
+      "loss": 1.9898,
+      "step": 2270
+    },
+    {
+      "epoch": 1.935243766282099,
+      "grad_norm": 0.9375,
+      "learning_rate": 1.7730496453900712e-06,
+      "loss": 1.9505,
+      "step": 2275
+    },
+    {
+      "epoch": 1.9394970492849168,
+      "grad_norm": 0.8984375,
+      "learning_rate": 1.6548463356973996e-06,
+      "loss": 1.9509,
+      "step": 2280
+    },
+    {
+      "epoch": 1.9437503322877348,
+      "grad_norm": 0.91796875,
+      "learning_rate": 1.5366430260047282e-06,
+      "loss": 1.9495,
+      "step": 2285
+    },
+    {
+      "epoch": 1.9480036152905524,
+      "grad_norm": 1.0390625,
+      "learning_rate": 1.4184397163120568e-06,
+      "loss": 1.9543,
+      "step": 2290
+    },
+    {
+      "epoch": 1.9522568982933701,
+      "grad_norm": 0.9375,
+      "learning_rate": 1.3002364066193854e-06,
+      "loss": 1.9254,
+      "step": 2295
+    },
+    {
+      "epoch": 1.956510181296188,
+      "grad_norm": 0.9375,
+      "learning_rate": 1.182033096926714e-06,
+      "loss": 1.8833,
+      "step": 2300
+    },
+    {
+      "epoch": 1.960763464299006,
+      "grad_norm": 0.90625,
+      "learning_rate": 1.0638297872340427e-06,
+      "loss": 1.9878,
+      "step": 2305
+    },
+    {
+      "epoch": 1.9650167473018236,
+      "grad_norm": 0.9375,
+      "learning_rate": 9.456264775413712e-07,
+      "loss": 1.9668,
+      "step": 2310
+    },
+    {
+      "epoch": 1.9692700303046413,
+      "grad_norm": 0.98828125,
+      "learning_rate": 8.274231678486998e-07,
+      "loss": 1.9359,
+      "step": 2315
+    },
+    {
+      "epoch": 1.9735233133074592,
+      "grad_norm": 0.9140625,
+      "learning_rate": 7.092198581560284e-07,
+      "loss": 1.947,
+      "step": 2320
+    },
+    {
+      "epoch": 1.9777765963102771,
+      "grad_norm": 0.921875,
+      "learning_rate": 5.91016548463357e-07,
+      "loss": 1.9126,
+      "step": 2325
+    },
+    {
+      "epoch": 1.9820298793130948,
+      "grad_norm": 0.90625,
+      "learning_rate": 4.728132387706856e-07,
+      "loss": 1.9494,
+      "step": 2330
+    },
+    {
+      "epoch": 1.9862831623159125,
+      "grad_norm": 0.91796875,
+      "learning_rate": 3.546099290780142e-07,
+      "loss": 1.9494,
+      "step": 2335
+    },
+    {
+      "epoch": 1.9905364453187304,
+      "grad_norm": 0.9765625,
+      "learning_rate": 2.364066193853428e-07,
+      "loss": 1.9362,
+      "step": 2340
+    },
+    {
+      "epoch": 1.9947897283215483,
+      "grad_norm": 0.92578125,
+      "learning_rate": 1.182033096926714e-07,
+      "loss": 1.9238,
+      "step": 2345
+    },
+    {
+      "epoch": 1.999043011324366,
+      "grad_norm": 0.96875,
+      "learning_rate": 0.0,
+      "loss": 1.9951,
+      "step": 2350
+    },
+    {
+      "epoch": 1.999043011324366,
+      "step": 2350,
+      "total_flos": 3.076816830283645e+19,
+      "train_loss": 2.254816828382776,
+      "train_runtime": 45442.7326,
+      "train_samples_per_second": 6.622,
+      "train_steps_per_second": 0.052
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 2350,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.076816830283645e+19,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:346e8a22ccdfc9195114d9a4fea809f18278ac3522e5eb6f96847b228de8b964
+size 5432