Training in progress, epoch 0

.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

added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-7B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.1",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c9292503240bdd62b3c4a03ec139d6bbac704d1e4448cbe59b2f760bbbcaf48
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b731a8761366ac2bc7160c58c78cf388d0f61172e2732315f89fce80d9773d49
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b1b9a563c6c809c4c55c8cb28599b1f2fb60f3a5b9c1a66e512cde987093058
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdd676708a6e39834c5c3ef9f3903f72181c89deb95d7c079980789e0a3139da
+size 1089994880

model.safetensors.index.json

@@ -0,0 +1,346 @@
+{
+  "metadata": {
+    "total_size": 15231233024
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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-00003-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00003-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-00003-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "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.bias": "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.bias": "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-00003-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00003-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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.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.self_attn.k_proj.bias": "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.bias": "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.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-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.self_attn.k_proj.bias": "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.bias": "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.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-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.self_attn.k_proj.bias": "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.bias": "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.bias": "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.self_attn.k_proj.bias": "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.bias": "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.bias": "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-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-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-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-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.bias": "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.bias": "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.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-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.self_attn.k_proj.bias": "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.bias": "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.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00003-of-00004.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_log.jsonl

@@ -0,0 +1,537 @@
+{"current_steps": 1, "total_steps": 1560, "loss": 0.4618, "lr": 6.41025641025641e-08, "epoch": 0.00192, "percentage": 0.06, "elapsed_time": "0:00:19", "remaining_time": "8:21:20"}
+{"current_steps": 2, "total_steps": 1560, "loss": 0.483, "lr": 1.282051282051282e-07, "epoch": 0.00384, "percentage": 0.13, "elapsed_time": "0:00:28", "remaining_time": "6:05:34"}
+{"current_steps": 3, "total_steps": 1560, "loss": 0.4064, "lr": 1.9230769230769234e-07, "epoch": 0.00576, "percentage": 0.19, "elapsed_time": "0:00:36", "remaining_time": "5:18:38"}
+{"current_steps": 4, "total_steps": 1560, "loss": 0.4284, "lr": 2.564102564102564e-07, "epoch": 0.00768, "percentage": 0.26, "elapsed_time": "0:00:45", "remaining_time": "4:55:17"}
+{"current_steps": 5, "total_steps": 1560, "loss": 0.4042, "lr": 3.205128205128205e-07, "epoch": 0.0096, "percentage": 0.32, "elapsed_time": "0:00:54", "remaining_time": "4:41:10"}
+{"current_steps": 6, "total_steps": 1560, "loss": 0.4145, "lr": 3.846153846153847e-07, "epoch": 0.01152, "percentage": 0.38, "elapsed_time": "0:01:03", "remaining_time": "4:32:14"}
+{"current_steps": 7, "total_steps": 1560, "loss": 0.4246, "lr": 4.4871794871794876e-07, "epoch": 0.01344, "percentage": 0.45, "elapsed_time": "0:01:11", "remaining_time": "4:25:57"}
+{"current_steps": 8, "total_steps": 1560, "loss": 0.3834, "lr": 5.128205128205128e-07, "epoch": 0.01536, "percentage": 0.51, "elapsed_time": "0:01:20", "remaining_time": "4:21:14"}
+{"current_steps": 9, "total_steps": 1560, "loss": 0.3825, "lr": 5.76923076923077e-07, "epoch": 0.01728, "percentage": 0.58, "elapsed_time": "0:01:29", "remaining_time": "4:17:09"}
+{"current_steps": 10, "total_steps": 1560, "loss": 0.3867, "lr": 6.41025641025641e-07, "epoch": 0.0192, "percentage": 0.64, "elapsed_time": "0:01:38", "remaining_time": "4:14:28"}
+{"current_steps": 11, "total_steps": 1560, "loss": 0.3923, "lr": 7.051282051282052e-07, "epoch": 0.02112, "percentage": 0.71, "elapsed_time": "0:01:47", "remaining_time": "4:12:34"}
+{"current_steps": 12, "total_steps": 1560, "loss": 0.4069, "lr": 7.692307692307694e-07, "epoch": 0.02304, "percentage": 0.77, "elapsed_time": "0:01:56", "remaining_time": "4:10:37"}
+{"current_steps": 13, "total_steps": 1560, "loss": 0.4416, "lr": 8.333333333333333e-07, "epoch": 0.02496, "percentage": 0.83, "elapsed_time": "0:02:05", "remaining_time": "4:09:06"}
+{"current_steps": 14, "total_steps": 1560, "loss": 0.4148, "lr": 8.974358974358975e-07, "epoch": 0.02688, "percentage": 0.9, "elapsed_time": "0:02:14", "remaining_time": "4:07:40"}
+{"current_steps": 15, "total_steps": 1560, "loss": 0.3897, "lr": 9.615384615384617e-07, "epoch": 0.0288, "percentage": 0.96, "elapsed_time": "0:02:23", "remaining_time": "4:06:27"}
+{"current_steps": 16, "total_steps": 1560, "loss": 0.4194, "lr": 1.0256410256410257e-06, "epoch": 0.03072, "percentage": 1.03, "elapsed_time": "0:02:32", "remaining_time": "4:05:12"}
+{"current_steps": 17, "total_steps": 1560, "loss": 0.3597, "lr": 1.0897435897435899e-06, "epoch": 0.03264, "percentage": 1.09, "elapsed_time": "0:02:41", "remaining_time": "4:04:12"}
+{"current_steps": 18, "total_steps": 1560, "loss": 0.349, "lr": 1.153846153846154e-06, "epoch": 0.03456, "percentage": 1.15, "elapsed_time": "0:02:50", "remaining_time": "4:03:09"}
+{"current_steps": 19, "total_steps": 1560, "loss": 0.3187, "lr": 1.217948717948718e-06, "epoch": 0.03648, "percentage": 1.22, "elapsed_time": "0:02:59", "remaining_time": "4:02:08"}
+{"current_steps": 20, "total_steps": 1560, "loss": 0.3572, "lr": 1.282051282051282e-06, "epoch": 0.0384, "percentage": 1.28, "elapsed_time": "0:03:08", "remaining_time": "4:01:17"}
+{"current_steps": 21, "total_steps": 1560, "loss": 0.2936, "lr": 1.3461538461538462e-06, "epoch": 0.04032, "percentage": 1.35, "elapsed_time": "0:03:16", "remaining_time": "4:00:26"}
+{"current_steps": 22, "total_steps": 1560, "loss": 0.2985, "lr": 1.4102564102564104e-06, "epoch": 0.04224, "percentage": 1.41, "elapsed_time": "0:03:25", "remaining_time": "3:59:41"}
+{"current_steps": 23, "total_steps": 1560, "loss": 0.3498, "lr": 1.4743589743589745e-06, "epoch": 0.04416, "percentage": 1.47, "elapsed_time": "0:03:34", "remaining_time": "3:59:04"}
+{"current_steps": 24, "total_steps": 1560, "loss": 0.2896, "lr": 1.5384615384615387e-06, "epoch": 0.04608, "percentage": 1.54, "elapsed_time": "0:03:43", "remaining_time": "3:58:27"}
+{"current_steps": 25, "total_steps": 1560, "loss": 0.3037, "lr": 1.602564102564103e-06, "epoch": 0.048, "percentage": 1.6, "elapsed_time": "0:03:52", "remaining_time": "3:57:53"}
+{"current_steps": 26, "total_steps": 1560, "loss": 0.2743, "lr": 1.6666666666666667e-06, "epoch": 0.04992, "percentage": 1.67, "elapsed_time": "0:04:01", "remaining_time": "3:57:23"}
+{"current_steps": 27, "total_steps": 1560, "loss": 0.2585, "lr": 1.7307692307692308e-06, "epoch": 0.05184, "percentage": 1.73, "elapsed_time": "0:04:10", "remaining_time": "3:56:47"}
+{"current_steps": 28, "total_steps": 1560, "loss": 0.2949, "lr": 1.794871794871795e-06, "epoch": 0.05376, "percentage": 1.79, "elapsed_time": "0:04:19", "remaining_time": "3:56:11"}
+{"current_steps": 29, "total_steps": 1560, "loss": 0.2928, "lr": 1.8589743589743592e-06, "epoch": 0.05568, "percentage": 1.86, "elapsed_time": "0:04:27", "remaining_time": "3:55:38"}
+{"current_steps": 30, "total_steps": 1560, "loss": 0.2819, "lr": 1.9230769230769234e-06, "epoch": 0.0576, "percentage": 1.92, "elapsed_time": "0:04:36", "remaining_time": "3:55:06"}
+{"current_steps": 31, "total_steps": 1560, "loss": 0.2601, "lr": 1.987179487179487e-06, "epoch": 0.05952, "percentage": 1.99, "elapsed_time": "0:04:45", "remaining_time": "3:54:35"}
+{"current_steps": 32, "total_steps": 1560, "loss": 0.2686, "lr": 2.0512820512820513e-06, "epoch": 0.06144, "percentage": 2.05, "elapsed_time": "0:04:54", "remaining_time": "3:54:06"}
+{"current_steps": 33, "total_steps": 1560, "loss": 0.2743, "lr": 2.1153846153846155e-06, "epoch": 0.06336, "percentage": 2.12, "elapsed_time": "0:05:03", "remaining_time": "3:53:43"}
+{"current_steps": 34, "total_steps": 1560, "loss": 0.278, "lr": 2.1794871794871797e-06, "epoch": 0.06528, "percentage": 2.18, "elapsed_time": "0:05:11", "remaining_time": "3:53:22"}
+{"current_steps": 35, "total_steps": 1560, "loss": 0.2777, "lr": 2.243589743589744e-06, "epoch": 0.0672, "percentage": 2.24, "elapsed_time": "0:05:20", "remaining_time": "3:53:02"}
+{"current_steps": 36, "total_steps": 1560, "loss": 0.2868, "lr": 2.307692307692308e-06, "epoch": 0.06912, "percentage": 2.31, "elapsed_time": "0:05:29", "remaining_time": "3:52:36"}
+{"current_steps": 37, "total_steps": 1560, "loss": 0.2571, "lr": 2.371794871794872e-06, "epoch": 0.07104, "percentage": 2.37, "elapsed_time": "0:05:38", "remaining_time": "3:52:16"}
+{"current_steps": 38, "total_steps": 1560, "loss": 0.2757, "lr": 2.435897435897436e-06, "epoch": 0.07296, "percentage": 2.44, "elapsed_time": "0:05:47", "remaining_time": "3:52:09"}
+{"current_steps": 39, "total_steps": 1560, "loss": 0.2654, "lr": 2.5e-06, "epoch": 0.07488, "percentage": 2.5, "elapsed_time": "0:05:57", "remaining_time": "3:52:03"}
+{"current_steps": 40, "total_steps": 1560, "loss": 0.2486, "lr": 2.564102564102564e-06, "epoch": 0.0768, "percentage": 2.56, "elapsed_time": "0:06:06", "remaining_time": "3:51:52"}
+{"current_steps": 41, "total_steps": 1560, "loss": 0.2655, "lr": 2.6282051282051286e-06, "epoch": 0.07872, "percentage": 2.63, "elapsed_time": "0:06:15", "remaining_time": "3:51:44"}
+{"current_steps": 42, "total_steps": 1560, "loss": 0.2766, "lr": 2.6923076923076923e-06, "epoch": 0.08064, "percentage": 2.69, "elapsed_time": "0:06:24", "remaining_time": "3:51:20"}
+{"current_steps": 43, "total_steps": 1560, "loss": 0.2311, "lr": 2.756410256410257e-06, "epoch": 0.08256, "percentage": 2.76, "elapsed_time": "0:06:32", "remaining_time": "3:50:57"}
+{"current_steps": 44, "total_steps": 1560, "loss": 0.2313, "lr": 2.8205128205128207e-06, "epoch": 0.08448, "percentage": 2.82, "elapsed_time": "0:06:41", "remaining_time": "3:50:34"}
+{"current_steps": 45, "total_steps": 1560, "loss": 0.2652, "lr": 2.8846153846153845e-06, "epoch": 0.0864, "percentage": 2.88, "elapsed_time": "0:06:50", "remaining_time": "3:50:12"}
+{"current_steps": 46, "total_steps": 1560, "loss": 0.2983, "lr": 2.948717948717949e-06, "epoch": 0.08832, "percentage": 2.95, "elapsed_time": "0:06:59", "remaining_time": "3:49:51"}
+{"current_steps": 47, "total_steps": 1560, "loss": 0.24, "lr": 3.012820512820513e-06, "epoch": 0.09024, "percentage": 3.01, "elapsed_time": "0:07:07", "remaining_time": "3:49:30"}
+{"current_steps": 48, "total_steps": 1560, "loss": 0.2302, "lr": 3.0769230769230774e-06, "epoch": 0.09216, "percentage": 3.08, "elapsed_time": "0:07:16", "remaining_time": "3:49:12"}
+{"current_steps": 49, "total_steps": 1560, "loss": 0.2325, "lr": 3.141025641025641e-06, "epoch": 0.09408, "percentage": 3.14, "elapsed_time": "0:07:25", "remaining_time": "3:48:52"}
+{"current_steps": 50, "total_steps": 1560, "loss": 0.2701, "lr": 3.205128205128206e-06, "epoch": 0.096, "percentage": 3.21, "elapsed_time": "0:07:34", "remaining_time": "3:48:34"}
+{"current_steps": 51, "total_steps": 1560, "loss": 0.2185, "lr": 3.2692307692307696e-06, "epoch": 0.09792, "percentage": 3.27, "elapsed_time": "0:07:42", "remaining_time": "3:48:15"}
+{"current_steps": 52, "total_steps": 1560, "loss": 0.2866, "lr": 3.3333333333333333e-06, "epoch": 0.09984, "percentage": 3.33, "elapsed_time": "0:07:51", "remaining_time": "3:47:58"}
+{"current_steps": 53, "total_steps": 1560, "loss": 0.2723, "lr": 3.397435897435898e-06, "epoch": 0.10176, "percentage": 3.4, "elapsed_time": "0:08:00", "remaining_time": "3:47:40"}
+{"current_steps": 54, "total_steps": 1560, "loss": 0.2189, "lr": 3.4615384615384617e-06, "epoch": 0.10368, "percentage": 3.46, "elapsed_time": "0:08:09", "remaining_time": "3:47:21"}
+{"current_steps": 55, "total_steps": 1560, "loss": 0.2734, "lr": 3.5256410256410263e-06, "epoch": 0.1056, "percentage": 3.53, "elapsed_time": "0:08:17", "remaining_time": "3:47:05"}
+{"current_steps": 56, "total_steps": 1560, "loss": 0.2855, "lr": 3.58974358974359e-06, "epoch": 0.10752, "percentage": 3.59, "elapsed_time": "0:08:26", "remaining_time": "3:46:50"}
+{"current_steps": 57, "total_steps": 1560, "loss": 0.2582, "lr": 3.653846153846154e-06, "epoch": 0.10944, "percentage": 3.65, "elapsed_time": "0:08:35", "remaining_time": "3:46:34"}
+{"current_steps": 58, "total_steps": 1560, "loss": 0.2714, "lr": 3.7179487179487184e-06, "epoch": 0.11136, "percentage": 3.72, "elapsed_time": "0:08:44", "remaining_time": "3:46:20"}
+{"current_steps": 59, "total_steps": 1560, "loss": 0.2351, "lr": 3.782051282051282e-06, "epoch": 0.11328, "percentage": 3.78, "elapsed_time": "0:08:53", "remaining_time": "3:46:03"}
+{"current_steps": 60, "total_steps": 1560, "loss": 0.2699, "lr": 3.846153846153847e-06, "epoch": 0.1152, "percentage": 3.85, "elapsed_time": "0:09:01", "remaining_time": "3:45:47"}
+{"current_steps": 61, "total_steps": 1560, "loss": 0.2103, "lr": 3.910256410256411e-06, "epoch": 0.11712, "percentage": 3.91, "elapsed_time": "0:09:10", "remaining_time": "3:45:31"}
+{"current_steps": 62, "total_steps": 1560, "loss": 0.2462, "lr": 3.974358974358974e-06, "epoch": 0.11904, "percentage": 3.97, "elapsed_time": "0:09:19", "remaining_time": "3:45:15"}
+{"current_steps": 63, "total_steps": 1560, "loss": 0.2367, "lr": 4.0384615384615385e-06, "epoch": 0.12096, "percentage": 4.04, "elapsed_time": "0:09:28", "remaining_time": "3:45:00"}
+{"current_steps": 64, "total_steps": 1560, "loss": 0.2428, "lr": 4.102564102564103e-06, "epoch": 0.12288, "percentage": 4.1, "elapsed_time": "0:09:36", "remaining_time": "3:44:46"}
+{"current_steps": 65, "total_steps": 1560, "loss": 0.2759, "lr": 4.166666666666667e-06, "epoch": 0.1248, "percentage": 4.17, "elapsed_time": "0:09:45", "remaining_time": "3:44:29"}
+{"current_steps": 66, "total_steps": 1560, "loss": 0.2896, "lr": 4.230769230769231e-06, "epoch": 0.12672, "percentage": 4.23, "elapsed_time": "0:09:54", "remaining_time": "3:44:14"}
+{"current_steps": 67, "total_steps": 1560, "loss": 0.2328, "lr": 4.294871794871795e-06, "epoch": 0.12864, "percentage": 4.29, "elapsed_time": "0:10:03", "remaining_time": "3:44:00"}
+{"current_steps": 68, "total_steps": 1560, "loss": 0.2442, "lr": 4.358974358974359e-06, "epoch": 0.13056, "percentage": 4.36, "elapsed_time": "0:10:11", "remaining_time": "3:43:46"}
+{"current_steps": 69, "total_steps": 1560, "loss": 0.2461, "lr": 4.423076923076924e-06, "epoch": 0.13248, "percentage": 4.42, "elapsed_time": "0:10:20", "remaining_time": "3:43:31"}
+{"current_steps": 70, "total_steps": 1560, "loss": 0.2856, "lr": 4.487179487179488e-06, "epoch": 0.1344, "percentage": 4.49, "elapsed_time": "0:10:29", "remaining_time": "3:43:17"}
+{"current_steps": 71, "total_steps": 1560, "loss": 0.2682, "lr": 4.551282051282052e-06, "epoch": 0.13632, "percentage": 4.55, "elapsed_time": "0:10:38", "remaining_time": "3:43:03"}
+{"current_steps": 72, "total_steps": 1560, "loss": 0.283, "lr": 4.615384615384616e-06, "epoch": 0.13824, "percentage": 4.62, "elapsed_time": "0:10:46", "remaining_time": "3:42:48"}
+{"current_steps": 73, "total_steps": 1560, "loss": 0.2003, "lr": 4.6794871794871795e-06, "epoch": 0.14016, "percentage": 4.68, "elapsed_time": "0:10:55", "remaining_time": "3:42:37"}
+{"current_steps": 74, "total_steps": 1560, "loss": 0.2997, "lr": 4.743589743589744e-06, "epoch": 0.14208, "percentage": 4.74, "elapsed_time": "0:11:04", "remaining_time": "3:42:28"}
+{"current_steps": 75, "total_steps": 1560, "loss": 0.2764, "lr": 4.807692307692308e-06, "epoch": 0.144, "percentage": 4.81, "elapsed_time": "0:11:13", "remaining_time": "3:42:17"}
+{"current_steps": 76, "total_steps": 1560, "loss": 0.2019, "lr": 4.871794871794872e-06, "epoch": 0.14592, "percentage": 4.87, "elapsed_time": "0:11:22", "remaining_time": "3:42:07"}
+{"current_steps": 77, "total_steps": 1560, "loss": 0.2833, "lr": 4.935897435897436e-06, "epoch": 0.14784, "percentage": 4.94, "elapsed_time": "0:11:31", "remaining_time": "3:41:54"}
+{"current_steps": 78, "total_steps": 1560, "loss": 0.2661, "lr": 5e-06, "epoch": 0.14976, "percentage": 5.0, "elapsed_time": "0:11:40", "remaining_time": "3:41:41"}
+{"current_steps": 79, "total_steps": 1560, "loss": 0.2571, "lr": 5.064102564102565e-06, "epoch": 0.15168, "percentage": 5.06, "elapsed_time": "0:11:48", "remaining_time": "3:41:30"}
+{"current_steps": 80, "total_steps": 1560, "loss": 0.265, "lr": 5.128205128205128e-06, "epoch": 0.1536, "percentage": 5.13, "elapsed_time": "0:11:57", "remaining_time": "3:41:17"}
+{"current_steps": 81, "total_steps": 1560, "loss": 0.2673, "lr": 5.192307692307693e-06, "epoch": 0.15552, "percentage": 5.19, "elapsed_time": "0:12:06", "remaining_time": "3:41:05"}
+{"current_steps": 82, "total_steps": 1560, "loss": 0.2732, "lr": 5.256410256410257e-06, "epoch": 0.15744, "percentage": 5.26, "elapsed_time": "0:12:15", "remaining_time": "3:40:54"}
+{"current_steps": 83, "total_steps": 1560, "loss": 0.2384, "lr": 5.320512820512821e-06, "epoch": 0.15936, "percentage": 5.32, "elapsed_time": "0:12:24", "remaining_time": "3:40:43"}
+{"current_steps": 84, "total_steps": 1560, "loss": 0.343, "lr": 5.384615384615385e-06, "epoch": 0.16128, "percentage": 5.38, "elapsed_time": "0:12:33", "remaining_time": "3:40:31"}
+{"current_steps": 85, "total_steps": 1560, "loss": 0.2167, "lr": 5.448717948717949e-06, "epoch": 0.1632, "percentage": 5.45, "elapsed_time": "0:12:41", "remaining_time": "3:40:20"}
+{"current_steps": 86, "total_steps": 1560, "loss": 0.2275, "lr": 5.512820512820514e-06, "epoch": 0.16512, "percentage": 5.51, "elapsed_time": "0:12:50", "remaining_time": "3:40:08"}
+{"current_steps": 87, "total_steps": 1560, "loss": 0.2692, "lr": 5.576923076923077e-06, "epoch": 0.16704, "percentage": 5.58, "elapsed_time": "0:12:59", "remaining_time": "3:39:56"}
+{"current_steps": 88, "total_steps": 1560, "loss": 0.2393, "lr": 5.641025641025641e-06, "epoch": 0.16896, "percentage": 5.64, "elapsed_time": "0:13:08", "remaining_time": "3:39:45"}
+{"current_steps": 89, "total_steps": 1560, "loss": 0.2178, "lr": 5.705128205128206e-06, "epoch": 0.17088, "percentage": 5.71, "elapsed_time": "0:13:17", "remaining_time": "3:39:34"}
+{"current_steps": 90, "total_steps": 1560, "loss": 0.2299, "lr": 5.769230769230769e-06, "epoch": 0.1728, "percentage": 5.77, "elapsed_time": "0:13:25", "remaining_time": "3:39:22"}
+{"current_steps": 91, "total_steps": 1560, "loss": 0.2436, "lr": 5.833333333333334e-06, "epoch": 0.17472, "percentage": 5.83, "elapsed_time": "0:13:34", "remaining_time": "3:39:09"}
+{"current_steps": 92, "total_steps": 1560, "loss": 0.271, "lr": 5.897435897435898e-06, "epoch": 0.17664, "percentage": 5.9, "elapsed_time": "0:13:43", "remaining_time": "3:38:57"}
+{"current_steps": 93, "total_steps": 1560, "loss": 0.2324, "lr": 5.961538461538462e-06, "epoch": 0.17856, "percentage": 5.96, "elapsed_time": "0:13:52", "remaining_time": "3:38:45"}
+{"current_steps": 94, "total_steps": 1560, "loss": 0.2531, "lr": 6.025641025641026e-06, "epoch": 0.18048, "percentage": 6.03, "elapsed_time": "0:14:00", "remaining_time": "3:38:33"}
+{"current_steps": 95, "total_steps": 1560, "loss": 0.2309, "lr": 6.08974358974359e-06, "epoch": 0.1824, "percentage": 6.09, "elapsed_time": "0:14:09", "remaining_time": "3:38:21"}
+{"current_steps": 96, "total_steps": 1560, "loss": 0.2381, "lr": 6.153846153846155e-06, "epoch": 0.18432, "percentage": 6.15, "elapsed_time": "0:14:18", "remaining_time": "3:38:09"}
+{"current_steps": 97, "total_steps": 1560, "loss": 0.2709, "lr": 6.217948717948718e-06, "epoch": 0.18624, "percentage": 6.22, "elapsed_time": "0:14:27", "remaining_time": "3:37:59"}
+{"current_steps": 98, "total_steps": 1560, "loss": 0.2953, "lr": 6.282051282051282e-06, "epoch": 0.18816, "percentage": 6.28, "elapsed_time": "0:14:35", "remaining_time": "3:37:48"}
+{"current_steps": 99, "total_steps": 1560, "loss": 0.2753, "lr": 6.3461538461538466e-06, "epoch": 0.19008, "percentage": 6.35, "elapsed_time": "0:14:44", "remaining_time": "3:37:37"}
+{"current_steps": 100, "total_steps": 1560, "loss": 0.2432, "lr": 6.410256410256412e-06, "epoch": 0.192, "percentage": 6.41, "elapsed_time": "0:14:53", "remaining_time": "3:37:24"}
+{"current_steps": 101, "total_steps": 1560, "loss": 0.204, "lr": 6.474358974358975e-06, "epoch": 0.19392, "percentage": 6.47, "elapsed_time": "0:15:02", "remaining_time": "3:37:13"}
+{"current_steps": 102, "total_steps": 1560, "loss": 0.2434, "lr": 6.538461538461539e-06, "epoch": 0.19584, "percentage": 6.54, "elapsed_time": "0:15:10", "remaining_time": "3:37:01"}
+{"current_steps": 103, "total_steps": 1560, "loss": 0.2282, "lr": 6.602564102564103e-06, "epoch": 0.19776, "percentage": 6.6, "elapsed_time": "0:15:19", "remaining_time": "3:36:49"}
+{"current_steps": 104, "total_steps": 1560, "loss": 0.2794, "lr": 6.666666666666667e-06, "epoch": 0.19968, "percentage": 6.67, "elapsed_time": "0:15:28", "remaining_time": "3:36:36"}
+{"current_steps": 105, "total_steps": 1560, "loss": 0.2408, "lr": 6.730769230769232e-06, "epoch": 0.2016, "percentage": 6.73, "elapsed_time": "0:15:37", "remaining_time": "3:36:25"}
+{"current_steps": 106, "total_steps": 1560, "loss": 0.263, "lr": 6.794871794871796e-06, "epoch": 0.20352, "percentage": 6.79, "elapsed_time": "0:15:45", "remaining_time": "3:36:15"}
+{"current_steps": 107, "total_steps": 1560, "loss": 0.2788, "lr": 6.858974358974359e-06, "epoch": 0.20544, "percentage": 6.86, "elapsed_time": "0:15:54", "remaining_time": "3:36:03"}
+{"current_steps": 108, "total_steps": 1560, "loss": 0.2113, "lr": 6.923076923076923e-06, "epoch": 0.20736, "percentage": 6.92, "elapsed_time": "0:16:03", "remaining_time": "3:35:56"}
+{"current_steps": 109, "total_steps": 1560, "loss": 0.2056, "lr": 6.9871794871794876e-06, "epoch": 0.20928, "percentage": 6.99, "elapsed_time": "0:16:12", "remaining_time": "3:35:45"}
+{"current_steps": 110, "total_steps": 1560, "loss": 0.2483, "lr": 7.051282051282053e-06, "epoch": 0.2112, "percentage": 7.05, "elapsed_time": "0:16:21", "remaining_time": "3:35:32"}
+{"current_steps": 111, "total_steps": 1560, "loss": 0.219, "lr": 7.115384615384616e-06, "epoch": 0.21312, "percentage": 7.12, "elapsed_time": "0:16:29", "remaining_time": "3:35:21"}
+{"current_steps": 112, "total_steps": 1560, "loss": 0.2574, "lr": 7.17948717948718e-06, "epoch": 0.21504, "percentage": 7.18, "elapsed_time": "0:16:38", "remaining_time": "3:35:10"}
+{"current_steps": 113, "total_steps": 1560, "loss": 0.2466, "lr": 7.243589743589744e-06, "epoch": 0.21696, "percentage": 7.24, "elapsed_time": "0:16:47", "remaining_time": "3:34:58"}
+{"current_steps": 114, "total_steps": 1560, "loss": 0.2323, "lr": 7.307692307692308e-06, "epoch": 0.21888, "percentage": 7.31, "elapsed_time": "0:16:55", "remaining_time": "3:34:46"}
+{"current_steps": 115, "total_steps": 1560, "loss": 0.2307, "lr": 7.371794871794873e-06, "epoch": 0.2208, "percentage": 7.37, "elapsed_time": "0:17:04", "remaining_time": "3:34:35"}
+{"current_steps": 116, "total_steps": 1560, "loss": 0.2128, "lr": 7.435897435897437e-06, "epoch": 0.22272, "percentage": 7.44, "elapsed_time": "0:17:13", "remaining_time": "3:34:24"}
+{"current_steps": 117, "total_steps": 1560, "loss": 0.2699, "lr": 7.500000000000001e-06, "epoch": 0.22464, "percentage": 7.5, "elapsed_time": "0:17:22", "remaining_time": "3:34:12"}
+{"current_steps": 118, "total_steps": 1560, "loss": 0.2514, "lr": 7.564102564102564e-06, "epoch": 0.22656, "percentage": 7.56, "elapsed_time": "0:17:30", "remaining_time": "3:34:01"}
+{"current_steps": 119, "total_steps": 1560, "loss": 0.2757, "lr": 7.6282051282051286e-06, "epoch": 0.22848, "percentage": 7.63, "elapsed_time": "0:17:39", "remaining_time": "3:33:51"}
+{"current_steps": 120, "total_steps": 1560, "loss": 0.2476, "lr": 7.692307692307694e-06, "epoch": 0.2304, "percentage": 7.69, "elapsed_time": "0:17:48", "remaining_time": "3:33:41"}
+{"current_steps": 121, "total_steps": 1560, "loss": 0.3399, "lr": 7.756410256410258e-06, "epoch": 0.23232, "percentage": 7.76, "elapsed_time": "0:17:57", "remaining_time": "3:33:30"}
+{"current_steps": 122, "total_steps": 1560, "loss": 0.2776, "lr": 7.820512820512822e-06, "epoch": 0.23424, "percentage": 7.82, "elapsed_time": "0:18:06", "remaining_time": "3:33:21"}
+{"current_steps": 123, "total_steps": 1560, "loss": 0.2598, "lr": 7.884615384615384e-06, "epoch": 0.23616, "percentage": 7.88, "elapsed_time": "0:18:14", "remaining_time": "3:33:11"}
+{"current_steps": 124, "total_steps": 1560, "loss": 0.2496, "lr": 7.948717948717949e-06, "epoch": 0.23808, "percentage": 7.95, "elapsed_time": "0:18:23", "remaining_time": "3:33:00"}
+{"current_steps": 125, "total_steps": 1560, "loss": 0.2426, "lr": 8.012820512820515e-06, "epoch": 0.24, "percentage": 8.01, "elapsed_time": "0:18:32", "remaining_time": "3:32:50"}
+{"current_steps": 126, "total_steps": 1560, "loss": 0.2309, "lr": 8.076923076923077e-06, "epoch": 0.24192, "percentage": 8.08, "elapsed_time": "0:18:41", "remaining_time": "3:32:40"}
+{"current_steps": 127, "total_steps": 1560, "loss": 0.2449, "lr": 8.141025641025641e-06, "epoch": 0.24384, "percentage": 8.14, "elapsed_time": "0:18:50", "remaining_time": "3:32:30"}
+{"current_steps": 128, "total_steps": 1560, "loss": 0.3192, "lr": 8.205128205128205e-06, "epoch": 0.24576, "percentage": 8.21, "elapsed_time": "0:18:58", "remaining_time": "3:32:21"}
+{"current_steps": 129, "total_steps": 1560, "loss": 0.2463, "lr": 8.26923076923077e-06, "epoch": 0.24768, "percentage": 8.27, "elapsed_time": "0:19:07", "remaining_time": "3:32:11"}
+{"current_steps": 130, "total_steps": 1560, "loss": 0.3115, "lr": 8.333333333333334e-06, "epoch": 0.2496, "percentage": 8.33, "elapsed_time": "0:19:16", "remaining_time": "3:32:02"}
+{"current_steps": 131, "total_steps": 1560, "loss": 0.2362, "lr": 8.397435897435898e-06, "epoch": 0.25152, "percentage": 8.4, "elapsed_time": "0:19:25", "remaining_time": "3:31:52"}
+{"current_steps": 132, "total_steps": 1560, "loss": 0.2945, "lr": 8.461538461538462e-06, "epoch": 0.25344, "percentage": 8.46, "elapsed_time": "0:19:34", "remaining_time": "3:31:43"}
+{"current_steps": 133, "total_steps": 1560, "loss": 0.2249, "lr": 8.525641025641026e-06, "epoch": 0.25536, "percentage": 8.53, "elapsed_time": "0:19:43", "remaining_time": "3:31:36"}
+{"current_steps": 134, "total_steps": 1560, "loss": 0.2482, "lr": 8.58974358974359e-06, "epoch": 0.25728, "percentage": 8.59, "elapsed_time": "0:19:52", "remaining_time": "3:31:30"}
+{"current_steps": 135, "total_steps": 1560, "loss": 0.2297, "lr": 8.653846153846155e-06, "epoch": 0.2592, "percentage": 8.65, "elapsed_time": "0:20:01", "remaining_time": "3:31:19"}
+{"current_steps": 136, "total_steps": 1560, "loss": 0.3128, "lr": 8.717948717948719e-06, "epoch": 0.26112, "percentage": 8.72, "elapsed_time": "0:20:09", "remaining_time": "3:31:09"}
+{"current_steps": 137, "total_steps": 1560, "loss": 0.2022, "lr": 8.782051282051283e-06, "epoch": 0.26304, "percentage": 8.78, "elapsed_time": "0:20:18", "remaining_time": "3:30:58"}
+{"current_steps": 138, "total_steps": 1560, "loss": 0.2522, "lr": 8.846153846153847e-06, "epoch": 0.26496, "percentage": 8.85, "elapsed_time": "0:20:27", "remaining_time": "3:30:48"}
+{"current_steps": 139, "total_steps": 1560, "loss": 0.2475, "lr": 8.910256410256411e-06, "epoch": 0.26688, "percentage": 8.91, "elapsed_time": "0:20:36", "remaining_time": "3:30:37"}
+{"current_steps": 140, "total_steps": 1560, "loss": 0.2063, "lr": 8.974358974358976e-06, "epoch": 0.2688, "percentage": 8.97, "elapsed_time": "0:20:44", "remaining_time": "3:30:27"}
+{"current_steps": 141, "total_steps": 1560, "loss": 0.2541, "lr": 9.03846153846154e-06, "epoch": 0.27072, "percentage": 9.04, "elapsed_time": "0:20:53", "remaining_time": "3:30:17"}
+{"current_steps": 142, "total_steps": 1560, "loss": 0.2469, "lr": 9.102564102564104e-06, "epoch": 0.27264, "percentage": 9.1, "elapsed_time": "0:21:02", "remaining_time": "3:30:07"}
+{"current_steps": 143, "total_steps": 1560, "loss": 0.2339, "lr": 9.166666666666666e-06, "epoch": 0.27456, "percentage": 9.17, "elapsed_time": "0:21:11", "remaining_time": "3:29:56"}
+{"current_steps": 144, "total_steps": 1560, "loss": 0.2771, "lr": 9.230769230769232e-06, "epoch": 0.27648, "percentage": 9.23, "elapsed_time": "0:21:19", "remaining_time": "3:29:46"}
+{"current_steps": 145, "total_steps": 1560, "loss": 0.2478, "lr": 9.294871794871796e-06, "epoch": 0.2784, "percentage": 9.29, "elapsed_time": "0:21:28", "remaining_time": "3:29:36"}
+{"current_steps": 146, "total_steps": 1560, "loss": 0.2553, "lr": 9.358974358974359e-06, "epoch": 0.28032, "percentage": 9.36, "elapsed_time": "0:21:37", "remaining_time": "3:29:25"}
+{"current_steps": 147, "total_steps": 1560, "loss": 0.2381, "lr": 9.423076923076923e-06, "epoch": 0.28224, "percentage": 9.42, "elapsed_time": "0:21:46", "remaining_time": "3:29:15"}
+{"current_steps": 148, "total_steps": 1560, "loss": 0.2321, "lr": 9.487179487179487e-06, "epoch": 0.28416, "percentage": 9.49, "elapsed_time": "0:21:54", "remaining_time": "3:29:05"}
+{"current_steps": 149, "total_steps": 1560, "loss": 0.2486, "lr": 9.551282051282053e-06, "epoch": 0.28608, "percentage": 9.55, "elapsed_time": "0:22:03", "remaining_time": "3:28:54"}
+{"current_steps": 150, "total_steps": 1560, "loss": 0.2405, "lr": 9.615384615384616e-06, "epoch": 0.288, "percentage": 9.62, "elapsed_time": "0:22:12", "remaining_time": "3:28:45"}
+{"current_steps": 151, "total_steps": 1560, "loss": 0.2417, "lr": 9.67948717948718e-06, "epoch": 0.28992, "percentage": 9.68, "elapsed_time": "0:22:21", "remaining_time": "3:28:35"}
+{"current_steps": 152, "total_steps": 1560, "loss": 0.2483, "lr": 9.743589743589744e-06, "epoch": 0.29184, "percentage": 9.74, "elapsed_time": "0:22:29", "remaining_time": "3:28:25"}
+{"current_steps": 153, "total_steps": 1560, "loss": 0.2428, "lr": 9.807692307692308e-06, "epoch": 0.29376, "percentage": 9.81, "elapsed_time": "0:22:38", "remaining_time": "3:28:15"}
+{"current_steps": 154, "total_steps": 1560, "loss": 0.2695, "lr": 9.871794871794872e-06, "epoch": 0.29568, "percentage": 9.87, "elapsed_time": "0:22:47", "remaining_time": "3:28:05"}
+{"current_steps": 155, "total_steps": 1560, "loss": 0.224, "lr": 9.935897435897437e-06, "epoch": 0.2976, "percentage": 9.94, "elapsed_time": "0:22:56", "remaining_time": "3:27:55"}
+{"current_steps": 156, "total_steps": 1560, "loss": 0.2664, "lr": 1e-05, "epoch": 0.29952, "percentage": 10.0, "elapsed_time": "0:23:05", "remaining_time": "3:27:45"}
+{"current_steps": 157, "total_steps": 1560, "loss": 0.212, "lr": 9.999987482852865e-06, "epoch": 0.30144, "percentage": 10.06, "elapsed_time": "0:23:13", "remaining_time": "3:27:35"}
+{"current_steps": 158, "total_steps": 1560, "loss": 0.2247, "lr": 9.99994993147413e-06, "epoch": 0.30336, "percentage": 10.13, "elapsed_time": "0:23:22", "remaining_time": "3:27:25"}
+{"current_steps": 159, "total_steps": 1560, "loss": 0.23, "lr": 9.99988734605181e-06, "epoch": 0.30528, "percentage": 10.19, "elapsed_time": "0:23:31", "remaining_time": "3:27:15"}
+{"current_steps": 160, "total_steps": 1560, "loss": 0.251, "lr": 9.999799726899261e-06, "epoch": 0.3072, "percentage": 10.26, "elapsed_time": "0:23:40", "remaining_time": "3:27:06"}
+{"current_steps": 161, "total_steps": 1560, "loss": 0.2649, "lr": 9.999687074455182e-06, "epoch": 0.30912, "percentage": 10.32, "elapsed_time": "0:23:48", "remaining_time": "3:26:56"}
+{"current_steps": 162, "total_steps": 1560, "loss": 0.2551, "lr": 9.999549389283605e-06, "epoch": 0.31104, "percentage": 10.38, "elapsed_time": "0:23:57", "remaining_time": "3:26:46"}
+{"current_steps": 163, "total_steps": 1560, "loss": 0.2669, "lr": 9.999386672073902e-06, "epoch": 0.31296, "percentage": 10.45, "elapsed_time": "0:24:06", "remaining_time": "3:26:36"}
+{"current_steps": 164, "total_steps": 1560, "loss": 0.216, "lr": 9.999198923640774e-06, "epoch": 0.31488, "percentage": 10.51, "elapsed_time": "0:24:15", "remaining_time": "3:26:26"}
+{"current_steps": 165, "total_steps": 1560, "loss": 0.2466, "lr": 9.998986144924253e-06, "epoch": 0.3168, "percentage": 10.58, "elapsed_time": "0:24:23", "remaining_time": "3:26:16"}
+{"current_steps": 166, "total_steps": 1560, "loss": 0.2555, "lr": 9.998748336989687e-06, "epoch": 0.31872, "percentage": 10.64, "elapsed_time": "0:24:32", "remaining_time": "3:26:05"}
+{"current_steps": 167, "total_steps": 1560, "loss": 0.2482, "lr": 9.998485501027752e-06, "epoch": 0.32064, "percentage": 10.71, "elapsed_time": "0:24:41", "remaining_time": "3:25:55"}
+{"current_steps": 168, "total_steps": 1560, "loss": 0.2684, "lr": 9.998197638354428e-06, "epoch": 0.32256, "percentage": 10.77, "elapsed_time": "0:24:49", "remaining_time": "3:25:44"}
+{"current_steps": 169, "total_steps": 1560, "loss": 0.236, "lr": 9.997884750411004e-06, "epoch": 0.32448, "percentage": 10.83, "elapsed_time": "0:24:58", "remaining_time": "3:25:34"}
+{"current_steps": 170, "total_steps": 1560, "loss": 0.2183, "lr": 9.997546838764066e-06, "epoch": 0.3264, "percentage": 10.9, "elapsed_time": "0:25:07", "remaining_time": "3:25:25"}
+{"current_steps": 171, "total_steps": 1560, "loss": 0.2517, "lr": 9.997183905105487e-06, "epoch": 0.32832, "percentage": 10.96, "elapsed_time": "0:25:16", "remaining_time": "3:25:15"}
+{"current_steps": 172, "total_steps": 1560, "loss": 0.2128, "lr": 9.996795951252427e-06, "epoch": 0.33024, "percentage": 11.03, "elapsed_time": "0:25:24", "remaining_time": "3:25:05"}
+{"current_steps": 173, "total_steps": 1560, "loss": 0.245, "lr": 9.996382979147316e-06, "epoch": 0.33216, "percentage": 11.09, "elapsed_time": "0:25:33", "remaining_time": "3:24:55"}
+{"current_steps": 174, "total_steps": 1560, "loss": 0.2194, "lr": 9.995944990857848e-06, "epoch": 0.33408, "percentage": 11.15, "elapsed_time": "0:25:42", "remaining_time": "3:24:46"}
+{"current_steps": 175, "total_steps": 1560, "loss": 0.2249, "lr": 9.995481988576968e-06, "epoch": 0.336, "percentage": 11.22, "elapsed_time": "0:25:51", "remaining_time": "3:24:36"}
+{"current_steps": 176, "total_steps": 1560, "loss": 0.2524, "lr": 9.994993974622863e-06, "epoch": 0.33792, "percentage": 11.28, "elapsed_time": "0:26:00", "remaining_time": "3:24:27"}
+{"current_steps": 177, "total_steps": 1560, "loss": 0.2426, "lr": 9.994480951438948e-06, "epoch": 0.33984, "percentage": 11.35, "elapsed_time": "0:26:08", "remaining_time": "3:24:18"}
+{"current_steps": 178, "total_steps": 1560, "loss": 0.2705, "lr": 9.993942921593858e-06, "epoch": 0.34176, "percentage": 11.41, "elapsed_time": "0:26:17", "remaining_time": "3:24:09"}
+{"current_steps": 179, "total_steps": 1560, "loss": 0.2892, "lr": 9.993379887781436e-06, "epoch": 0.34368, "percentage": 11.47, "elapsed_time": "0:26:26", "remaining_time": "3:23:59"}
+{"current_steps": 180, "total_steps": 1560, "loss": 0.2796, "lr": 9.992791852820709e-06, "epoch": 0.3456, "percentage": 11.54, "elapsed_time": "0:26:35", "remaining_time": "3:23:50"}
+{"current_steps": 181, "total_steps": 1560, "loss": 0.2224, "lr": 9.992178819655889e-06, "epoch": 0.34752, "percentage": 11.6, "elapsed_time": "0:26:44", "remaining_time": "3:23:41"}
+{"current_steps": 182, "total_steps": 1560, "loss": 0.226, "lr": 9.991540791356342e-06, "epoch": 0.34944, "percentage": 11.67, "elapsed_time": "0:26:52", "remaining_time": "3:23:31"}
+{"current_steps": 183, "total_steps": 1560, "loss": 0.2259, "lr": 9.990877771116588e-06, "epoch": 0.35136, "percentage": 11.73, "elapsed_time": "0:27:01", "remaining_time": "3:23:22"}
+{"current_steps": 184, "total_steps": 1560, "loss": 0.2296, "lr": 9.990189762256275e-06, "epoch": 0.35328, "percentage": 11.79, "elapsed_time": "0:27:10", "remaining_time": "3:23:13"}
+{"current_steps": 185, "total_steps": 1560, "loss": 0.2123, "lr": 9.98947676822017e-06, "epoch": 0.3552, "percentage": 11.86, "elapsed_time": "0:27:19", "remaining_time": "3:23:03"}
+{"current_steps": 186, "total_steps": 1560, "loss": 0.2402, "lr": 9.988738792578126e-06, "epoch": 0.35712, "percentage": 11.92, "elapsed_time": "0:27:28", "remaining_time": "3:22:54"}
+{"current_steps": 187, "total_steps": 1560, "loss": 0.2418, "lr": 9.987975839025091e-06, "epoch": 0.35904, "percentage": 11.99, "elapsed_time": "0:27:36", "remaining_time": "3:22:44"}
+{"current_steps": 188, "total_steps": 1560, "loss": 0.2588, "lr": 9.987187911381059e-06, "epoch": 0.36096, "percentage": 12.05, "elapsed_time": "0:27:45", "remaining_time": "3:22:34"}
+{"current_steps": 189, "total_steps": 1560, "loss": 0.2396, "lr": 9.986375013591077e-06, "epoch": 0.36288, "percentage": 12.12, "elapsed_time": "0:27:54", "remaining_time": "3:22:25"}
+{"current_steps": 190, "total_steps": 1560, "loss": 0.2021, "lr": 9.985537149725207e-06, "epoch": 0.3648, "percentage": 12.18, "elapsed_time": "0:28:03", "remaining_time": "3:22:16"}
+{"current_steps": 191, "total_steps": 1560, "loss": 0.2169, "lr": 9.984674323978517e-06, "epoch": 0.36672, "percentage": 12.24, "elapsed_time": "0:28:11", "remaining_time": "3:22:06"}
+{"current_steps": 192, "total_steps": 1560, "loss": 0.2129, "lr": 9.983786540671052e-06, "epoch": 0.36864, "percentage": 12.31, "elapsed_time": "0:28:20", "remaining_time": "3:21:56"}
+{"current_steps": 193, "total_steps": 1560, "loss": 0.3003, "lr": 9.982873804247817e-06, "epoch": 0.37056, "percentage": 12.37, "elapsed_time": "0:28:29", "remaining_time": "3:21:47"}
+{"current_steps": 194, "total_steps": 1560, "loss": 0.2613, "lr": 9.981936119278758e-06, "epoch": 0.37248, "percentage": 12.44, "elapsed_time": "0:28:38", "remaining_time": "3:21:39"}
+{"current_steps": 195, "total_steps": 1560, "loss": 0.2609, "lr": 9.980973490458728e-06, "epoch": 0.3744, "percentage": 12.5, "elapsed_time": "0:28:47", "remaining_time": "3:21:30"}
+{"current_steps": 196, "total_steps": 1560, "loss": 0.2488, "lr": 9.979985922607476e-06, "epoch": 0.37632, "percentage": 12.56, "elapsed_time": "0:28:56", "remaining_time": "3:21:22"}
+{"current_steps": 197, "total_steps": 1560, "loss": 0.2117, "lr": 9.978973420669613e-06, "epoch": 0.37824, "percentage": 12.63, "elapsed_time": "0:29:05", "remaining_time": "3:21:13"}
+{"current_steps": 198, "total_steps": 1560, "loss": 0.3104, "lr": 9.977935989714594e-06, "epoch": 0.38016, "percentage": 12.69, "elapsed_time": "0:29:14", "remaining_time": "3:21:05"}
+{"current_steps": 199, "total_steps": 1560, "loss": 0.2319, "lr": 9.97687363493669e-06, "epoch": 0.38208, "percentage": 12.76, "elapsed_time": "0:29:22", "remaining_time": "3:20:57"}
+{"current_steps": 200, "total_steps": 1560, "loss": 0.2154, "lr": 9.975786361654959e-06, "epoch": 0.384, "percentage": 12.82, "elapsed_time": "0:29:31", "remaining_time": "3:20:48"}
+{"current_steps": 201, "total_steps": 1560, "loss": 0.2229, "lr": 9.974674175313228e-06, "epoch": 0.38592, "percentage": 12.88, "elapsed_time": "0:29:40", "remaining_time": "3:20:41"}
+{"current_steps": 202, "total_steps": 1560, "loss": 0.2301, "lr": 9.973537081480056e-06, "epoch": 0.38784, "percentage": 12.95, "elapsed_time": "0:29:49", "remaining_time": "3:20:32"}
+{"current_steps": 203, "total_steps": 1560, "loss": 0.2287, "lr": 9.972375085848712e-06, "epoch": 0.38976, "percentage": 13.01, "elapsed_time": "0:29:58", "remaining_time": "3:20:22"}
+{"current_steps": 204, "total_steps": 1560, "loss": 0.1856, "lr": 9.971188194237141e-06, "epoch": 0.39168, "percentage": 13.08, "elapsed_time": "0:30:07", "remaining_time": "3:20:13"}
+{"current_steps": 205, "total_steps": 1560, "loss": 0.2867, "lr": 9.969976412587943e-06, "epoch": 0.3936, "percentage": 13.14, "elapsed_time": "0:30:16", "remaining_time": "3:20:04"}
+{"current_steps": 206, "total_steps": 1560, "loss": 0.2461, "lr": 9.96873974696834e-06, "epoch": 0.39552, "percentage": 13.21, "elapsed_time": "0:30:24", "remaining_time": "3:19:54"}
+{"current_steps": 207, "total_steps": 1560, "loss": 0.2656, "lr": 9.967478203570143e-06, "epoch": 0.39744, "percentage": 13.27, "elapsed_time": "0:30:33", "remaining_time": "3:19:45"}
+{"current_steps": 208, "total_steps": 1560, "loss": 0.2487, "lr": 9.966191788709716e-06, "epoch": 0.39936, "percentage": 13.33, "elapsed_time": "0:30:42", "remaining_time": "3:19:36"}
+{"current_steps": 209, "total_steps": 1560, "loss": 0.2085, "lr": 9.96488050882796e-06, "epoch": 0.40128, "percentage": 13.4, "elapsed_time": "0:30:51", "remaining_time": "3:19:27"}
+{"current_steps": 210, "total_steps": 1560, "loss": 0.2608, "lr": 9.96354437049027e-06, "epoch": 0.4032, "percentage": 13.46, "elapsed_time": "0:31:00", "remaining_time": "3:19:17"}
+{"current_steps": 211, "total_steps": 1560, "loss": 0.2438, "lr": 9.9621833803865e-06, "epoch": 0.40512, "percentage": 13.53, "elapsed_time": "0:31:08", "remaining_time": "3:19:08"}
+{"current_steps": 212, "total_steps": 1560, "loss": 0.2788, "lr": 9.960797545330936e-06, "epoch": 0.40704, "percentage": 13.59, "elapsed_time": "0:31:18", "remaining_time": "3:19:01"}
+{"current_steps": 213, "total_steps": 1560, "loss": 0.2642, "lr": 9.95938687226226e-06, "epoch": 0.40896, "percentage": 13.65, "elapsed_time": "0:31:26", "remaining_time": "3:18:52"}
+{"current_steps": 214, "total_steps": 1560, "loss": 0.2463, "lr": 9.95795136824351e-06, "epoch": 0.41088, "percentage": 13.72, "elapsed_time": "0:31:35", "remaining_time": "3:18:44"}
+{"current_steps": 215, "total_steps": 1560, "loss": 0.2721, "lr": 9.956491040462051e-06, "epoch": 0.4128, "percentage": 13.78, "elapsed_time": "0:31:44", "remaining_time": "3:18:36"}
+{"current_steps": 216, "total_steps": 1560, "loss": 0.3134, "lr": 9.955005896229543e-06, "epoch": 0.41472, "percentage": 13.85, "elapsed_time": "0:31:53", "remaining_time": "3:18:27"}
+{"current_steps": 217, "total_steps": 1560, "loss": 0.2955, "lr": 9.95349594298189e-06, "epoch": 0.41664, "percentage": 13.91, "elapsed_time": "0:32:02", "remaining_time": "3:18:19"}
+{"current_steps": 218, "total_steps": 1560, "loss": 0.2684, "lr": 9.951961188279216e-06, "epoch": 0.41856, "percentage": 13.97, "elapsed_time": "0:32:11", "remaining_time": "3:18:11"}
+{"current_steps": 219, "total_steps": 1560, "loss": 0.2682, "lr": 9.950401639805822e-06, "epoch": 0.42048, "percentage": 14.04, "elapsed_time": "0:32:20", "remaining_time": "3:18:02"}
+{"current_steps": 220, "total_steps": 1560, "loss": 0.2926, "lr": 9.948817305370145e-06, "epoch": 0.4224, "percentage": 14.1, "elapsed_time": "0:32:29", "remaining_time": "3:17:53"}
+{"current_steps": 221, "total_steps": 1560, "loss": 0.2757, "lr": 9.947208192904722e-06, "epoch": 0.42432, "percentage": 14.17, "elapsed_time": "0:32:38", "remaining_time": "3:17:44"}
+{"current_steps": 222, "total_steps": 1560, "loss": 0.2266, "lr": 9.94557431046616e-06, "epoch": 0.42624, "percentage": 14.23, "elapsed_time": "0:32:47", "remaining_time": "3:17:36"}
+{"current_steps": 223, "total_steps": 1560, "loss": 0.2525, "lr": 9.943915666235068e-06, "epoch": 0.42816, "percentage": 14.29, "elapsed_time": "0:32:56", "remaining_time": "3:17:28"}
+{"current_steps": 224, "total_steps": 1560, "loss": 0.3127, "lr": 9.942232268516051e-06, "epoch": 0.43008, "percentage": 14.36, "elapsed_time": "0:33:05", "remaining_time": "3:17:19"}
+{"current_steps": 225, "total_steps": 1560, "loss": 0.219, "lr": 9.940524125737641e-06, "epoch": 0.432, "percentage": 14.42, "elapsed_time": "0:33:14", "remaining_time": "3:17:11"}
+{"current_steps": 226, "total_steps": 1560, "loss": 0.2066, "lr": 9.938791246452267e-06, "epoch": 0.43392, "percentage": 14.49, "elapsed_time": "0:33:22", "remaining_time": "3:17:02"}
+{"current_steps": 227, "total_steps": 1560, "loss": 0.2142, "lr": 9.937033639336212e-06, "epoch": 0.43584, "percentage": 14.55, "elapsed_time": "0:33:31", "remaining_time": "3:16:54"}
+{"current_steps": 228, "total_steps": 1560, "loss": 0.2202, "lr": 9.935251313189564e-06, "epoch": 0.43776, "percentage": 14.62, "elapsed_time": "0:33:40", "remaining_time": "3:16:45"}
+{"current_steps": 229, "total_steps": 1560, "loss": 0.2642, "lr": 9.933444276936185e-06, "epoch": 0.43968, "percentage": 14.68, "elapsed_time": "0:33:49", "remaining_time": "3:16:37"}
+{"current_steps": 230, "total_steps": 1560, "loss": 0.3228, "lr": 9.931612539623643e-06, "epoch": 0.4416, "percentage": 14.74, "elapsed_time": "0:33:58", "remaining_time": "3:16:28"}
+{"current_steps": 231, "total_steps": 1560, "loss": 0.2109, "lr": 9.929756110423195e-06, "epoch": 0.44352, "percentage": 14.81, "elapsed_time": "0:34:07", "remaining_time": "3:16:20"}
+{"current_steps": 232, "total_steps": 1560, "loss": 0.3153, "lr": 9.927874998629714e-06, "epoch": 0.44544, "percentage": 14.87, "elapsed_time": "0:34:16", "remaining_time": "3:16:11"}
+{"current_steps": 233, "total_steps": 1560, "loss": 0.2277, "lr": 9.925969213661664e-06, "epoch": 0.44736, "percentage": 14.94, "elapsed_time": "0:34:25", "remaining_time": "3:16:03"}
+{"current_steps": 234, "total_steps": 1560, "loss": 0.2711, "lr": 9.924038765061042e-06, "epoch": 0.44928, "percentage": 15.0, "elapsed_time": "0:34:34", "remaining_time": "3:15:54"}
+{"current_steps": 235, "total_steps": 1560, "loss": 0.329, "lr": 9.92208366249333e-06, "epoch": 0.4512, "percentage": 15.06, "elapsed_time": "0:34:43", "remaining_time": "3:15:46"}
+{"current_steps": 236, "total_steps": 1560, "loss": 0.2506, "lr": 9.920103915747452e-06, "epoch": 0.45312, "percentage": 15.13, "elapsed_time": "0:34:52", "remaining_time": "3:15:37"}
+{"current_steps": 237, "total_steps": 1560, "loss": 0.2404, "lr": 9.91809953473572e-06, "epoch": 0.45504, "percentage": 15.19, "elapsed_time": "0:35:01", "remaining_time": "3:15:29"}
+{"current_steps": 238, "total_steps": 1560, "loss": 0.2438, "lr": 9.916070529493785e-06, "epoch": 0.45696, "percentage": 15.26, "elapsed_time": "0:35:09", "remaining_time": "3:15:20"}
+{"current_steps": 239, "total_steps": 1560, "loss": 0.2678, "lr": 9.914016910180593e-06, "epoch": 0.45888, "percentage": 15.32, "elapsed_time": "0:35:18", "remaining_time": "3:15:11"}
+{"current_steps": 240, "total_steps": 1560, "loss": 0.2353, "lr": 9.911938687078324e-06, "epoch": 0.4608, "percentage": 15.38, "elapsed_time": "0:35:27", "remaining_time": "3:15:02"}
+{"current_steps": 241, "total_steps": 1560, "loss": 0.2483, "lr": 9.90983587059235e-06, "epoch": 0.46272, "percentage": 15.45, "elapsed_time": "0:35:36", "remaining_time": "3:14:54"}
+{"current_steps": 242, "total_steps": 1560, "loss": 0.2358, "lr": 9.907708471251173e-06, "epoch": 0.46464, "percentage": 15.51, "elapsed_time": "0:35:45", "remaining_time": "3:14:46"}
+{"current_steps": 243, "total_steps": 1560, "loss": 0.2316, "lr": 9.905556499706384e-06, "epoch": 0.46656, "percentage": 15.58, "elapsed_time": "0:35:54", "remaining_time": "3:14:38"}
+{"current_steps": 244, "total_steps": 1560, "loss": 0.2781, "lr": 9.9033799667326e-06, "epoch": 0.46848, "percentage": 15.64, "elapsed_time": "0:36:03", "remaining_time": "3:14:31"}
+{"current_steps": 245, "total_steps": 1560, "loss": 0.2393, "lr": 9.901178883227413e-06, "epoch": 0.4704, "percentage": 15.71, "elapsed_time": "0:36:13", "remaining_time": "3:14:23"}
+{"current_steps": 246, "total_steps": 1560, "loss": 0.2374, "lr": 9.89895326021134e-06, "epoch": 0.47232, "percentage": 15.77, "elapsed_time": "0:36:22", "remaining_time": "3:14:15"}
+{"current_steps": 247, "total_steps": 1560, "loss": 0.3059, "lr": 9.896703108827758e-06, "epoch": 0.47424, "percentage": 15.83, "elapsed_time": "0:36:31", "remaining_time": "3:14:07"}
+{"current_steps": 248, "total_steps": 1560, "loss": 0.2663, "lr": 9.89442844034286e-06, "epoch": 0.47616, "percentage": 15.9, "elapsed_time": "0:36:40", "remaining_time": "3:13:59"}
+{"current_steps": 249, "total_steps": 1560, "loss": 0.2775, "lr": 9.89212926614559e-06, "epoch": 0.47808, "percentage": 15.96, "elapsed_time": "0:36:49", "remaining_time": "3:13:50"}
+{"current_steps": 250, "total_steps": 1560, "loss": 0.2131, "lr": 9.889805597747588e-06, "epoch": 0.48, "percentage": 16.03, "elapsed_time": "0:36:57", "remaining_time": "3:13:41"}
+{"current_steps": 251, "total_steps": 1560, "loss": 0.2532, "lr": 9.887457446783133e-06, "epoch": 0.48192, "percentage": 16.09, "elapsed_time": "0:37:06", "remaining_time": "3:13:33"}
+{"current_steps": 252, "total_steps": 1560, "loss": 0.263, "lr": 9.885084825009085e-06, "epoch": 0.48384, "percentage": 16.15, "elapsed_time": "0:37:16", "remaining_time": "3:13:26"}
+{"current_steps": 253, "total_steps": 1560, "loss": 0.2543, "lr": 9.88268774430483e-06, "epoch": 0.48576, "percentage": 16.22, "elapsed_time": "0:37:25", "remaining_time": "3:13:19"}
+{"current_steps": 254, "total_steps": 1560, "loss": 0.2572, "lr": 9.88026621667221e-06, "epoch": 0.48768, "percentage": 16.28, "elapsed_time": "0:37:34", "remaining_time": "3:13:12"}
+{"current_steps": 255, "total_steps": 1560, "loss": 0.2467, "lr": 9.87782025423547e-06, "epoch": 0.4896, "percentage": 16.35, "elapsed_time": "0:37:43", "remaining_time": "3:13:05"}
+{"current_steps": 256, "total_steps": 1560, "loss": 0.231, "lr": 9.875349869241202e-06, "epoch": 0.49152, "percentage": 16.41, "elapsed_time": "0:37:52", "remaining_time": "3:12:56"}
+{"current_steps": 257, "total_steps": 1560, "loss": 0.2807, "lr": 9.872855074058274e-06, "epoch": 0.49344, "percentage": 16.47, "elapsed_time": "0:38:01", "remaining_time": "3:12:47"}
+{"current_steps": 258, "total_steps": 1560, "loss": 0.2478, "lr": 9.870335881177774e-06, "epoch": 0.49536, "percentage": 16.54, "elapsed_time": "0:38:10", "remaining_time": "3:12:39"}
+{"current_steps": 259, "total_steps": 1560, "loss": 0.2977, "lr": 9.867792303212944e-06, "epoch": 0.49728, "percentage": 16.6, "elapsed_time": "0:38:19", "remaining_time": "3:12:30"}
+{"current_steps": 260, "total_steps": 1560, "loss": 0.2454, "lr": 9.86522435289912e-06, "epoch": 0.4992, "percentage": 16.67, "elapsed_time": "0:38:28", "remaining_time": "3:12:21"}
+{"current_steps": 261, "total_steps": 1560, "loss": 0.2201, "lr": 9.862632043093667e-06, "epoch": 0.50112, "percentage": 16.73, "elapsed_time": "0:38:37", "remaining_time": "3:12:13"}
+{"current_steps": 262, "total_steps": 1560, "loss": 0.2765, "lr": 9.860015386775915e-06, "epoch": 0.50304, "percentage": 16.79, "elapsed_time": "0:38:46", "remaining_time": "3:12:04"}
+{"current_steps": 263, "total_steps": 1560, "loss": 0.2618, "lr": 9.857374397047093e-06, "epoch": 0.50496, "percentage": 16.86, "elapsed_time": "0:38:55", "remaining_time": "3:11:57"}
+{"current_steps": 264, "total_steps": 1560, "loss": 0.2449, "lr": 9.854709087130261e-06, "epoch": 0.50688, "percentage": 16.92, "elapsed_time": "0:39:04", "remaining_time": "3:11:50"}
+{"current_steps": 265, "total_steps": 1560, "loss": 0.2252, "lr": 9.852019470370254e-06, "epoch": 0.5088, "percentage": 16.99, "elapsed_time": "0:39:14", "remaining_time": "3:11:44"}
+{"current_steps": 266, "total_steps": 1560, "loss": 0.2446, "lr": 9.849305560233598e-06, "epoch": 0.51072, "percentage": 17.05, "elapsed_time": "0:39:23", "remaining_time": "3:11:37"}
+{"current_steps": 267, "total_steps": 1560, "loss": 0.2047, "lr": 9.846567370308462e-06, "epoch": 0.51264, "percentage": 17.12, "elapsed_time": "0:39:32", "remaining_time": "3:11:30"}
+{"current_steps": 268, "total_steps": 1560, "loss": 0.2869, "lr": 9.843804914304578e-06, "epoch": 0.51456, "percentage": 17.18, "elapsed_time": "0:39:41", "remaining_time": "3:11:22"}
+{"current_steps": 269, "total_steps": 1560, "loss": 0.24, "lr": 9.841018206053167e-06, "epoch": 0.51648, "percentage": 17.24, "elapsed_time": "0:39:50", "remaining_time": "3:11:14"}
+{"current_steps": 270, "total_steps": 1560, "loss": 0.2289, "lr": 9.838207259506891e-06, "epoch": 0.5184, "percentage": 17.31, "elapsed_time": "0:40:00", "remaining_time": "3:11:06"}
+{"current_steps": 271, "total_steps": 1560, "loss": 0.2737, "lr": 9.835372088739758e-06, "epoch": 0.52032, "percentage": 17.37, "elapsed_time": "0:40:09", "remaining_time": "3:10:58"}
+{"current_steps": 272, "total_steps": 1560, "loss": 0.2401, "lr": 9.83251270794707e-06, "epoch": 0.52224, "percentage": 17.44, "elapsed_time": "0:40:18", "remaining_time": "3:10:50"}
+{"current_steps": 273, "total_steps": 1560, "loss": 0.259, "lr": 9.829629131445342e-06, "epoch": 0.52416, "percentage": 17.5, "elapsed_time": "0:40:27", "remaining_time": "3:10:42"}
+{"current_steps": 274, "total_steps": 1560, "loss": 0.2832, "lr": 9.826721373672235e-06, "epoch": 0.52608, "percentage": 17.56, "elapsed_time": "0:40:36", "remaining_time": "3:10:35"}
+{"current_steps": 275, "total_steps": 1560, "loss": 0.232, "lr": 9.82378944918648e-06, "epoch": 0.528, "percentage": 17.63, "elapsed_time": "0:40:45", "remaining_time": "3:10:27"}
+{"current_steps": 276, "total_steps": 1560, "loss": 0.2507, "lr": 9.820833372667813e-06, "epoch": 0.52992, "percentage": 17.69, "elapsed_time": "0:40:54", "remaining_time": "3:10:19"}
+{"current_steps": 277, "total_steps": 1560, "loss": 0.2564, "lr": 9.817853158916889e-06, "epoch": 0.53184, "percentage": 17.76, "elapsed_time": "0:41:03", "remaining_time": "3:10:11"}
+{"current_steps": 278, "total_steps": 1560, "loss": 0.2451, "lr": 9.814848822855216e-06, "epoch": 0.53376, "percentage": 17.82, "elapsed_time": "0:41:12", "remaining_time": "3:10:03"}
+{"current_steps": 279, "total_steps": 1560, "loss": 0.2482, "lr": 9.811820379525085e-06, "epoch": 0.53568, "percentage": 17.88, "elapsed_time": "0:41:21", "remaining_time": "3:09:55"}
+{"current_steps": 280, "total_steps": 1560, "loss": 0.2532, "lr": 9.80876784408948e-06, "epoch": 0.5376, "percentage": 17.95, "elapsed_time": "0:41:31", "remaining_time": "3:09:47"}
+{"current_steps": 281, "total_steps": 1560, "loss": 0.227, "lr": 9.805691231832018e-06, "epoch": 0.53952, "percentage": 18.01, "elapsed_time": "0:41:40", "remaining_time": "3:09:39"}
+{"current_steps": 282, "total_steps": 1560, "loss": 0.2132, "lr": 9.802590558156863e-06, "epoch": 0.54144, "percentage": 18.08, "elapsed_time": "0:41:49", "remaining_time": "3:09:32"}
+{"current_steps": 283, "total_steps": 1560, "loss": 0.2228, "lr": 9.799465838588646e-06, "epoch": 0.54336, "percentage": 18.14, "elapsed_time": "0:41:58", "remaining_time": "3:09:24"}
+{"current_steps": 284, "total_steps": 1560, "loss": 0.3078, "lr": 9.796317088772402e-06, "epoch": 0.54528, "percentage": 18.21, "elapsed_time": "0:42:07", "remaining_time": "3:09:16"}
+{"current_steps": 285, "total_steps": 1560, "loss": 0.2625, "lr": 9.793144324473473e-06, "epoch": 0.5472, "percentage": 18.27, "elapsed_time": "0:42:16", "remaining_time": "3:09:08"}
+{"current_steps": 286, "total_steps": 1560, "loss": 0.267, "lr": 9.789947561577445e-06, "epoch": 0.54912, "percentage": 18.33, "elapsed_time": "0:42:25", "remaining_time": "3:09:00"}
+{"current_steps": 287, "total_steps": 1560, "loss": 0.2634, "lr": 9.786726816090058e-06, "epoch": 0.55104, "percentage": 18.4, "elapsed_time": "0:42:34", "remaining_time": "3:08:52"}
+{"current_steps": 288, "total_steps": 1560, "loss": 0.2592, "lr": 9.783482104137127e-06, "epoch": 0.55296, "percentage": 18.46, "elapsed_time": "0:42:44", "remaining_time": "3:08:44"}
+{"current_steps": 289, "total_steps": 1560, "loss": 0.2691, "lr": 9.78021344196447e-06, "epoch": 0.55488, "percentage": 18.53, "elapsed_time": "0:42:53", "remaining_time": "3:08:36"}
+{"current_steps": 290, "total_steps": 1560, "loss": 0.2382, "lr": 9.776920845937816e-06, "epoch": 0.5568, "percentage": 18.59, "elapsed_time": "0:43:02", "remaining_time": "3:08:28"}
+{"current_steps": 291, "total_steps": 1560, "loss": 0.2586, "lr": 9.77360433254273e-06, "epoch": 0.55872, "percentage": 18.65, "elapsed_time": "0:43:11", "remaining_time": "3:08:20"}
+{"current_steps": 292, "total_steps": 1560, "loss": 0.229, "lr": 9.770263918384523e-06, "epoch": 0.56064, "percentage": 18.72, "elapsed_time": "0:43:20", "remaining_time": "3:08:12"}
+{"current_steps": 293, "total_steps": 1560, "loss": 0.3025, "lr": 9.766899620188181e-06, "epoch": 0.56256, "percentage": 18.78, "elapsed_time": "0:43:29", "remaining_time": "3:08:04"}
+{"current_steps": 294, "total_steps": 1560, "loss": 0.2369, "lr": 9.763511454798268e-06, "epoch": 0.56448, "percentage": 18.85, "elapsed_time": "0:43:38", "remaining_time": "3:07:56"}
+{"current_steps": 295, "total_steps": 1560, "loss": 0.3104, "lr": 9.760099439178852e-06, "epoch": 0.5664, "percentage": 18.91, "elapsed_time": "0:43:47", "remaining_time": "3:07:48"}
+{"current_steps": 296, "total_steps": 1560, "loss": 0.2098, "lr": 9.75666359041341e-06, "epoch": 0.56832, "percentage": 18.97, "elapsed_time": "0:43:56", "remaining_time": "3:07:40"}
+{"current_steps": 297, "total_steps": 1560, "loss": 0.2525, "lr": 9.753203925704756e-06, "epoch": 0.57024, "percentage": 19.04, "elapsed_time": "0:44:06", "remaining_time": "3:07:32"}
+{"current_steps": 298, "total_steps": 1560, "loss": 0.2153, "lr": 9.749720462374939e-06, "epoch": 0.57216, "percentage": 19.1, "elapsed_time": "0:44:15", "remaining_time": "3:07:24"}
+{"current_steps": 299, "total_steps": 1560, "loss": 0.2685, "lr": 9.74621321786517e-06, "epoch": 0.57408, "percentage": 19.17, "elapsed_time": "0:44:24", "remaining_time": "3:07:16"}
+{"current_steps": 300, "total_steps": 1560, "loss": 0.2324, "lr": 9.742682209735727e-06, "epoch": 0.576, "percentage": 19.23, "elapsed_time": "0:44:33", "remaining_time": "3:07:08"}
+{"current_steps": 301, "total_steps": 1560, "loss": 0.2474, "lr": 9.739127455665872e-06, "epoch": 0.57792, "percentage": 19.29, "elapsed_time": "0:44:42", "remaining_time": "3:07:00"}
+{"current_steps": 302, "total_steps": 1560, "loss": 0.2634, "lr": 9.735548973453753e-06, "epoch": 0.57984, "percentage": 19.36, "elapsed_time": "0:44:51", "remaining_time": "3:06:50"}
+{"current_steps": 303, "total_steps": 1560, "loss": 0.2708, "lr": 9.731946781016328e-06, "epoch": 0.58176, "percentage": 19.42, "elapsed_time": "0:45:00", "remaining_time": "3:06:42"}
+{"current_steps": 304, "total_steps": 1560, "loss": 0.234, "lr": 9.728320896389263e-06, "epoch": 0.58368, "percentage": 19.49, "elapsed_time": "0:45:09", "remaining_time": "3:06:33"}
+{"current_steps": 305, "total_steps": 1560, "loss": 0.2443, "lr": 9.724671337726855e-06, "epoch": 0.5856, "percentage": 19.55, "elapsed_time": "0:45:18", "remaining_time": "3:06:25"}
+{"current_steps": 306, "total_steps": 1560, "loss": 0.2274, "lr": 9.720998123301924e-06, "epoch": 0.58752, "percentage": 19.62, "elapsed_time": "0:45:27", "remaining_time": "3:06:17"}
+{"current_steps": 307, "total_steps": 1560, "loss": 0.2662, "lr": 9.717301271505739e-06, "epoch": 0.58944, "percentage": 19.68, "elapsed_time": "0:45:36", "remaining_time": "3:06:09"}
+{"current_steps": 308, "total_steps": 1560, "loss": 0.2759, "lr": 9.713580800847917e-06, "epoch": 0.59136, "percentage": 19.74, "elapsed_time": "0:45:45", "remaining_time": "3:06:01"}
+{"current_steps": 309, "total_steps": 1560, "loss": 0.2825, "lr": 9.709836729956326e-06, "epoch": 0.59328, "percentage": 19.81, "elapsed_time": "0:45:54", "remaining_time": "3:05:53"}
+{"current_steps": 310, "total_steps": 1560, "loss": 0.187, "lr": 9.706069077577e-06, "epoch": 0.5952, "percentage": 19.87, "elapsed_time": "0:46:03", "remaining_time": "3:05:44"}
+{"current_steps": 311, "total_steps": 1560, "loss": 0.2461, "lr": 9.702277862574048e-06, "epoch": 0.59712, "percentage": 19.94, "elapsed_time": "0:46:13", "remaining_time": "3:05:36"}
+{"current_steps": 312, "total_steps": 1560, "loss": 0.2962, "lr": 9.698463103929542e-06, "epoch": 0.59904, "percentage": 20.0, "elapsed_time": "0:46:22", "remaining_time": "3:05:28"}
+{"current_steps": 313, "total_steps": 1560, "loss": 0.2364, "lr": 9.694624820743446e-06, "epoch": 0.60096, "percentage": 20.06, "elapsed_time": "0:46:31", "remaining_time": "3:05:20"}
+{"current_steps": 314, "total_steps": 1560, "loss": 0.2071, "lr": 9.690763032233498e-06, "epoch": 0.60288, "percentage": 20.13, "elapsed_time": "0:46:40", "remaining_time": "3:05:12"}
+{"current_steps": 315, "total_steps": 1560, "loss": 0.2162, "lr": 9.686877757735126e-06, "epoch": 0.6048, "percentage": 20.19, "elapsed_time": "0:46:49", "remaining_time": "3:05:04"}
+{"current_steps": 316, "total_steps": 1560, "loss": 0.2353, "lr": 9.682969016701357e-06, "epoch": 0.60672, "percentage": 20.26, "elapsed_time": "0:46:58", "remaining_time": "3:04:56"}
+{"current_steps": 317, "total_steps": 1560, "loss": 0.2603, "lr": 9.679036828702704e-06, "epoch": 0.60864, "percentage": 20.32, "elapsed_time": "0:47:07", "remaining_time": "3:04:48"}
+{"current_steps": 318, "total_steps": 1560, "loss": 0.2977, "lr": 9.675081213427076e-06, "epoch": 0.61056, "percentage": 20.38, "elapsed_time": "0:47:17", "remaining_time": "3:04:40"}
+{"current_steps": 319, "total_steps": 1560, "loss": 0.2437, "lr": 9.671102190679678e-06, "epoch": 0.61248, "percentage": 20.45, "elapsed_time": "0:47:26", "remaining_time": "3:04:32"}
+{"current_steps": 320, "total_steps": 1560, "loss": 0.3418, "lr": 9.66709978038292e-06, "epoch": 0.6144, "percentage": 20.51, "elapsed_time": "0:47:35", "remaining_time": "3:04:23"}
+{"current_steps": 321, "total_steps": 1560, "loss": 0.2355, "lr": 9.663074002576303e-06, "epoch": 0.61632, "percentage": 20.58, "elapsed_time": "0:47:44", "remaining_time": "3:04:15"}
+{"current_steps": 322, "total_steps": 1560, "loss": 0.2992, "lr": 9.659024877416328e-06, "epoch": 0.61824, "percentage": 20.64, "elapsed_time": "0:47:53", "remaining_time": "3:04:07"}
+{"current_steps": 323, "total_steps": 1560, "loss": 0.2228, "lr": 9.654952425176392e-06, "epoch": 0.62016, "percentage": 20.71, "elapsed_time": "0:48:02", "remaining_time": "3:03:58"}
+{"current_steps": 324, "total_steps": 1560, "loss": 0.2411, "lr": 9.650856666246693e-06, "epoch": 0.62208, "percentage": 20.77, "elapsed_time": "0:48:11", "remaining_time": "3:03:49"}
+{"current_steps": 325, "total_steps": 1560, "loss": 0.2739, "lr": 9.646737621134112e-06, "epoch": 0.624, "percentage": 20.83, "elapsed_time": "0:48:20", "remaining_time": "3:03:41"}
+{"current_steps": 326, "total_steps": 1560, "loss": 0.2345, "lr": 9.642595310462133e-06, "epoch": 0.62592, "percentage": 20.9, "elapsed_time": "0:48:29", "remaining_time": "3:03:34"}
+{"current_steps": 327, "total_steps": 1560, "loss": 0.2388, "lr": 9.638429754970715e-06, "epoch": 0.62784, "percentage": 20.96, "elapsed_time": "0:48:39", "remaining_time": "3:03:27"}
+{"current_steps": 328, "total_steps": 1560, "loss": 0.2207, "lr": 9.63424097551621e-06, "epoch": 0.62976, "percentage": 21.03, "elapsed_time": "0:48:48", "remaining_time": "3:03:19"}
+{"current_steps": 329, "total_steps": 1560, "loss": 0.2702, "lr": 9.630028993071244e-06, "epoch": 0.63168, "percentage": 21.09, "elapsed_time": "0:48:57", "remaining_time": "3:03:10"}
+{"current_steps": 330, "total_steps": 1560, "loss": 0.2273, "lr": 9.62579382872462e-06, "epoch": 0.6336, "percentage": 21.15, "elapsed_time": "0:49:06", "remaining_time": "3:03:02"}
+{"current_steps": 331, "total_steps": 1560, "loss": 0.2386, "lr": 9.621535503681205e-06, "epoch": 0.63552, "percentage": 21.22, "elapsed_time": "0:49:15", "remaining_time": "3:02:55"}
+{"current_steps": 332, "total_steps": 1560, "loss": 0.2184, "lr": 9.617254039261835e-06, "epoch": 0.63744, "percentage": 21.28, "elapsed_time": "0:49:25", "remaining_time": "3:02:47"}
+{"current_steps": 333, "total_steps": 1560, "loss": 0.2079, "lr": 9.612949456903195e-06, "epoch": 0.63936, "percentage": 21.35, "elapsed_time": "0:49:34", "remaining_time": "3:02:39"}
+{"current_steps": 334, "total_steps": 1560, "loss": 0.2433, "lr": 9.608621778157722e-06, "epoch": 0.64128, "percentage": 21.41, "elapsed_time": "0:49:43", "remaining_time": "3:02:31"}
+{"current_steps": 335, "total_steps": 1560, "loss": 0.2219, "lr": 9.604271024693495e-06, "epoch": 0.6432, "percentage": 21.47, "elapsed_time": "0:49:52", "remaining_time": "3:02:23"}
+{"current_steps": 336, "total_steps": 1560, "loss": 0.23, "lr": 9.599897218294122e-06, "epoch": 0.64512, "percentage": 21.54, "elapsed_time": "0:50:01", "remaining_time": "3:02:15"}
+{"current_steps": 337, "total_steps": 1560, "loss": 0.2779, "lr": 9.595500380858632e-06, "epoch": 0.64704, "percentage": 21.6, "elapsed_time": "0:50:10", "remaining_time": "3:02:06"}
+{"current_steps": 338, "total_steps": 1560, "loss": 0.2338, "lr": 9.591080534401371e-06, "epoch": 0.64896, "percentage": 21.67, "elapsed_time": "0:50:19", "remaining_time": "3:01:57"}
+{"current_steps": 339, "total_steps": 1560, "loss": 0.2695, "lr": 9.586637701051886e-06, "epoch": 0.65088, "percentage": 21.73, "elapsed_time": "0:50:28", "remaining_time": "3:01:48"}
+{"current_steps": 340, "total_steps": 1560, "loss": 0.2579, "lr": 9.582171903054815e-06, "epoch": 0.6528, "percentage": 21.79, "elapsed_time": "0:50:37", "remaining_time": "3:01:39"}
+{"current_steps": 341, "total_steps": 1560, "loss": 0.2276, "lr": 9.577683162769781e-06, "epoch": 0.65472, "percentage": 21.86, "elapsed_time": "0:50:46", "remaining_time": "3:01:30"}
+{"current_steps": 342, "total_steps": 1560, "loss": 0.2544, "lr": 9.573171502671273e-06, "epoch": 0.65664, "percentage": 21.92, "elapsed_time": "0:50:55", "remaining_time": "3:01:21"}
+{"current_steps": 343, "total_steps": 1560, "loss": 0.272, "lr": 9.568636945348534e-06, "epoch": 0.65856, "percentage": 21.99, "elapsed_time": "0:51:04", "remaining_time": "3:01:13"}
+{"current_steps": 344, "total_steps": 1560, "loss": 0.2599, "lr": 9.564079513505455e-06, "epoch": 0.66048, "percentage": 22.05, "elapsed_time": "0:51:13", "remaining_time": "3:01:04"}
+{"current_steps": 345, "total_steps": 1560, "loss": 0.2152, "lr": 9.55949922996045e-06, "epoch": 0.6624, "percentage": 22.12, "elapsed_time": "0:51:22", "remaining_time": "3:00:56"}
+{"current_steps": 346, "total_steps": 1560, "loss": 0.2811, "lr": 9.554896117646357e-06, "epoch": 0.66432, "percentage": 22.18, "elapsed_time": "0:51:31", "remaining_time": "3:00:47"}
+{"current_steps": 347, "total_steps": 1560, "loss": 0.2272, "lr": 9.550270199610307e-06, "epoch": 0.66624, "percentage": 22.24, "elapsed_time": "0:51:40", "remaining_time": "3:00:38"}
+{"current_steps": 348, "total_steps": 1560, "loss": 0.2291, "lr": 9.54562149901362e-06, "epoch": 0.66816, "percentage": 22.31, "elapsed_time": "0:51:49", "remaining_time": "3:00:30"}
+{"current_steps": 349, "total_steps": 1560, "loss": 0.2448, "lr": 9.54095003913168e-06, "epoch": 0.67008, "percentage": 22.37, "elapsed_time": "0:51:58", "remaining_time": "3:00:21"}
+{"current_steps": 350, "total_steps": 1560, "loss": 0.2286, "lr": 9.536255843353832e-06, "epoch": 0.672, "percentage": 22.44, "elapsed_time": "0:52:07", "remaining_time": "3:00:12"}
+{"current_steps": 351, "total_steps": 1560, "loss": 0.2127, "lr": 9.531538935183252e-06, "epoch": 0.67392, "percentage": 22.5, "elapsed_time": "0:52:16", "remaining_time": "3:00:04"}
+{"current_steps": 352, "total_steps": 1560, "loss": 0.2203, "lr": 9.526799338236828e-06, "epoch": 0.67584, "percentage": 22.56, "elapsed_time": "0:52:25", "remaining_time": "2:59:55"}
+{"current_steps": 353, "total_steps": 1560, "loss": 0.2186, "lr": 9.522037076245057e-06, "epoch": 0.67776, "percentage": 22.63, "elapsed_time": "0:52:34", "remaining_time": "2:59:46"}
+{"current_steps": 354, "total_steps": 1560, "loss": 0.2277, "lr": 9.517252173051912e-06, "epoch": 0.67968, "percentage": 22.69, "elapsed_time": "0:52:43", "remaining_time": "2:59:38"}
+{"current_steps": 355, "total_steps": 1560, "loss": 0.2633, "lr": 9.512444652614728e-06, "epoch": 0.6816, "percentage": 22.76, "elapsed_time": "0:52:53", "remaining_time": "2:59:30"}
+{"current_steps": 356, "total_steps": 1560, "loss": 0.2331, "lr": 9.507614539004082e-06, "epoch": 0.68352, "percentage": 22.82, "elapsed_time": "0:53:02", "remaining_time": "2:59:23"}
+{"current_steps": 357, "total_steps": 1560, "loss": 0.2154, "lr": 9.502761856403668e-06, "epoch": 0.68544, "percentage": 22.88, "elapsed_time": "0:53:11", "remaining_time": "2:59:15"}
+{"current_steps": 358, "total_steps": 1560, "loss": 0.2413, "lr": 9.497886629110187e-06, "epoch": 0.68736, "percentage": 22.95, "elapsed_time": "0:53:21", "remaining_time": "2:59:08"}
+{"current_steps": 359, "total_steps": 1560, "loss": 0.2609, "lr": 9.49298888153321e-06, "epoch": 0.68928, "percentage": 23.01, "elapsed_time": "0:53:30", "remaining_time": "2:59:00"}
+{"current_steps": 360, "total_steps": 1560, "loss": 0.2489, "lr": 9.488068638195072e-06, "epoch": 0.6912, "percentage": 23.08, "elapsed_time": "0:53:39", "remaining_time": "2:58:52"}
+{"current_steps": 361, "total_steps": 1560, "loss": 0.2371, "lr": 9.483125923730731e-06, "epoch": 0.69312, "percentage": 23.14, "elapsed_time": "0:53:49", "remaining_time": "2:58:45"}
+{"current_steps": 362, "total_steps": 1560, "loss": 0.2693, "lr": 9.478160762887668e-06, "epoch": 0.69504, "percentage": 23.21, "elapsed_time": "0:53:58", "remaining_time": "2:58:37"}
+{"current_steps": 363, "total_steps": 1560, "loss": 0.2894, "lr": 9.473173180525737e-06, "epoch": 0.69696, "percentage": 23.27, "elapsed_time": "0:54:07", "remaining_time": "2:58:29"}
+{"current_steps": 364, "total_steps": 1560, "loss": 0.1996, "lr": 9.468163201617063e-06, "epoch": 0.69888, "percentage": 23.33, "elapsed_time": "0:54:17", "remaining_time": "2:58:21"}
+{"current_steps": 365, "total_steps": 1560, "loss": 0.2431, "lr": 9.463130851245899e-06, "epoch": 0.7008, "percentage": 23.4, "elapsed_time": "0:54:26", "remaining_time": "2:58:13"}
+{"current_steps": 366, "total_steps": 1560, "loss": 0.2791, "lr": 9.458076154608515e-06, "epoch": 0.70272, "percentage": 23.46, "elapsed_time": "0:54:35", "remaining_time": "2:58:05"}
+{"current_steps": 367, "total_steps": 1560, "loss": 0.2147, "lr": 9.452999137013064e-06, "epoch": 0.70464, "percentage": 23.53, "elapsed_time": "0:54:44", "remaining_time": "2:57:58"}
+{"current_steps": 368, "total_steps": 1560, "loss": 0.2337, "lr": 9.447899823879456e-06, "epoch": 0.70656, "percentage": 23.59, "elapsed_time": "0:54:53", "remaining_time": "2:57:49"}
+{"current_steps": 369, "total_steps": 1560, "loss": 0.2626, "lr": 9.442778240739234e-06, "epoch": 0.70848, "percentage": 23.65, "elapsed_time": "0:55:02", "remaining_time": "2:57:40"}
+{"current_steps": 370, "total_steps": 1560, "loss": 0.2668, "lr": 9.437634413235437e-06, "epoch": 0.7104, "percentage": 23.72, "elapsed_time": "0:55:12", "remaining_time": "2:57:33"}
+{"current_steps": 371, "total_steps": 1560, "loss": 0.2261, "lr": 9.43246836712249e-06, "epoch": 0.71232, "percentage": 23.78, "elapsed_time": "0:55:21", "remaining_time": "2:57:25"}
+{"current_steps": 372, "total_steps": 1560, "loss": 0.2481, "lr": 9.427280128266049e-06, "epoch": 0.71424, "percentage": 23.85, "elapsed_time": "0:55:31", "remaining_time": "2:57:18"}
+{"current_steps": 373, "total_steps": 1560, "loss": 0.2454, "lr": 9.4220697226429e-06, "epoch": 0.71616, "percentage": 23.91, "elapsed_time": "0:55:40", "remaining_time": "2:57:10"}
+{"current_steps": 374, "total_steps": 1560, "loss": 0.2586, "lr": 9.416837176340807e-06, "epoch": 0.71808, "percentage": 23.97, "elapsed_time": "0:55:49", "remaining_time": "2:57:01"}
+{"current_steps": 375, "total_steps": 1560, "loss": 0.2808, "lr": 9.411582515558391e-06, "epoch": 0.72, "percentage": 24.04, "elapsed_time": "0:55:58", "remaining_time": "2:56:53"}
+{"current_steps": 376, "total_steps": 1560, "loss": 0.2344, "lr": 9.406305766604996e-06, "epoch": 0.72192, "percentage": 24.1, "elapsed_time": "0:56:07", "remaining_time": "2:56:45"}
+{"current_steps": 377, "total_steps": 1560, "loss": 0.2239, "lr": 9.401006955900555e-06, "epoch": 0.72384, "percentage": 24.17, "elapsed_time": "0:56:17", "remaining_time": "2:56:37"}
+{"current_steps": 378, "total_steps": 1560, "loss": 0.2578, "lr": 9.395686109975475e-06, "epoch": 0.72576, "percentage": 24.23, "elapsed_time": "0:56:26", "remaining_time": "2:56:29"}
+{"current_steps": 379, "total_steps": 1560, "loss": 0.2652, "lr": 9.390343255470471e-06, "epoch": 0.72768, "percentage": 24.29, "elapsed_time": "0:56:35", "remaining_time": "2:56:21"}
+{"current_steps": 380, "total_steps": 1560, "loss": 0.2274, "lr": 9.384978419136469e-06, "epoch": 0.7296, "percentage": 24.36, "elapsed_time": "0:56:45", "remaining_time": "2:56:13"}
+{"current_steps": 381, "total_steps": 1560, "loss": 0.2769, "lr": 9.37959162783444e-06, "epoch": 0.73152, "percentage": 24.42, "elapsed_time": "0:56:54", "remaining_time": "2:56:06"}
+{"current_steps": 382, "total_steps": 1560, "loss": 0.2246, "lr": 9.374182908535293e-06, "epoch": 0.73344, "percentage": 24.49, "elapsed_time": "0:57:03", "remaining_time": "2:55:58"}
+{"current_steps": 383, "total_steps": 1560, "loss": 0.2339, "lr": 9.368752288319722e-06, "epoch": 0.73536, "percentage": 24.55, "elapsed_time": "0:57:13", "remaining_time": "2:55:50"}
+{"current_steps": 384, "total_steps": 1560, "loss": 0.2307, "lr": 9.363299794378072e-06, "epoch": 0.73728, "percentage": 24.62, "elapsed_time": "0:57:22", "remaining_time": "2:55:41"}
+{"current_steps": 385, "total_steps": 1560, "loss": 0.245, "lr": 9.357825454010214e-06, "epoch": 0.7392, "percentage": 24.68, "elapsed_time": "0:57:31", "remaining_time": "2:55:32"}
+{"current_steps": 386, "total_steps": 1560, "loss": 0.267, "lr": 9.352329294625397e-06, "epoch": 0.74112, "percentage": 24.74, "elapsed_time": "0:57:40", "remaining_time": "2:55:23"}
+{"current_steps": 387, "total_steps": 1560, "loss": 0.2342, "lr": 9.346811343742115e-06, "epoch": 0.74304, "percentage": 24.81, "elapsed_time": "0:57:49", "remaining_time": "2:55:16"}
+{"current_steps": 388, "total_steps": 1560, "loss": 0.2405, "lr": 9.34127162898797e-06, "epoch": 0.74496, "percentage": 24.87, "elapsed_time": "0:57:58", "remaining_time": "2:55:07"}
+{"current_steps": 389, "total_steps": 1560, "loss": 0.2047, "lr": 9.335710178099528e-06, "epoch": 0.74688, "percentage": 24.94, "elapsed_time": "0:58:07", "remaining_time": "2:54:59"}
+{"current_steps": 390, "total_steps": 1560, "loss": 0.2278, "lr": 9.330127018922195e-06, "epoch": 0.7488, "percentage": 25.0, "elapsed_time": "0:58:16", "remaining_time": "2:54:50"}
+{"current_steps": 391, "total_steps": 1560, "loss": 0.2394, "lr": 9.324522179410054e-06, "epoch": 0.75072, "percentage": 25.06, "elapsed_time": "0:58:25", "remaining_time": "2:54:41"}
+{"current_steps": 392, "total_steps": 1560, "loss": 0.2602, "lr": 9.318895687625752e-06, "epoch": 0.75264, "percentage": 25.13, "elapsed_time": "0:58:34", "remaining_time": "2:54:32"}
+{"current_steps": 393, "total_steps": 1560, "loss": 0.2327, "lr": 9.313247571740336e-06, "epoch": 0.75456, "percentage": 25.19, "elapsed_time": "0:58:43", "remaining_time": "2:54:23"}
+{"current_steps": 394, "total_steps": 1560, "loss": 0.3077, "lr": 9.307577860033123e-06, "epoch": 0.75648, "percentage": 25.26, "elapsed_time": "0:58:52", "remaining_time": "2:54:14"}
+{"current_steps": 395, "total_steps": 1560, "loss": 0.22, "lr": 9.301886580891563e-06, "epoch": 0.7584, "percentage": 25.32, "elapsed_time": "0:59:01", "remaining_time": "2:54:05"}
+{"current_steps": 396, "total_steps": 1560, "loss": 0.2392, "lr": 9.296173762811084e-06, "epoch": 0.76032, "percentage": 25.38, "elapsed_time": "0:59:10", "remaining_time": "2:53:55"}
+{"current_steps": 397, "total_steps": 1560, "loss": 0.2732, "lr": 9.290439434394965e-06, "epoch": 0.76224, "percentage": 25.45, "elapsed_time": "0:59:19", "remaining_time": "2:53:46"}
+{"current_steps": 398, "total_steps": 1560, "loss": 0.2372, "lr": 9.284683624354172e-06, "epoch": 0.76416, "percentage": 25.51, "elapsed_time": "0:59:28", "remaining_time": "2:53:37"}
+{"current_steps": 399, "total_steps": 1560, "loss": 0.2265, "lr": 9.278906361507238e-06, "epoch": 0.76608, "percentage": 25.58, "elapsed_time": "0:59:36", "remaining_time": "2:53:28"}
+{"current_steps": 400, "total_steps": 1560, "loss": 0.2037, "lr": 9.273107674780102e-06, "epoch": 0.768, "percentage": 25.64, "elapsed_time": "0:59:46", "remaining_time": "2:53:20"}
+{"current_steps": 401, "total_steps": 1560, "loss": 0.2637, "lr": 9.26728759320597e-06, "epoch": 0.76992, "percentage": 25.71, "elapsed_time": "0:59:55", "remaining_time": "2:53:10"}
+{"current_steps": 402, "total_steps": 1560, "loss": 0.2869, "lr": 9.261446145925167e-06, "epoch": 0.77184, "percentage": 25.77, "elapsed_time": "1:00:04", "remaining_time": "2:53:01"}
+{"current_steps": 403, "total_steps": 1560, "loss": 0.322, "lr": 9.255583362184998e-06, "epoch": 0.77376, "percentage": 25.83, "elapsed_time": "1:00:12", "remaining_time": "2:52:52"}
+{"current_steps": 404, "total_steps": 1560, "loss": 0.2302, "lr": 9.249699271339594e-06, "epoch": 0.77568, "percentage": 25.9, "elapsed_time": "1:00:21", "remaining_time": "2:52:43"}
+{"current_steps": 405, "total_steps": 1560, "loss": 0.2705, "lr": 9.243793902849764e-06, "epoch": 0.7776, "percentage": 25.96, "elapsed_time": "1:00:30", "remaining_time": "2:52:34"}
+{"current_steps": 406, "total_steps": 1560, "loss": 0.2287, "lr": 9.237867286282855e-06, "epoch": 0.77952, "percentage": 26.03, "elapsed_time": "1:00:39", "remaining_time": "2:52:25"}
+{"current_steps": 407, "total_steps": 1560, "loss": 0.2882, "lr": 9.231919451312603e-06, "epoch": 0.78144, "percentage": 26.09, "elapsed_time": "1:00:48", "remaining_time": "2:52:16"}
+{"current_steps": 408, "total_steps": 1560, "loss": 0.2119, "lr": 9.225950427718974e-06, "epoch": 0.78336, "percentage": 26.15, "elapsed_time": "1:00:58", "remaining_time": "2:52:08"}
+{"current_steps": 409, "total_steps": 1560, "loss": 0.2188, "lr": 9.21996024538803e-06, "epoch": 0.78528, "percentage": 26.22, "elapsed_time": "1:01:07", "remaining_time": "2:51:59"}
+{"current_steps": 410, "total_steps": 1560, "loss": 0.2727, "lr": 9.213948934311767e-06, "epoch": 0.7872, "percentage": 26.28, "elapsed_time": "1:01:16", "remaining_time": "2:51:51"}
+{"current_steps": 411, "total_steps": 1560, "loss": 0.2552, "lr": 9.207916524587971e-06, "epoch": 0.78912, "percentage": 26.35, "elapsed_time": "1:01:25", "remaining_time": "2:51:42"}
+{"current_steps": 412, "total_steps": 1560, "loss": 0.207, "lr": 9.201863046420065e-06, "epoch": 0.79104, "percentage": 26.41, "elapsed_time": "1:01:34", "remaining_time": "2:51:33"}
+{"current_steps": 413, "total_steps": 1560, "loss": 0.2239, "lr": 9.195788530116962e-06, "epoch": 0.79296, "percentage": 26.47, "elapsed_time": "1:01:43", "remaining_time": "2:51:25"}
+{"current_steps": 414, "total_steps": 1560, "loss": 0.2013, "lr": 9.189693006092907e-06, "epoch": 0.79488, "percentage": 26.54, "elapsed_time": "1:01:52", "remaining_time": "2:51:16"}
+{"current_steps": 415, "total_steps": 1560, "loss": 0.2246, "lr": 9.183576504867328e-06, "epoch": 0.7968, "percentage": 26.6, "elapsed_time": "1:02:01", "remaining_time": "2:51:07"}
+{"current_steps": 416, "total_steps": 1560, "loss": 0.2432, "lr": 9.177439057064684e-06, "epoch": 0.79872, "percentage": 26.67, "elapsed_time": "1:02:10", "remaining_time": "2:50:59"}
+{"current_steps": 417, "total_steps": 1560, "loss": 0.2401, "lr": 9.171280693414307e-06, "epoch": 0.80064, "percentage": 26.73, "elapsed_time": "1:02:19", "remaining_time": "2:50:50"}
+{"current_steps": 418, "total_steps": 1560, "loss": 0.2778, "lr": 9.165101444750259e-06, "epoch": 0.80256, "percentage": 26.79, "elapsed_time": "1:02:28", "remaining_time": "2:50:41"}
+{"current_steps": 419, "total_steps": 1560, "loss": 0.2437, "lr": 9.158901342011163e-06, "epoch": 0.80448, "percentage": 26.86, "elapsed_time": "1:02:37", "remaining_time": "2:50:32"}
+{"current_steps": 420, "total_steps": 1560, "loss": 0.2411, "lr": 9.152680416240059e-06, "epoch": 0.8064, "percentage": 26.92, "elapsed_time": "1:02:46", "remaining_time": "2:50:24"}
+{"current_steps": 421, "total_steps": 1560, "loss": 0.2771, "lr": 9.146438698584245e-06, "epoch": 0.80832, "percentage": 26.99, "elapsed_time": "1:02:55", "remaining_time": "2:50:15"}
+{"current_steps": 422, "total_steps": 1560, "loss": 0.273, "lr": 9.140176220295118e-06, "epoch": 0.81024, "percentage": 27.05, "elapsed_time": "1:03:04", "remaining_time": "2:50:06"}
+{"current_steps": 423, "total_steps": 1560, "loss": 0.2371, "lr": 9.133893012728027e-06, "epoch": 0.81216, "percentage": 27.12, "elapsed_time": "1:03:13", "remaining_time": "2:49:57"}
+{"current_steps": 424, "total_steps": 1560, "loss": 0.2589, "lr": 9.1275891073421e-06, "epoch": 0.81408, "percentage": 27.18, "elapsed_time": "1:03:23", "remaining_time": "2:49:49"}
+{"current_steps": 425, "total_steps": 1560, "loss": 0.1931, "lr": 9.121264535700107e-06, "epoch": 0.816, "percentage": 27.24, "elapsed_time": "1:03:32", "remaining_time": "2:49:40"}
+{"current_steps": 426, "total_steps": 1560, "loss": 0.2583, "lr": 9.114919329468283e-06, "epoch": 0.81792, "percentage": 27.31, "elapsed_time": "1:03:41", "remaining_time": "2:49:32"}
+{"current_steps": 427, "total_steps": 1560, "loss": 0.2112, "lr": 9.10855352041618e-06, "epoch": 0.81984, "percentage": 27.37, "elapsed_time": "1:03:50", "remaining_time": "2:49:23"}
+{"current_steps": 428, "total_steps": 1560, "loss": 0.308, "lr": 9.102167140416503e-06, "epoch": 0.82176, "percentage": 27.44, "elapsed_time": "1:03:59", "remaining_time": "2:49:14"}
+{"current_steps": 429, "total_steps": 1560, "loss": 0.2577, "lr": 9.09576022144496e-06, "epoch": 0.82368, "percentage": 27.5, "elapsed_time": "1:04:08", "remaining_time": "2:49:05"}
+{"current_steps": 430, "total_steps": 1560, "loss": 0.2715, "lr": 9.089332795580085e-06, "epoch": 0.8256, "percentage": 27.56, "elapsed_time": "1:04:17", "remaining_time": "2:48:56"}
+{"current_steps": 431, "total_steps": 1560, "loss": 0.2401, "lr": 9.082884895003098e-06, "epoch": 0.82752, "percentage": 27.63, "elapsed_time": "1:04:26", "remaining_time": "2:48:47"}
+{"current_steps": 432, "total_steps": 1560, "loss": 0.2509, "lr": 9.076416551997721e-06, "epoch": 0.82944, "percentage": 27.69, "elapsed_time": "1:04:35", "remaining_time": "2:48:38"}
+{"current_steps": 433, "total_steps": 1560, "loss": 0.2181, "lr": 9.06992779895004e-06, "epoch": 0.83136, "percentage": 27.76, "elapsed_time": "1:04:44", "remaining_time": "2:48:29"}
+{"current_steps": 434, "total_steps": 1560, "loss": 0.276, "lr": 9.063418668348323e-06, "epoch": 0.83328, "percentage": 27.82, "elapsed_time": "1:04:53", "remaining_time": "2:48:20"}
+{"current_steps": 435, "total_steps": 1560, "loss": 0.2321, "lr": 9.056889192782865e-06, "epoch": 0.8352, "percentage": 27.88, "elapsed_time": "1:05:02", "remaining_time": "2:48:11"}
+{"current_steps": 436, "total_steps": 1560, "loss": 0.2403, "lr": 9.050339404945834e-06, "epoch": 0.83712, "percentage": 27.95, "elapsed_time": "1:05:11", "remaining_time": "2:48:02"}
+{"current_steps": 437, "total_steps": 1560, "loss": 0.2147, "lr": 9.04376933763109e-06, "epoch": 0.83904, "percentage": 28.01, "elapsed_time": "1:05:20", "remaining_time": "2:47:53"}
+{"current_steps": 438, "total_steps": 1560, "loss": 0.2683, "lr": 9.037179023734036e-06, "epoch": 0.84096, "percentage": 28.08, "elapsed_time": "1:05:29", "remaining_time": "2:47:44"}
+{"current_steps": 439, "total_steps": 1560, "loss": 0.2338, "lr": 9.030568496251437e-06, "epoch": 0.84288, "percentage": 28.14, "elapsed_time": "1:05:38", "remaining_time": "2:47:35"}
+{"current_steps": 440, "total_steps": 1560, "loss": 0.2691, "lr": 9.023937788281278e-06, "epoch": 0.8448, "percentage": 28.21, "elapsed_time": "1:05:46", "remaining_time": "2:47:26"}
+{"current_steps": 441, "total_steps": 1560, "loss": 0.2653, "lr": 9.017286933022573e-06, "epoch": 0.84672, "percentage": 28.27, "elapsed_time": "1:05:55", "remaining_time": "2:47:17"}
+{"current_steps": 442, "total_steps": 1560, "loss": 0.2334, "lr": 9.01061596377522e-06, "epoch": 0.84864, "percentage": 28.33, "elapsed_time": "1:06:04", "remaining_time": "2:47:08"}
+{"current_steps": 443, "total_steps": 1560, "loss": 0.3047, "lr": 9.003924913939816e-06, "epoch": 0.85056, "percentage": 28.4, "elapsed_time": "1:06:13", "remaining_time": "2:46:59"}
+{"current_steps": 444, "total_steps": 1560, "loss": 0.2504, "lr": 8.997213817017508e-06, "epoch": 0.85248, "percentage": 28.46, "elapsed_time": "1:06:22", "remaining_time": "2:46:50"}
+{"current_steps": 445, "total_steps": 1560, "loss": 0.2769, "lr": 8.990482706609805e-06, "epoch": 0.8544, "percentage": 28.53, "elapsed_time": "1:06:31", "remaining_time": "2:46:41"}
+{"current_steps": 446, "total_steps": 1560, "loss": 0.2186, "lr": 8.98373161641843e-06, "epoch": 0.85632, "percentage": 28.59, "elapsed_time": "1:06:40", "remaining_time": "2:46:33"}
+{"current_steps": 447, "total_steps": 1560, "loss": 0.2401, "lr": 8.97696058024514e-06, "epoch": 0.85824, "percentage": 28.65, "elapsed_time": "1:06:49", "remaining_time": "2:46:24"}
+{"current_steps": 448, "total_steps": 1560, "loss": 0.2109, "lr": 8.970169631991556e-06, "epoch": 0.86016, "percentage": 28.72, "elapsed_time": "1:06:58", "remaining_time": "2:46:15"}
+{"current_steps": 449, "total_steps": 1560, "loss": 0.2049, "lr": 8.963358805658998e-06, "epoch": 0.86208, "percentage": 28.78, "elapsed_time": "1:07:07", "remaining_time": "2:46:06"}
+{"current_steps": 450, "total_steps": 1560, "loss": 0.2398, "lr": 8.95652813534831e-06, "epoch": 0.864, "percentage": 28.85, "elapsed_time": "1:07:16", "remaining_time": "2:45:57"}
+{"current_steps": 451, "total_steps": 1560, "loss": 0.2639, "lr": 8.949677655259696e-06, "epoch": 0.86592, "percentage": 28.91, "elapsed_time": "1:07:25", "remaining_time": "2:45:48"}
+{"current_steps": 452, "total_steps": 1560, "loss": 0.2547, "lr": 8.942807399692543e-06, "epoch": 0.86784, "percentage": 28.97, "elapsed_time": "1:07:34", "remaining_time": "2:45:38"}
+{"current_steps": 453, "total_steps": 1560, "loss": 0.2519, "lr": 8.935917403045251e-06, "epoch": 0.86976, "percentage": 29.04, "elapsed_time": "1:07:43", "remaining_time": "2:45:29"}
+{"current_steps": 454, "total_steps": 1560, "loss": 0.248, "lr": 8.92900769981506e-06, "epoch": 0.87168, "percentage": 29.1, "elapsed_time": "1:07:52", "remaining_time": "2:45:20"}
+{"current_steps": 455, "total_steps": 1560, "loss": 0.2423, "lr": 8.92207832459788e-06, "epoch": 0.8736, "percentage": 29.17, "elapsed_time": "1:08:01", "remaining_time": "2:45:11"}
+{"current_steps": 456, "total_steps": 1560, "loss": 0.257, "lr": 8.915129312088112e-06, "epoch": 0.87552, "percentage": 29.23, "elapsed_time": "1:08:10", "remaining_time": "2:45:02"}
+{"current_steps": 457, "total_steps": 1560, "loss": 0.264, "lr": 8.908160697078482e-06, "epoch": 0.87744, "percentage": 29.29, "elapsed_time": "1:08:19", "remaining_time": "2:44:53"}
+{"current_steps": 458, "total_steps": 1560, "loss": 0.2499, "lr": 8.901172514459864e-06, "epoch": 0.87936, "percentage": 29.36, "elapsed_time": "1:08:27", "remaining_time": "2:44:44"}
+{"current_steps": 459, "total_steps": 1560, "loss": 0.2294, "lr": 8.894164799221099e-06, "epoch": 0.88128, "percentage": 29.42, "elapsed_time": "1:08:36", "remaining_time": "2:44:35"}
+{"current_steps": 460, "total_steps": 1560, "loss": 0.2477, "lr": 8.88713758644883e-06, "epoch": 0.8832, "percentage": 29.49, "elapsed_time": "1:08:45", "remaining_time": "2:44:26"}
+{"current_steps": 461, "total_steps": 1560, "loss": 0.2327, "lr": 8.88009091132732e-06, "epoch": 0.88512, "percentage": 29.55, "elapsed_time": "1:08:54", "remaining_time": "2:44:16"}
+{"current_steps": 462, "total_steps": 1560, "loss": 0.2594, "lr": 8.873024809138272e-06, "epoch": 0.88704, "percentage": 29.62, "elapsed_time": "1:09:03", "remaining_time": "2:44:07"}
+{"current_steps": 463, "total_steps": 1560, "loss": 0.2306, "lr": 8.86593931526067e-06, "epoch": 0.88896, "percentage": 29.68, "elapsed_time": "1:09:12", "remaining_time": "2:43:58"}
+{"current_steps": 464, "total_steps": 1560, "loss": 0.2541, "lr": 8.858834465170576e-06, "epoch": 0.89088, "percentage": 29.74, "elapsed_time": "1:09:21", "remaining_time": "2:43:49"}
+{"current_steps": 465, "total_steps": 1560, "loss": 0.2468, "lr": 8.851710294440974e-06, "epoch": 0.8928, "percentage": 29.81, "elapsed_time": "1:09:30", "remaining_time": "2:43:41"}
+{"current_steps": 466, "total_steps": 1560, "loss": 0.2246, "lr": 8.84456683874158e-06, "epoch": 0.89472, "percentage": 29.87, "elapsed_time": "1:09:39", "remaining_time": "2:43:32"}
+{"current_steps": 467, "total_steps": 1560, "loss": 0.2259, "lr": 8.837404133838667e-06, "epoch": 0.89664, "percentage": 29.94, "elapsed_time": "1:09:48", "remaining_time": "2:43:23"}
+{"current_steps": 468, "total_steps": 1560, "loss": 0.2354, "lr": 8.83022221559489e-06, "epoch": 0.89856, "percentage": 30.0, "elapsed_time": "1:09:57", "remaining_time": "2:43:14"}
+{"current_steps": 469, "total_steps": 1560, "loss": 0.2746, "lr": 8.823021119969102e-06, "epoch": 0.90048, "percentage": 30.06, "elapsed_time": "1:10:06", "remaining_time": "2:43:05"}
+{"current_steps": 470, "total_steps": 1560, "loss": 0.2464, "lr": 8.815800883016168e-06, "epoch": 0.9024, "percentage": 30.13, "elapsed_time": "1:10:15", "remaining_time": "2:42:56"}
+{"current_steps": 471, "total_steps": 1560, "loss": 0.3011, "lr": 8.808561540886796e-06, "epoch": 0.90432, "percentage": 30.19, "elapsed_time": "1:10:24", "remaining_time": "2:42:47"}
+{"current_steps": 472, "total_steps": 1560, "loss": 0.227, "lr": 8.801303129827352e-06, "epoch": 0.90624, "percentage": 30.26, "elapsed_time": "1:10:33", "remaining_time": "2:42:38"}
+{"current_steps": 473, "total_steps": 1560, "loss": 0.2256, "lr": 8.794025686179677e-06, "epoch": 0.90816, "percentage": 30.32, "elapsed_time": "1:10:42", "remaining_time": "2:42:29"}
+{"current_steps": 474, "total_steps": 1560, "loss": 0.2225, "lr": 8.786729246380901e-06, "epoch": 0.91008, "percentage": 30.38, "elapsed_time": "1:10:51", "remaining_time": "2:42:20"}
+{"current_steps": 475, "total_steps": 1560, "loss": 0.2245, "lr": 8.779413846963267e-06, "epoch": 0.912, "percentage": 30.45, "elapsed_time": "1:11:00", "remaining_time": "2:42:11"}
+{"current_steps": 476, "total_steps": 1560, "loss": 0.2016, "lr": 8.772079524553951e-06, "epoch": 0.91392, "percentage": 30.51, "elapsed_time": "1:11:09", "remaining_time": "2:42:02"}
+{"current_steps": 477, "total_steps": 1560, "loss": 0.2695, "lr": 8.764726315874872e-06, "epoch": 0.91584, "percentage": 30.58, "elapsed_time": "1:11:18", "remaining_time": "2:41:53"}
+{"current_steps": 478, "total_steps": 1560, "loss": 0.244, "lr": 8.757354257742501e-06, "epoch": 0.91776, "percentage": 30.64, "elapsed_time": "1:11:27", "remaining_time": "2:41:44"}
+{"current_steps": 479, "total_steps": 1560, "loss": 0.2717, "lr": 8.749963387067697e-06, "epoch": 0.91968, "percentage": 30.71, "elapsed_time": "1:11:36", "remaining_time": "2:41:35"}
+{"current_steps": 480, "total_steps": 1560, "loss": 0.2505, "lr": 8.742553740855507e-06, "epoch": 0.9216, "percentage": 30.77, "elapsed_time": "1:11:45", "remaining_time": "2:41:26"}
+{"current_steps": 481, "total_steps": 1560, "loss": 0.234, "lr": 8.735125356204982e-06, "epoch": 0.92352, "percentage": 30.83, "elapsed_time": "1:11:54", "remaining_time": "2:41:17"}
+{"current_steps": 482, "total_steps": 1560, "loss": 0.3257, "lr": 8.727678270308994e-06, "epoch": 0.92544, "percentage": 30.9, "elapsed_time": "1:12:03", "remaining_time": "2:41:08"}
+{"current_steps": 483, "total_steps": 1560, "loss": 0.2591, "lr": 8.720212520454054e-06, "epoch": 0.92736, "percentage": 30.96, "elapsed_time": "1:12:11", "remaining_time": "2:40:59"}
+{"current_steps": 484, "total_steps": 1560, "loss": 0.2401, "lr": 8.712728144020118e-06, "epoch": 0.92928, "percentage": 31.03, "elapsed_time": "1:12:20", "remaining_time": "2:40:50"}
+{"current_steps": 485, "total_steps": 1560, "loss": 0.2422, "lr": 8.705225178480397e-06, "epoch": 0.9312, "percentage": 31.09, "elapsed_time": "1:12:29", "remaining_time": "2:40:41"}
+{"current_steps": 486, "total_steps": 1560, "loss": 0.2216, "lr": 8.697703661401187e-06, "epoch": 0.93312, "percentage": 31.15, "elapsed_time": "1:12:38", "remaining_time": "2:40:32"}
+{"current_steps": 487, "total_steps": 1560, "loss": 0.2647, "lr": 8.69016363044166e-06, "epoch": 0.93504, "percentage": 31.22, "elapsed_time": "1:12:47", "remaining_time": "2:40:23"}
+{"current_steps": 488, "total_steps": 1560, "loss": 0.2257, "lr": 8.682605123353685e-06, "epoch": 0.93696, "percentage": 31.28, "elapsed_time": "1:12:56", "remaining_time": "2:40:14"}
+{"current_steps": 489, "total_steps": 1560, "loss": 0.2175, "lr": 8.675028177981643e-06, "epoch": 0.93888, "percentage": 31.35, "elapsed_time": "1:13:05", "remaining_time": "2:40:05"}
+{"current_steps": 490, "total_steps": 1560, "loss": 0.225, "lr": 8.66743283226223e-06, "epoch": 0.9408, "percentage": 31.41, "elapsed_time": "1:13:14", "remaining_time": "2:39:56"}
+{"current_steps": 491, "total_steps": 1560, "loss": 0.2179, "lr": 8.659819124224266e-06, "epoch": 0.94272, "percentage": 31.47, "elapsed_time": "1:13:23", "remaining_time": "2:39:47"}
+{"current_steps": 492, "total_steps": 1560, "loss": 0.2556, "lr": 8.652187091988516e-06, "epoch": 0.94464, "percentage": 31.54, "elapsed_time": "1:13:32", "remaining_time": "2:39:38"}
+{"current_steps": 493, "total_steps": 1560, "loss": 0.2508, "lr": 8.644536773767488e-06, "epoch": 0.94656, "percentage": 31.6, "elapsed_time": "1:13:41", "remaining_time": "2:39:29"}
+{"current_steps": 494, "total_steps": 1560, "loss": 0.2241, "lr": 8.636868207865244e-06, "epoch": 0.94848, "percentage": 31.67, "elapsed_time": "1:13:50", "remaining_time": "2:39:20"}
+{"current_steps": 495, "total_steps": 1560, "loss": 0.2353, "lr": 8.629181432677213e-06, "epoch": 0.9504, "percentage": 31.73, "elapsed_time": "1:13:59", "remaining_time": "2:39:11"}
+{"current_steps": 496, "total_steps": 1560, "loss": 0.2552, "lr": 8.621476486689991e-06, "epoch": 0.95232, "percentage": 31.79, "elapsed_time": "1:14:08", "remaining_time": "2:39:02"}
+{"current_steps": 497, "total_steps": 1560, "loss": 0.3039, "lr": 8.613753408481158e-06, "epoch": 0.95424, "percentage": 31.86, "elapsed_time": "1:14:17", "remaining_time": "2:38:53"}
+{"current_steps": 498, "total_steps": 1560, "loss": 0.2456, "lr": 8.606012236719073e-06, "epoch": 0.95616, "percentage": 31.92, "elapsed_time": "1:14:26", "remaining_time": "2:38:44"}
+{"current_steps": 499, "total_steps": 1560, "loss": 0.2224, "lr": 8.598253010162693e-06, "epoch": 0.95808, "percentage": 31.99, "elapsed_time": "1:14:35", "remaining_time": "2:38:35"}
+{"current_steps": 500, "total_steps": 1560, "loss": 0.2396, "lr": 8.590475767661371e-06, "epoch": 0.96, "percentage": 32.05, "elapsed_time": "1:14:44", "remaining_time": "2:38:26"}
+{"current_steps": 501, "total_steps": 1560, "loss": 0.2371, "lr": 8.58268054815466e-06, "epoch": 0.96192, "percentage": 32.12, "elapsed_time": "1:14:53", "remaining_time": "2:38:17"}
+{"current_steps": 502, "total_steps": 1560, "loss": 0.253, "lr": 8.574867390672124e-06, "epoch": 0.96384, "percentage": 32.18, "elapsed_time": "1:15:02", "remaining_time": "2:38:09"}
+{"current_steps": 503, "total_steps": 1560, "loss": 0.3291, "lr": 8.567036334333142e-06, "epoch": 0.96576, "percentage": 32.24, "elapsed_time": "1:15:12", "remaining_time": "2:38:01"}
+{"current_steps": 504, "total_steps": 1560, "loss": 0.2468, "lr": 8.559187418346703e-06, "epoch": 0.96768, "percentage": 32.31, "elapsed_time": "1:15:21", "remaining_time": "2:37:53"}
+{"current_steps": 505, "total_steps": 1560, "loss": 0.2499, "lr": 8.551320682011227e-06, "epoch": 0.9696, "percentage": 32.37, "elapsed_time": "1:15:30", "remaining_time": "2:37:45"}
+{"current_steps": 506, "total_steps": 1560, "loss": 0.2367, "lr": 8.543436164714351e-06, "epoch": 0.97152, "percentage": 32.44, "elapsed_time": "1:15:40", "remaining_time": "2:37:36"}
+{"current_steps": 507, "total_steps": 1560, "loss": 0.2502, "lr": 8.535533905932739e-06, "epoch": 0.97344, "percentage": 32.5, "elapsed_time": "1:15:49", "remaining_time": "2:37:28"}
+{"current_steps": 508, "total_steps": 1560, "loss": 0.2552, "lr": 8.527613945231886e-06, "epoch": 0.97536, "percentage": 32.56, "elapsed_time": "1:15:58", "remaining_time": "2:37:20"}
+{"current_steps": 509, "total_steps": 1560, "loss": 0.2094, "lr": 8.519676322265914e-06, "epoch": 0.97728, "percentage": 32.63, "elapsed_time": "1:16:08", "remaining_time": "2:37:12"}
+{"current_steps": 510, "total_steps": 1560, "loss": 0.2023, "lr": 8.511721076777388e-06, "epoch": 0.9792, "percentage": 32.69, "elapsed_time": "1:16:17", "remaining_time": "2:37:03"}
+{"current_steps": 511, "total_steps": 1560, "loss": 0.2355, "lr": 8.503748248597095e-06, "epoch": 0.98112, "percentage": 32.76, "elapsed_time": "1:16:26", "remaining_time": "2:36:54"}
+{"current_steps": 512, "total_steps": 1560, "loss": 0.2603, "lr": 8.495757877643857e-06, "epoch": 0.98304, "percentage": 32.82, "elapsed_time": "1:16:35", "remaining_time": "2:36:46"}
+{"current_steps": 513, "total_steps": 1560, "loss": 0.2243, "lr": 8.487750003924341e-06, "epoch": 0.98496, "percentage": 32.88, "elapsed_time": "1:16:44", "remaining_time": "2:36:37"}
+{"current_steps": 514, "total_steps": 1560, "loss": 0.2454, "lr": 8.479724667532836e-06, "epoch": 0.98688, "percentage": 32.95, "elapsed_time": "1:16:53", "remaining_time": "2:36:28"}
+{"current_steps": 515, "total_steps": 1560, "loss": 0.2731, "lr": 8.471681908651067e-06, "epoch": 0.9888, "percentage": 33.01, "elapsed_time": "1:17:02", "remaining_time": "2:36:19"}
+{"current_steps": 516, "total_steps": 1560, "loss": 0.2466, "lr": 8.463621767547998e-06, "epoch": 0.99072, "percentage": 33.08, "elapsed_time": "1:17:11", "remaining_time": "2:36:10"}
+{"current_steps": 517, "total_steps": 1560, "loss": 0.2964, "lr": 8.455544284579614e-06, "epoch": 0.99264, "percentage": 33.14, "elapsed_time": "1:17:20", "remaining_time": "2:36:00"}
+{"current_steps": 518, "total_steps": 1560, "loss": 0.2284, "lr": 8.447449500188731e-06, "epoch": 0.99456, "percentage": 33.21, "elapsed_time": "1:17:28", "remaining_time": "2:35:51"}
+{"current_steps": 519, "total_steps": 1560, "loss": 0.2533, "lr": 8.439337454904794e-06, "epoch": 0.99648, "percentage": 33.27, "elapsed_time": "1:17:37", "remaining_time": "2:35:42"}
+{"current_steps": 520, "total_steps": 1560, "loss": 0.2624, "lr": 8.43120818934367e-06, "epoch": 0.9984, "percentage": 33.33, "elapsed_time": "1:17:46", "remaining_time": "2:35:33"}
+{"current_steps": 521, "total_steps": 1560, "loss": 0.3766, "lr": 8.42306174420744e-06, "epoch": 1.00128, "percentage": 33.4, "elapsed_time": "1:18:46", "remaining_time": "2:37:06"}
+{"current_steps": 522, "total_steps": 1560, "loss": 0.2299, "lr": 8.414898160284208e-06, "epoch": 1.0032, "percentage": 33.46, "elapsed_time": "1:18:57", "remaining_time": "2:36:59"}
+{"current_steps": 523, "total_steps": 1560, "loss": 0.2069, "lr": 8.406717478447889e-06, "epoch": 1.00512, "percentage": 33.53, "elapsed_time": "1:19:07", "remaining_time": "2:36:52"}
+{"current_steps": 524, "total_steps": 1560, "loss": 0.2519, "lr": 8.398519739657997e-06, "epoch": 1.00704, "percentage": 33.59, "elapsed_time": "1:19:16", "remaining_time": "2:36:44"}
+{"current_steps": 525, "total_steps": 1560, "loss": 0.1837, "lr": 8.390304984959455e-06, "epoch": 1.00896, "percentage": 33.65, "elapsed_time": "1:19:26", "remaining_time": "2:36:36"}
+{"current_steps": 526, "total_steps": 1560, "loss": 0.2254, "lr": 8.382073255482381e-06, "epoch": 1.01088, "percentage": 33.72, "elapsed_time": "1:19:35", "remaining_time": "2:36:28"}
+{"current_steps": 527, "total_steps": 1560, "loss": 0.1789, "lr": 8.373824592441884e-06, "epoch": 1.0128, "percentage": 33.78, "elapsed_time": "1:19:45", "remaining_time": "2:36:19"}
+{"current_steps": 528, "total_steps": 1560, "loss": 0.1988, "lr": 8.36555903713785e-06, "epoch": 1.01472, "percentage": 33.85, "elapsed_time": "1:19:54", "remaining_time": "2:36:11"}
+{"current_steps": 529, "total_steps": 1560, "loss": 0.2206, "lr": 8.357276630954756e-06, "epoch": 1.01664, "percentage": 33.91, "elapsed_time": "1:20:03", "remaining_time": "2:36:02"}
+{"current_steps": 530, "total_steps": 1560, "loss": 0.2044, "lr": 8.348977415361435e-06, "epoch": 1.01856, "percentage": 33.97, "elapsed_time": "1:20:13", "remaining_time": "2:35:54"}
+{"current_steps": 531, "total_steps": 1560, "loss": 0.1792, "lr": 8.340661431910888e-06, "epoch": 1.02048, "percentage": 34.04, "elapsed_time": "1:20:22", "remaining_time": "2:35:46"}
+{"current_steps": 532, "total_steps": 1560, "loss": 0.2211, "lr": 8.332328722240072e-06, "epoch": 1.0224, "percentage": 34.1, "elapsed_time": "1:20:32", "remaining_time": "2:35:37"}
+{"current_steps": 533, "total_steps": 1560, "loss": 0.2055, "lr": 8.323979328069689e-06, "epoch": 1.02432, "percentage": 34.17, "elapsed_time": "1:20:41", "remaining_time": "2:35:29"}
+{"current_steps": 534, "total_steps": 1560, "loss": 0.1872, "lr": 8.315613291203977e-06, "epoch": 1.02624, "percentage": 34.23, "elapsed_time": "1:20:51", "remaining_time": "2:35:20"}
+{"current_steps": 535, "total_steps": 1560, "loss": 0.2539, "lr": 8.307230653530501e-06, "epoch": 1.02816, "percentage": 34.29, "elapsed_time": "1:21:00", "remaining_time": "2:35:11"}
+{"current_steps": 536, "total_steps": 1560, "loss": 0.279, "lr": 8.298831457019943e-06, "epoch": 1.03008, "percentage": 34.36, "elapsed_time": "1:21:09", "remaining_time": "2:35:02"}
+{"current_steps": 537, "total_steps": 1560, "loss": 0.198, "lr": 8.290415743725894e-06, "epoch": 1.032, "percentage": 34.42, "elapsed_time": "1:21:18", "remaining_time": "2:34:53"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:074bf7facf12d6d05d2b46fc171e3693e7378b5d8fa085869ed8405d5096ac39
+size 7416