Training in progress, step 400

.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 +1,28 @@
+{
+  "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": 131072,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "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 sha256:24aad0340781e9c3aa96eebbab32c29e06efdfbbaff08aa0ce2d65cecc1b89e3
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:111b2ac7c4df6d7c18aa3086a0345dbf51faeef0b6ff1d8fecb0910a9d584931
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99407b4f7786931fc3d68a05341e953c9be18d6046ef963a38c4cd8d162b045b
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6da83eab2988953b21378b27d4a5f083181da4ab7889752c0d036328d31b92d0
+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": "<|im_end|>",
+    "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 +1,209 @@
+{
+  "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": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "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 +1,415 @@
+{"current_steps": 1, "total_steps": 2148, "loss": 1.2425, "lr": 0.0, "epoch": 0.0013950244129272262, "percentage": 0.05, "elapsed_time": "0:00:10", "remaining_time": "6:23:32"}
+{"current_steps": 2, "total_steps": 2148, "loss": 1.1023, "lr": 4.6511627906976744e-08, "epoch": 0.0027900488258544524, "percentage": 0.09, "elapsed_time": "0:00:23", "remaining_time": "7:00:50"}
+{"current_steps": 3, "total_steps": 2148, "loss": 0.9672, "lr": 9.302325581395349e-08, "epoch": 0.0041850732387816785, "percentage": 0.14, "elapsed_time": "0:00:30", "remaining_time": "6:03:04"}
+{"current_steps": 4, "total_steps": 2148, "loss": 1.1729, "lr": 1.3953488372093024e-07, "epoch": 0.005580097651708905, "percentage": 0.19, "elapsed_time": "0:00:36", "remaining_time": "5:26:55"}
+{"current_steps": 5, "total_steps": 2148, "loss": 1.0899, "lr": 1.8604651162790698e-07, "epoch": 0.006975122064636131, "percentage": 0.23, "elapsed_time": "0:00:42", "remaining_time": "5:05:34"}
+{"current_steps": 6, "total_steps": 2148, "loss": 1.1316, "lr": 2.3255813953488374e-07, "epoch": 0.008370146477563357, "percentage": 0.28, "elapsed_time": "0:00:49", "remaining_time": "4:55:07"}
+{"current_steps": 7, "total_steps": 2148, "loss": 1.218, "lr": 2.790697674418605e-07, "epoch": 0.009765170890490584, "percentage": 0.33, "elapsed_time": "0:00:57", "remaining_time": "4:51:38"}
+{"current_steps": 8, "total_steps": 2148, "loss": 1.2305, "lr": 3.2558139534883724e-07, "epoch": 0.01116019530341781, "percentage": 0.37, "elapsed_time": "0:01:04", "remaining_time": "4:49:00"}
+{"current_steps": 9, "total_steps": 2148, "loss": 1.0201, "lr": 3.7209302325581396e-07, "epoch": 0.012555219716345037, "percentage": 0.42, "elapsed_time": "0:01:11", "remaining_time": "4:42:58"}
+{"current_steps": 10, "total_steps": 2148, "loss": 1.0511, "lr": 4.186046511627907e-07, "epoch": 0.013950244129272262, "percentage": 0.47, "elapsed_time": "0:01:18", "remaining_time": "4:38:49"}
+{"current_steps": 11, "total_steps": 2148, "loss": 1.1006, "lr": 4.651162790697675e-07, "epoch": 0.015345268542199489, "percentage": 0.51, "elapsed_time": "0:01:24", "remaining_time": "4:33:22"}
+{"current_steps": 12, "total_steps": 2148, "loss": 1.0077, "lr": 5.116279069767442e-07, "epoch": 0.016740292955126714, "percentage": 0.56, "elapsed_time": "0:01:35", "remaining_time": "4:42:07"}
+{"current_steps": 13, "total_steps": 2148, "loss": 1.1322, "lr": 5.58139534883721e-07, "epoch": 0.01813531736805394, "percentage": 0.61, "elapsed_time": "0:01:41", "remaining_time": "4:37:10"}
+{"current_steps": 14, "total_steps": 2148, "loss": 1.185, "lr": 6.046511627906977e-07, "epoch": 0.01953034178098117, "percentage": 0.65, "elapsed_time": "0:01:47", "remaining_time": "4:33:56"}
+{"current_steps": 15, "total_steps": 2148, "loss": 0.9731, "lr": 6.511627906976745e-07, "epoch": 0.020925366193908394, "percentage": 0.7, "elapsed_time": "0:01:54", "remaining_time": "4:31:54"}
+{"current_steps": 16, "total_steps": 2148, "loss": 1.0501, "lr": 6.976744186046513e-07, "epoch": 0.02232039060683562, "percentage": 0.74, "elapsed_time": "0:02:01", "remaining_time": "4:30:03"}
+{"current_steps": 17, "total_steps": 2148, "loss": 0.8553, "lr": 7.441860465116279e-07, "epoch": 0.023715415019762844, "percentage": 0.79, "elapsed_time": "0:02:08", "remaining_time": "4:29:03"}
+{"current_steps": 18, "total_steps": 2148, "loss": 1.1987, "lr": 7.906976744186047e-07, "epoch": 0.025110439432690073, "percentage": 0.84, "elapsed_time": "0:02:15", "remaining_time": "4:27:56"}
+{"current_steps": 19, "total_steps": 2148, "loss": 0.9201, "lr": 8.372093023255814e-07, "epoch": 0.0265054638456173, "percentage": 0.88, "elapsed_time": "0:02:22", "remaining_time": "4:27:01"}
+{"current_steps": 20, "total_steps": 2148, "loss": 1.044, "lr": 8.837209302325582e-07, "epoch": 0.027900488258544524, "percentage": 0.93, "elapsed_time": "0:02:30", "remaining_time": "4:27:17"}
+{"current_steps": 21, "total_steps": 2148, "loss": 0.9848, "lr": 9.30232558139535e-07, "epoch": 0.029295512671471752, "percentage": 0.98, "elapsed_time": "0:02:39", "remaining_time": "4:29:30"}
+{"current_steps": 22, "total_steps": 2148, "loss": 1.0273, "lr": 9.767441860465117e-07, "epoch": 0.030690537084398978, "percentage": 1.02, "elapsed_time": "0:02:45", "remaining_time": "4:27:17"}
+{"current_steps": 23, "total_steps": 2148, "loss": 1.0212, "lr": 1.0232558139534884e-06, "epoch": 0.03208556149732621, "percentage": 1.07, "elapsed_time": "0:02:54", "remaining_time": "4:29:05"}
+{"current_steps": 24, "total_steps": 2148, "loss": 0.8341, "lr": 1.0697674418604653e-06, "epoch": 0.03348058591025343, "percentage": 1.12, "elapsed_time": "0:03:05", "remaining_time": "4:33:31"}
+{"current_steps": 25, "total_steps": 2148, "loss": 0.9715, "lr": 1.116279069767442e-06, "epoch": 0.03487561032318066, "percentage": 1.16, "elapsed_time": "0:03:11", "remaining_time": "4:31:27"}
+{"current_steps": 26, "total_steps": 2148, "loss": 0.9561, "lr": 1.1627906976744188e-06, "epoch": 0.03627063473610788, "percentage": 1.21, "elapsed_time": "0:03:17", "remaining_time": "4:29:16"}
+{"current_steps": 27, "total_steps": 2148, "loss": 1.0269, "lr": 1.2093023255813954e-06, "epoch": 0.03766565914903511, "percentage": 1.26, "elapsed_time": "0:03:25", "remaining_time": "4:28:42"}
+{"current_steps": 28, "total_steps": 2148, "loss": 0.9129, "lr": 1.2558139534883723e-06, "epoch": 0.03906068356196234, "percentage": 1.3, "elapsed_time": "0:03:39", "remaining_time": "4:36:38"}
+{"current_steps": 29, "total_steps": 2148, "loss": 0.889, "lr": 1.302325581395349e-06, "epoch": 0.04045570797488956, "percentage": 1.35, "elapsed_time": "0:03:49", "remaining_time": "4:39:00"}
+{"current_steps": 30, "total_steps": 2148, "loss": 0.8861, "lr": 1.3488372093023258e-06, "epoch": 0.04185073238781679, "percentage": 1.4, "elapsed_time": "0:03:56", "remaining_time": "4:37:47"}
+{"current_steps": 31, "total_steps": 2148, "loss": 0.8176, "lr": 1.3953488372093025e-06, "epoch": 0.043245756800744016, "percentage": 1.44, "elapsed_time": "0:04:07", "remaining_time": "4:41:40"}
+{"current_steps": 32, "total_steps": 2148, "loss": 0.8987, "lr": 1.4418604651162794e-06, "epoch": 0.04464078121367124, "percentage": 1.49, "elapsed_time": "0:04:13", "remaining_time": "4:39:20"}
+{"current_steps": 33, "total_steps": 2148, "loss": 0.9591, "lr": 1.4883720930232558e-06, "epoch": 0.04603580562659847, "percentage": 1.54, "elapsed_time": "0:04:21", "remaining_time": "4:38:49"}
+{"current_steps": 34, "total_steps": 2148, "loss": 0.9171, "lr": 1.534883720930233e-06, "epoch": 0.04743083003952569, "percentage": 1.58, "elapsed_time": "0:04:28", "remaining_time": "4:38:20"}
+{"current_steps": 35, "total_steps": 2148, "loss": 0.9138, "lr": 1.5813953488372093e-06, "epoch": 0.04882585445245292, "percentage": 1.63, "elapsed_time": "0:04:38", "remaining_time": "4:40:22"}
+{"current_steps": 36, "total_steps": 2148, "loss": 0.9461, "lr": 1.6279069767441862e-06, "epoch": 0.050220878865380146, "percentage": 1.68, "elapsed_time": "0:04:48", "remaining_time": "4:42:05"}
+{"current_steps": 37, "total_steps": 2148, "loss": 0.8543, "lr": 1.6744186046511629e-06, "epoch": 0.05161590327830737, "percentage": 1.72, "elapsed_time": "0:04:54", "remaining_time": "4:40:25"}
+{"current_steps": 38, "total_steps": 2148, "loss": 0.7944, "lr": 1.7209302325581397e-06, "epoch": 0.0530109276912346, "percentage": 1.77, "elapsed_time": "0:05:04", "remaining_time": "4:41:48"}
+{"current_steps": 39, "total_steps": 2148, "loss": 0.9039, "lr": 1.7674418604651164e-06, "epoch": 0.054405952104161825, "percentage": 1.82, "elapsed_time": "0:05:14", "remaining_time": "4:43:08"}
+{"current_steps": 40, "total_steps": 2148, "loss": 0.8522, "lr": 1.8139534883720933e-06, "epoch": 0.05580097651708905, "percentage": 1.86, "elapsed_time": "0:05:21", "remaining_time": "4:42:02"}
+{"current_steps": 41, "total_steps": 2148, "loss": 0.8171, "lr": 1.86046511627907e-06, "epoch": 0.057196000930016276, "percentage": 1.91, "elapsed_time": "0:05:28", "remaining_time": "4:41:40"}
+{"current_steps": 42, "total_steps": 2148, "loss": 0.8097, "lr": 1.9069767441860468e-06, "epoch": 0.058591025342943505, "percentage": 1.96, "elapsed_time": "0:05:36", "remaining_time": "4:41:01"}
+{"current_steps": 43, "total_steps": 2148, "loss": 0.7917, "lr": 1.9534883720930235e-06, "epoch": 0.05998604975587073, "percentage": 2.0, "elapsed_time": "0:05:46", "remaining_time": "4:42:44"}
+{"current_steps": 44, "total_steps": 2148, "loss": 0.8263, "lr": 2.0000000000000003e-06, "epoch": 0.061381074168797956, "percentage": 2.05, "elapsed_time": "0:05:53", "remaining_time": "4:41:30"}
+{"current_steps": 45, "total_steps": 2148, "loss": 0.8794, "lr": 2.0465116279069768e-06, "epoch": 0.06277609858172518, "percentage": 2.09, "elapsed_time": "0:05:59", "remaining_time": "4:39:59"}
+{"current_steps": 46, "total_steps": 2148, "loss": 0.7589, "lr": 2.0930232558139536e-06, "epoch": 0.06417112299465241, "percentage": 2.14, "elapsed_time": "0:06:06", "remaining_time": "4:39:25"}
+{"current_steps": 47, "total_steps": 2148, "loss": 0.8537, "lr": 2.1395348837209305e-06, "epoch": 0.06556614740757963, "percentage": 2.19, "elapsed_time": "0:06:13", "remaining_time": "4:38:34"}
+{"current_steps": 48, "total_steps": 2148, "loss": 0.8453, "lr": 2.1860465116279074e-06, "epoch": 0.06696117182050686, "percentage": 2.23, "elapsed_time": "0:06:20", "remaining_time": "4:37:24"}
+{"current_steps": 49, "total_steps": 2148, "loss": 0.8777, "lr": 2.232558139534884e-06, "epoch": 0.06835619623343409, "percentage": 2.28, "elapsed_time": "0:06:29", "remaining_time": "4:37:48"}
+{"current_steps": 50, "total_steps": 2148, "loss": 0.8138, "lr": 2.2790697674418607e-06, "epoch": 0.06975122064636131, "percentage": 2.33, "elapsed_time": "0:06:36", "remaining_time": "4:36:59"}
+{"current_steps": 51, "total_steps": 2148, "loss": 0.8686, "lr": 2.3255813953488376e-06, "epoch": 0.07114624505928854, "percentage": 2.37, "elapsed_time": "0:06:45", "remaining_time": "4:37:42"}
+{"current_steps": 52, "total_steps": 2148, "loss": 0.8089, "lr": 2.3720930232558144e-06, "epoch": 0.07254126947221576, "percentage": 2.42, "elapsed_time": "0:06:52", "remaining_time": "4:37:04"}
+{"current_steps": 53, "total_steps": 2148, "loss": 0.8212, "lr": 2.418604651162791e-06, "epoch": 0.07393629388514299, "percentage": 2.47, "elapsed_time": "0:06:58", "remaining_time": "4:35:58"}
+{"current_steps": 54, "total_steps": 2148, "loss": 0.8565, "lr": 2.4651162790697678e-06, "epoch": 0.07533131829807022, "percentage": 2.51, "elapsed_time": "0:07:07", "remaining_time": "4:36:36"}
+{"current_steps": 55, "total_steps": 2148, "loss": 0.7982, "lr": 2.5116279069767446e-06, "epoch": 0.07672634271099744, "percentage": 2.56, "elapsed_time": "0:07:17", "remaining_time": "4:37:39"}
+{"current_steps": 56, "total_steps": 2148, "loss": 0.8422, "lr": 2.558139534883721e-06, "epoch": 0.07812136712392467, "percentage": 2.61, "elapsed_time": "0:07:25", "remaining_time": "4:37:09"}
+{"current_steps": 57, "total_steps": 2148, "loss": 0.8189, "lr": 2.604651162790698e-06, "epoch": 0.07951639153685189, "percentage": 2.65, "elapsed_time": "0:07:31", "remaining_time": "4:36:00"}
+{"current_steps": 58, "total_steps": 2148, "loss": 0.7868, "lr": 2.6511627906976744e-06, "epoch": 0.08091141594977912, "percentage": 2.7, "elapsed_time": "0:07:38", "remaining_time": "4:35:06"}
+{"current_steps": 59, "total_steps": 2148, "loss": 0.8908, "lr": 2.6976744186046517e-06, "epoch": 0.08230644036270635, "percentage": 2.75, "elapsed_time": "0:07:48", "remaining_time": "4:36:36"}
+{"current_steps": 60, "total_steps": 2148, "loss": 0.7486, "lr": 2.744186046511628e-06, "epoch": 0.08370146477563357, "percentage": 2.79, "elapsed_time": "0:07:55", "remaining_time": "4:35:52"}
+{"current_steps": 61, "total_steps": 2148, "loss": 0.7484, "lr": 2.790697674418605e-06, "epoch": 0.0850964891885608, "percentage": 2.84, "elapsed_time": "0:08:04", "remaining_time": "4:36:11"}
+{"current_steps": 62, "total_steps": 2148, "loss": 0.8414, "lr": 2.8372093023255815e-06, "epoch": 0.08649151360148803, "percentage": 2.89, "elapsed_time": "0:08:13", "remaining_time": "4:36:32"}
+{"current_steps": 63, "total_steps": 2148, "loss": 0.7389, "lr": 2.8837209302325587e-06, "epoch": 0.08788653801441525, "percentage": 2.93, "elapsed_time": "0:08:21", "remaining_time": "4:36:22"}
+{"current_steps": 64, "total_steps": 2148, "loss": 0.8335, "lr": 2.930232558139535e-06, "epoch": 0.08928156242734248, "percentage": 2.98, "elapsed_time": "0:08:32", "remaining_time": "4:38:05"}
+{"current_steps": 65, "total_steps": 2148, "loss": 0.7524, "lr": 2.9767441860465116e-06, "epoch": 0.0906765868402697, "percentage": 3.03, "elapsed_time": "0:08:41", "remaining_time": "4:38:33"}
+{"current_steps": 66, "total_steps": 2148, "loss": 0.8056, "lr": 3.0232558139534885e-06, "epoch": 0.09207161125319693, "percentage": 3.07, "elapsed_time": "0:08:51", "remaining_time": "4:39:31"}
+{"current_steps": 67, "total_steps": 2148, "loss": 0.8627, "lr": 3.069767441860466e-06, "epoch": 0.09346663566612416, "percentage": 3.12, "elapsed_time": "0:09:02", "remaining_time": "4:40:40"}
+{"current_steps": 68, "total_steps": 2148, "loss": 0.8127, "lr": 3.1162790697674423e-06, "epoch": 0.09486166007905138, "percentage": 3.17, "elapsed_time": "0:09:09", "remaining_time": "4:40:00"}
+{"current_steps": 69, "total_steps": 2148, "loss": 0.757, "lr": 3.1627906976744187e-06, "epoch": 0.0962566844919786, "percentage": 3.21, "elapsed_time": "0:09:17", "remaining_time": "4:39:56"}
+{"current_steps": 70, "total_steps": 2148, "loss": 0.722, "lr": 3.2093023255813956e-06, "epoch": 0.09765170890490583, "percentage": 3.26, "elapsed_time": "0:09:25", "remaining_time": "4:39:42"}
+{"current_steps": 71, "total_steps": 2148, "loss": 0.7583, "lr": 3.2558139534883724e-06, "epoch": 0.09904673331783306, "percentage": 3.31, "elapsed_time": "0:09:35", "remaining_time": "4:40:32"}
+{"current_steps": 72, "total_steps": 2148, "loss": 0.7642, "lr": 3.3023255813953493e-06, "epoch": 0.10044175773076029, "percentage": 3.35, "elapsed_time": "0:09:41", "remaining_time": "4:39:27"}
+{"current_steps": 73, "total_steps": 2148, "loss": 0.7746, "lr": 3.3488372093023258e-06, "epoch": 0.10183678214368752, "percentage": 3.4, "elapsed_time": "0:09:49", "remaining_time": "4:39:09"}
+{"current_steps": 74, "total_steps": 2148, "loss": 0.7831, "lr": 3.3953488372093026e-06, "epoch": 0.10323180655661474, "percentage": 3.45, "elapsed_time": "0:09:55", "remaining_time": "4:38:05"}
+{"current_steps": 75, "total_steps": 2148, "loss": 0.7607, "lr": 3.4418604651162795e-06, "epoch": 0.10462683096954196, "percentage": 3.49, "elapsed_time": "0:10:05", "remaining_time": "4:38:54"}
+{"current_steps": 76, "total_steps": 2148, "loss": 0.6625, "lr": 3.4883720930232564e-06, "epoch": 0.1060218553824692, "percentage": 3.54, "elapsed_time": "0:10:11", "remaining_time": "4:37:56"}
+{"current_steps": 77, "total_steps": 2148, "loss": 0.6498, "lr": 3.534883720930233e-06, "epoch": 0.10741687979539642, "percentage": 3.58, "elapsed_time": "0:10:18", "remaining_time": "4:37:08"}
+{"current_steps": 78, "total_steps": 2148, "loss": 0.7939, "lr": 3.5813953488372093e-06, "epoch": 0.10881190420832365, "percentage": 3.63, "elapsed_time": "0:10:28", "remaining_time": "4:37:59"}
+{"current_steps": 79, "total_steps": 2148, "loss": 0.763, "lr": 3.6279069767441866e-06, "epoch": 0.11020692862125087, "percentage": 3.68, "elapsed_time": "0:10:35", "remaining_time": "4:37:12"}
+{"current_steps": 80, "total_steps": 2148, "loss": 0.8045, "lr": 3.674418604651163e-06, "epoch": 0.1116019530341781, "percentage": 3.72, "elapsed_time": "0:10:41", "remaining_time": "4:36:27"}
+{"current_steps": 81, "total_steps": 2148, "loss": 0.7621, "lr": 3.72093023255814e-06, "epoch": 0.11299697744710532, "percentage": 3.77, "elapsed_time": "0:10:49", "remaining_time": "4:36:08"}
+{"current_steps": 82, "total_steps": 2148, "loss": 0.8571, "lr": 3.7674418604651163e-06, "epoch": 0.11439200186003255, "percentage": 3.82, "elapsed_time": "0:10:56", "remaining_time": "4:35:44"}
+{"current_steps": 83, "total_steps": 2148, "loss": 0.7575, "lr": 3.8139534883720936e-06, "epoch": 0.11578702627295978, "percentage": 3.86, "elapsed_time": "0:11:03", "remaining_time": "4:35:01"}
+{"current_steps": 84, "total_steps": 2148, "loss": 0.7546, "lr": 3.86046511627907e-06, "epoch": 0.11718205068588701, "percentage": 3.91, "elapsed_time": "0:11:14", "remaining_time": "4:36:10"}
+{"current_steps": 85, "total_steps": 2148, "loss": 0.7393, "lr": 3.906976744186047e-06, "epoch": 0.11857707509881422, "percentage": 3.96, "elapsed_time": "0:11:24", "remaining_time": "4:36:53"}
+{"current_steps": 86, "total_steps": 2148, "loss": 0.6678, "lr": 3.953488372093024e-06, "epoch": 0.11997209951174145, "percentage": 4.0, "elapsed_time": "0:11:31", "remaining_time": "4:36:17"}
+{"current_steps": 87, "total_steps": 2148, "loss": 0.8062, "lr": 4.000000000000001e-06, "epoch": 0.12136712392466868, "percentage": 4.05, "elapsed_time": "0:11:38", "remaining_time": "4:35:43"}
+{"current_steps": 88, "total_steps": 2148, "loss": 0.821, "lr": 4.0465116279069775e-06, "epoch": 0.12276214833759591, "percentage": 4.1, "elapsed_time": "0:11:48", "remaining_time": "4:36:23"}
+{"current_steps": 89, "total_steps": 2148, "loss": 0.8603, "lr": 4.0930232558139536e-06, "epoch": 0.12415717275052314, "percentage": 4.14, "elapsed_time": "0:11:58", "remaining_time": "4:36:59"}
+{"current_steps": 90, "total_steps": 2148, "loss": 0.7435, "lr": 4.1395348837209304e-06, "epoch": 0.12555219716345037, "percentage": 4.19, "elapsed_time": "0:12:05", "remaining_time": "4:36:24"}
+{"current_steps": 91, "total_steps": 2148, "loss": 0.7425, "lr": 4.186046511627907e-06, "epoch": 0.12694722157637758, "percentage": 4.24, "elapsed_time": "0:12:12", "remaining_time": "4:35:47"}
+{"current_steps": 92, "total_steps": 2148, "loss": 0.7943, "lr": 4.232558139534884e-06, "epoch": 0.12834224598930483, "percentage": 4.28, "elapsed_time": "0:12:18", "remaining_time": "4:35:07"}
+{"current_steps": 93, "total_steps": 2148, "loss": 0.7444, "lr": 4.279069767441861e-06, "epoch": 0.12973727040223204, "percentage": 4.33, "elapsed_time": "0:12:25", "remaining_time": "4:34:35"}
+{"current_steps": 94, "total_steps": 2148, "loss": 0.7103, "lr": 4.325581395348837e-06, "epoch": 0.13113229481515926, "percentage": 4.38, "elapsed_time": "0:12:32", "remaining_time": "4:33:59"}
+{"current_steps": 95, "total_steps": 2148, "loss": 0.6561, "lr": 4.372093023255815e-06, "epoch": 0.1325273192280865, "percentage": 4.42, "elapsed_time": "0:12:41", "remaining_time": "4:34:06"}
+{"current_steps": 96, "total_steps": 2148, "loss": 0.7494, "lr": 4.418604651162791e-06, "epoch": 0.1339223436410137, "percentage": 4.47, "elapsed_time": "0:12:49", "remaining_time": "4:34:11"}
+{"current_steps": 97, "total_steps": 2148, "loss": 0.7291, "lr": 4.465116279069768e-06, "epoch": 0.13531736805394096, "percentage": 4.52, "elapsed_time": "0:12:56", "remaining_time": "4:33:32"}
+{"current_steps": 98, "total_steps": 2148, "loss": 0.6947, "lr": 4.5116279069767445e-06, "epoch": 0.13671239246686817, "percentage": 4.56, "elapsed_time": "0:13:04", "remaining_time": "4:33:23"}
+{"current_steps": 99, "total_steps": 2148, "loss": 0.7622, "lr": 4.558139534883721e-06, "epoch": 0.13810741687979539, "percentage": 4.61, "elapsed_time": "0:13:11", "remaining_time": "4:32:59"}
+{"current_steps": 100, "total_steps": 2148, "loss": 0.6875, "lr": 4.604651162790698e-06, "epoch": 0.13950244129272263, "percentage": 4.66, "elapsed_time": "0:13:19", "remaining_time": "4:32:55"}
+{"current_steps": 101, "total_steps": 2148, "loss": 0.7712, "lr": 4.651162790697675e-06, "epoch": 0.14089746570564984, "percentage": 4.7, "elapsed_time": "0:13:27", "remaining_time": "4:32:40"}
+{"current_steps": 102, "total_steps": 2148, "loss": 0.7644, "lr": 4.697674418604651e-06, "epoch": 0.1422924901185771, "percentage": 4.75, "elapsed_time": "0:13:35", "remaining_time": "4:32:33"}
+{"current_steps": 103, "total_steps": 2148, "loss": 0.7202, "lr": 4.744186046511629e-06, "epoch": 0.1436875145315043, "percentage": 4.8, "elapsed_time": "0:13:43", "remaining_time": "4:32:29"}
+{"current_steps": 104, "total_steps": 2148, "loss": 0.7174, "lr": 4.790697674418605e-06, "epoch": 0.14508253894443152, "percentage": 4.84, "elapsed_time": "0:13:54", "remaining_time": "4:33:13"}
+{"current_steps": 105, "total_steps": 2148, "loss": 0.7038, "lr": 4.837209302325582e-06, "epoch": 0.14647756335735876, "percentage": 4.89, "elapsed_time": "0:14:02", "remaining_time": "4:33:22"}
+{"current_steps": 106, "total_steps": 2148, "loss": 0.6988, "lr": 4.883720930232559e-06, "epoch": 0.14787258777028597, "percentage": 4.93, "elapsed_time": "0:14:12", "remaining_time": "4:33:51"}
+{"current_steps": 107, "total_steps": 2148, "loss": 0.7673, "lr": 4.9302325581395355e-06, "epoch": 0.14926761218321322, "percentage": 4.98, "elapsed_time": "0:14:19", "remaining_time": "4:33:15"}
+{"current_steps": 108, "total_steps": 2148, "loss": 0.7376, "lr": 4.976744186046512e-06, "epoch": 0.15066263659614043, "percentage": 5.03, "elapsed_time": "0:14:29", "remaining_time": "4:33:37"}
+{"current_steps": 109, "total_steps": 2148, "loss": 0.9254, "lr": 5.023255813953489e-06, "epoch": 0.15205766100906765, "percentage": 5.07, "elapsed_time": "0:14:36", "remaining_time": "4:33:11"}
+{"current_steps": 110, "total_steps": 2148, "loss": 0.6927, "lr": 5.069767441860466e-06, "epoch": 0.1534526854219949, "percentage": 5.12, "elapsed_time": "0:14:44", "remaining_time": "4:33:00"}
+{"current_steps": 111, "total_steps": 2148, "loss": 0.7171, "lr": 5.116279069767442e-06, "epoch": 0.1548477098349221, "percentage": 5.17, "elapsed_time": "0:14:50", "remaining_time": "4:32:27"}
+{"current_steps": 112, "total_steps": 2148, "loss": 0.7676, "lr": 5.162790697674419e-06, "epoch": 0.15624273424784935, "percentage": 5.21, "elapsed_time": "0:14:58", "remaining_time": "4:32:13"}
+{"current_steps": 113, "total_steps": 2148, "loss": 0.7407, "lr": 5.209302325581396e-06, "epoch": 0.15763775866077656, "percentage": 5.26, "elapsed_time": "0:15:08", "remaining_time": "4:32:32"}
+{"current_steps": 114, "total_steps": 2148, "loss": 0.7664, "lr": 5.255813953488372e-06, "epoch": 0.15903278307370378, "percentage": 5.31, "elapsed_time": "0:15:17", "remaining_time": "4:32:47"}
+{"current_steps": 115, "total_steps": 2148, "loss": 0.751, "lr": 5.302325581395349e-06, "epoch": 0.16042780748663102, "percentage": 5.35, "elapsed_time": "0:15:23", "remaining_time": "4:32:09"}
+{"current_steps": 116, "total_steps": 2148, "loss": 0.7237, "lr": 5.348837209302326e-06, "epoch": 0.16182283189955823, "percentage": 5.4, "elapsed_time": "0:15:33", "remaining_time": "4:32:29"}
+{"current_steps": 117, "total_steps": 2148, "loss": 0.7932, "lr": 5.395348837209303e-06, "epoch": 0.16321785631248548, "percentage": 5.45, "elapsed_time": "0:15:43", "remaining_time": "4:32:53"}
+{"current_steps": 118, "total_steps": 2148, "loss": 0.7457, "lr": 5.44186046511628e-06, "epoch": 0.1646128807254127, "percentage": 5.49, "elapsed_time": "0:15:53", "remaining_time": "4:33:22"}
+{"current_steps": 119, "total_steps": 2148, "loss": 0.7163, "lr": 5.488372093023256e-06, "epoch": 0.16600790513833993, "percentage": 5.54, "elapsed_time": "0:16:03", "remaining_time": "4:33:55"}
+{"current_steps": 120, "total_steps": 2148, "loss": 0.7197, "lr": 5.534883720930233e-06, "epoch": 0.16740292955126715, "percentage": 5.59, "elapsed_time": "0:16:14", "remaining_time": "4:34:21"}
+{"current_steps": 121, "total_steps": 2148, "loss": 0.7522, "lr": 5.58139534883721e-06, "epoch": 0.16879795396419436, "percentage": 5.63, "elapsed_time": "0:16:26", "remaining_time": "4:35:32"}
+{"current_steps": 122, "total_steps": 2148, "loss": 0.7701, "lr": 5.627906976744186e-06, "epoch": 0.1701929783771216, "percentage": 5.68, "elapsed_time": "0:16:39", "remaining_time": "4:36:38"}
+{"current_steps": 123, "total_steps": 2148, "loss": 0.7893, "lr": 5.674418604651163e-06, "epoch": 0.17158800279004882, "percentage": 5.73, "elapsed_time": "0:16:47", "remaining_time": "4:36:20"}
+{"current_steps": 124, "total_steps": 2148, "loss": 0.7317, "lr": 5.72093023255814e-06, "epoch": 0.17298302720297606, "percentage": 5.77, "elapsed_time": "0:16:53", "remaining_time": "4:35:39"}
+{"current_steps": 125, "total_steps": 2148, "loss": 0.7706, "lr": 5.7674418604651175e-06, "epoch": 0.17437805161590328, "percentage": 5.82, "elapsed_time": "0:17:00", "remaining_time": "4:35:15"}
+{"current_steps": 126, "total_steps": 2148, "loss": 0.7555, "lr": 5.8139534883720935e-06, "epoch": 0.1757730760288305, "percentage": 5.87, "elapsed_time": "0:17:07", "remaining_time": "4:34:44"}
+{"current_steps": 127, "total_steps": 2148, "loss": 0.7651, "lr": 5.86046511627907e-06, "epoch": 0.17716810044175774, "percentage": 5.91, "elapsed_time": "0:17:13", "remaining_time": "4:34:09"}
+{"current_steps": 128, "total_steps": 2148, "loss": 0.7595, "lr": 5.906976744186047e-06, "epoch": 0.17856312485468495, "percentage": 5.96, "elapsed_time": "0:17:22", "remaining_time": "4:34:11"}
+{"current_steps": 129, "total_steps": 2148, "loss": 0.8296, "lr": 5.953488372093023e-06, "epoch": 0.1799581492676122, "percentage": 6.01, "elapsed_time": "0:17:32", "remaining_time": "4:34:31"}
+{"current_steps": 130, "total_steps": 2148, "loss": 0.745, "lr": 6e-06, "epoch": 0.1813531736805394, "percentage": 6.05, "elapsed_time": "0:17:38", "remaining_time": "4:33:54"}
+{"current_steps": 131, "total_steps": 2148, "loss": 0.6755, "lr": 6.046511627906977e-06, "epoch": 0.18274819809346662, "percentage": 6.1, "elapsed_time": "0:17:46", "remaining_time": "4:33:35"}
+{"current_steps": 132, "total_steps": 2148, "loss": 0.7682, "lr": 6.093023255813954e-06, "epoch": 0.18414322250639387, "percentage": 6.15, "elapsed_time": "0:17:53", "remaining_time": "4:33:09"}
+{"current_steps": 133, "total_steps": 2148, "loss": 0.7427, "lr": 6.139534883720932e-06, "epoch": 0.18553824691932108, "percentage": 6.19, "elapsed_time": "0:18:02", "remaining_time": "4:33:23"}
+{"current_steps": 134, "total_steps": 2148, "loss": 0.7355, "lr": 6.186046511627908e-06, "epoch": 0.18693327133224832, "percentage": 6.24, "elapsed_time": "0:18:14", "remaining_time": "4:34:02"}
+{"current_steps": 135, "total_steps": 2148, "loss": 0.8027, "lr": 6.2325581395348845e-06, "epoch": 0.18832829574517554, "percentage": 6.28, "elapsed_time": "0:18:21", "remaining_time": "4:33:43"}
+{"current_steps": 136, "total_steps": 2148, "loss": 0.7745, "lr": 6.279069767441861e-06, "epoch": 0.18972332015810275, "percentage": 6.33, "elapsed_time": "0:18:30", "remaining_time": "4:33:55"}
+{"current_steps": 137, "total_steps": 2148, "loss": 0.7536, "lr": 6.325581395348837e-06, "epoch": 0.19111834457103, "percentage": 6.38, "elapsed_time": "0:18:38", "remaining_time": "4:33:32"}
+{"current_steps": 138, "total_steps": 2148, "loss": 0.7661, "lr": 6.372093023255814e-06, "epoch": 0.1925133689839572, "percentage": 6.42, "elapsed_time": "0:18:44", "remaining_time": "4:32:56"}
+{"current_steps": 139, "total_steps": 2148, "loss": 0.7113, "lr": 6.418604651162791e-06, "epoch": 0.19390839339688445, "percentage": 6.47, "elapsed_time": "0:18:51", "remaining_time": "4:32:40"}
+{"current_steps": 140, "total_steps": 2148, "loss": 0.7844, "lr": 6.465116279069767e-06, "epoch": 0.19530341780981167, "percentage": 6.52, "elapsed_time": "0:18:58", "remaining_time": "4:32:02"}
+{"current_steps": 141, "total_steps": 2148, "loss": 0.7984, "lr": 6.511627906976745e-06, "epoch": 0.1966984422227389, "percentage": 6.56, "elapsed_time": "0:19:04", "remaining_time": "4:31:30"}
+{"current_steps": 142, "total_steps": 2148, "loss": 0.749, "lr": 6.558139534883722e-06, "epoch": 0.19809346663566613, "percentage": 6.61, "elapsed_time": "0:19:11", "remaining_time": "4:31:04"}
+{"current_steps": 143, "total_steps": 2148, "loss": 0.7404, "lr": 6.604651162790699e-06, "epoch": 0.19948849104859334, "percentage": 6.66, "elapsed_time": "0:19:18", "remaining_time": "4:30:46"}
+{"current_steps": 144, "total_steps": 2148, "loss": 0.7011, "lr": 6.651162790697675e-06, "epoch": 0.20088351546152058, "percentage": 6.7, "elapsed_time": "0:19:26", "remaining_time": "4:30:26"}
+{"current_steps": 145, "total_steps": 2148, "loss": 0.7989, "lr": 6.6976744186046515e-06, "epoch": 0.2022785398744478, "percentage": 6.75, "elapsed_time": "0:19:33", "remaining_time": "4:30:10"}
+{"current_steps": 146, "total_steps": 2148, "loss": 0.7381, "lr": 6.744186046511628e-06, "epoch": 0.20367356428737504, "percentage": 6.8, "elapsed_time": "0:19:40", "remaining_time": "4:29:40"}
+{"current_steps": 147, "total_steps": 2148, "loss": 0.8333, "lr": 6.790697674418605e-06, "epoch": 0.20506858870030226, "percentage": 6.84, "elapsed_time": "0:19:47", "remaining_time": "4:29:20"}
+{"current_steps": 148, "total_steps": 2148, "loss": 0.7275, "lr": 6.837209302325581e-06, "epoch": 0.20646361311322947, "percentage": 6.89, "elapsed_time": "0:19:54", "remaining_time": "4:29:03"}
+{"current_steps": 149, "total_steps": 2148, "loss": 0.8271, "lr": 6.883720930232559e-06, "epoch": 0.20785863752615671, "percentage": 6.94, "elapsed_time": "0:20:01", "remaining_time": "4:28:39"}
+{"current_steps": 150, "total_steps": 2148, "loss": 0.7624, "lr": 6.930232558139536e-06, "epoch": 0.20925366193908393, "percentage": 6.98, "elapsed_time": "0:20:08", "remaining_time": "4:28:12"}
+{"current_steps": 151, "total_steps": 2148, "loss": 0.7584, "lr": 6.976744186046513e-06, "epoch": 0.21064868635201117, "percentage": 7.03, "elapsed_time": "0:20:15", "remaining_time": "4:27:55"}
+{"current_steps": 152, "total_steps": 2148, "loss": 0.7009, "lr": 7.023255813953489e-06, "epoch": 0.2120437107649384, "percentage": 7.08, "elapsed_time": "0:20:23", "remaining_time": "4:27:40"}
+{"current_steps": 153, "total_steps": 2148, "loss": 0.7554, "lr": 7.069767441860466e-06, "epoch": 0.2134387351778656, "percentage": 7.12, "elapsed_time": "0:20:30", "remaining_time": "4:27:25"}
+{"current_steps": 154, "total_steps": 2148, "loss": 0.727, "lr": 7.1162790697674425e-06, "epoch": 0.21483375959079284, "percentage": 7.17, "elapsed_time": "0:20:36", "remaining_time": "4:26:55"}
+{"current_steps": 155, "total_steps": 2148, "loss": 0.7333, "lr": 7.1627906976744185e-06, "epoch": 0.21622878400372006, "percentage": 7.22, "elapsed_time": "0:20:43", "remaining_time": "4:26:27"}
+{"current_steps": 156, "total_steps": 2148, "loss": 0.755, "lr": 7.209302325581395e-06, "epoch": 0.2176238084166473, "percentage": 7.26, "elapsed_time": "0:20:55", "remaining_time": "4:27:14"}
+{"current_steps": 157, "total_steps": 2148, "loss": 0.6771, "lr": 7.255813953488373e-06, "epoch": 0.21901883282957452, "percentage": 7.31, "elapsed_time": "0:21:03", "remaining_time": "4:26:59"}
+{"current_steps": 158, "total_steps": 2148, "loss": 0.7357, "lr": 7.30232558139535e-06, "epoch": 0.22041385724250173, "percentage": 7.36, "elapsed_time": "0:21:11", "remaining_time": "4:26:51"}
+{"current_steps": 159, "total_steps": 2148, "loss": 0.7254, "lr": 7.348837209302326e-06, "epoch": 0.22180888165542897, "percentage": 7.4, "elapsed_time": "0:21:19", "remaining_time": "4:26:48"}
+{"current_steps": 160, "total_steps": 2148, "loss": 0.7488, "lr": 7.395348837209303e-06, "epoch": 0.2232039060683562, "percentage": 7.45, "elapsed_time": "0:21:26", "remaining_time": "4:26:19"}
+{"current_steps": 161, "total_steps": 2148, "loss": 0.7413, "lr": 7.44186046511628e-06, "epoch": 0.22459893048128343, "percentage": 7.5, "elapsed_time": "0:21:34", "remaining_time": "4:26:21"}
+{"current_steps": 162, "total_steps": 2148, "loss": 0.7363, "lr": 7.488372093023256e-06, "epoch": 0.22599395489421065, "percentage": 7.54, "elapsed_time": "0:21:42", "remaining_time": "4:26:11"}
+{"current_steps": 163, "total_steps": 2148, "loss": 0.7673, "lr": 7.534883720930233e-06, "epoch": 0.22738897930713786, "percentage": 7.59, "elapsed_time": "0:21:53", "remaining_time": "4:26:32"}
+{"current_steps": 164, "total_steps": 2148, "loss": 0.6675, "lr": 7.5813953488372095e-06, "epoch": 0.2287840037200651, "percentage": 7.64, "elapsed_time": "0:22:00", "remaining_time": "4:26:11"}
+{"current_steps": 165, "total_steps": 2148, "loss": 0.7483, "lr": 7.627906976744187e-06, "epoch": 0.23017902813299232, "percentage": 7.68, "elapsed_time": "0:22:11", "remaining_time": "4:26:45"}
+{"current_steps": 166, "total_steps": 2148, "loss": 0.6732, "lr": 7.674418604651164e-06, "epoch": 0.23157405254591956, "percentage": 7.73, "elapsed_time": "0:22:19", "remaining_time": "4:26:31"}
+{"current_steps": 167, "total_steps": 2148, "loss": 0.7009, "lr": 7.72093023255814e-06, "epoch": 0.23296907695884678, "percentage": 7.77, "elapsed_time": "0:22:26", "remaining_time": "4:26:12"}
+{"current_steps": 168, "total_steps": 2148, "loss": 0.7252, "lr": 7.767441860465116e-06, "epoch": 0.23436410137177402, "percentage": 7.82, "elapsed_time": "0:22:36", "remaining_time": "4:26:23"}
+{"current_steps": 169, "total_steps": 2148, "loss": 0.733, "lr": 7.813953488372094e-06, "epoch": 0.23575912578470123, "percentage": 7.87, "elapsed_time": "0:22:42", "remaining_time": "4:25:59"}
+{"current_steps": 170, "total_steps": 2148, "loss": 0.6615, "lr": 7.86046511627907e-06, "epoch": 0.23715415019762845, "percentage": 7.91, "elapsed_time": "0:22:55", "remaining_time": "4:26:45"}
+{"current_steps": 171, "total_steps": 2148, "loss": 0.7284, "lr": 7.906976744186048e-06, "epoch": 0.2385491746105557, "percentage": 7.96, "elapsed_time": "0:23:08", "remaining_time": "4:27:29"}
+{"current_steps": 172, "total_steps": 2148, "loss": 0.7081, "lr": 7.953488372093024e-06, "epoch": 0.2399441990234829, "percentage": 8.01, "elapsed_time": "0:23:19", "remaining_time": "4:27:56"}
+{"current_steps": 173, "total_steps": 2148, "loss": 0.7747, "lr": 8.000000000000001e-06, "epoch": 0.24133922343641015, "percentage": 8.05, "elapsed_time": "0:23:25", "remaining_time": "4:27:29"}
+{"current_steps": 174, "total_steps": 2148, "loss": 0.7386, "lr": 8.046511627906977e-06, "epoch": 0.24273424784933736, "percentage": 8.1, "elapsed_time": "0:23:33", "remaining_time": "4:27:11"}
+{"current_steps": 175, "total_steps": 2148, "loss": 0.7658, "lr": 8.093023255813955e-06, "epoch": 0.24412927226226458, "percentage": 8.15, "elapsed_time": "0:23:41", "remaining_time": "4:27:11"}
+{"current_steps": 176, "total_steps": 2148, "loss": 0.7733, "lr": 8.139534883720931e-06, "epoch": 0.24552429667519182, "percentage": 8.19, "elapsed_time": "0:23:51", "remaining_time": "4:27:20"}
+{"current_steps": 177, "total_steps": 2148, "loss": 0.7699, "lr": 8.186046511627907e-06, "epoch": 0.24691932108811904, "percentage": 8.24, "elapsed_time": "0:23:58", "remaining_time": "4:26:58"}
+{"current_steps": 178, "total_steps": 2148, "loss": 0.774, "lr": 8.232558139534885e-06, "epoch": 0.24831434550104628, "percentage": 8.29, "elapsed_time": "0:24:05", "remaining_time": "4:26:35"}
+{"current_steps": 179, "total_steps": 2148, "loss": 0.7115, "lr": 8.279069767441861e-06, "epoch": 0.2497093699139735, "percentage": 8.33, "elapsed_time": "0:24:12", "remaining_time": "4:26:12"}
+{"current_steps": 180, "total_steps": 2148, "loss": 0.7587, "lr": 8.325581395348837e-06, "epoch": 0.25110439432690074, "percentage": 8.38, "elapsed_time": "0:24:21", "remaining_time": "4:26:20"}
+{"current_steps": 181, "total_steps": 2148, "loss": 0.75, "lr": 8.372093023255815e-06, "epoch": 0.25249941873982795, "percentage": 8.43, "elapsed_time": "0:24:27", "remaining_time": "4:25:50"}
+{"current_steps": 182, "total_steps": 2148, "loss": 0.7088, "lr": 8.418604651162792e-06, "epoch": 0.25389444315275517, "percentage": 8.47, "elapsed_time": "0:24:34", "remaining_time": "4:25:30"}
+{"current_steps": 183, "total_steps": 2148, "loss": 0.7468, "lr": 8.465116279069768e-06, "epoch": 0.2552894675656824, "percentage": 8.52, "elapsed_time": "0:24:41", "remaining_time": "4:25:09"}
+{"current_steps": 184, "total_steps": 2148, "loss": 0.7142, "lr": 8.511627906976744e-06, "epoch": 0.25668449197860965, "percentage": 8.57, "elapsed_time": "0:24:49", "remaining_time": "4:25:01"}
+{"current_steps": 185, "total_steps": 2148, "loss": 0.7071, "lr": 8.558139534883722e-06, "epoch": 0.25807951639153687, "percentage": 8.61, "elapsed_time": "0:24:55", "remaining_time": "4:24:32"}
+{"current_steps": 186, "total_steps": 2148, "loss": 0.6276, "lr": 8.604651162790698e-06, "epoch": 0.2594745408044641, "percentage": 8.66, "elapsed_time": "0:25:02", "remaining_time": "4:24:04"}
+{"current_steps": 187, "total_steps": 2148, "loss": 0.7856, "lr": 8.651162790697674e-06, "epoch": 0.2608695652173913, "percentage": 8.71, "elapsed_time": "0:25:08", "remaining_time": "4:23:42"}
+{"current_steps": 188, "total_steps": 2148, "loss": 0.7252, "lr": 8.697674418604652e-06, "epoch": 0.2622645896303185, "percentage": 8.75, "elapsed_time": "0:25:14", "remaining_time": "4:23:13"}
+{"current_steps": 189, "total_steps": 2148, "loss": 0.6964, "lr": 8.74418604651163e-06, "epoch": 0.2636596140432458, "percentage": 8.8, "elapsed_time": "0:25:22", "remaining_time": "4:23:04"}
+{"current_steps": 190, "total_steps": 2148, "loss": 0.7203, "lr": 8.790697674418606e-06, "epoch": 0.265054638456173, "percentage": 8.85, "elapsed_time": "0:25:30", "remaining_time": "4:22:52"}
+{"current_steps": 191, "total_steps": 2148, "loss": 0.7592, "lr": 8.837209302325582e-06, "epoch": 0.2664496628691002, "percentage": 8.89, "elapsed_time": "0:25:38", "remaining_time": "4:22:42"}
+{"current_steps": 192, "total_steps": 2148, "loss": 0.701, "lr": 8.88372093023256e-06, "epoch": 0.2678446872820274, "percentage": 8.94, "elapsed_time": "0:25:44", "remaining_time": "4:22:17"}
+{"current_steps": 193, "total_steps": 2148, "loss": 0.7542, "lr": 8.930232558139535e-06, "epoch": 0.26923971169495464, "percentage": 8.99, "elapsed_time": "0:25:54", "remaining_time": "4:22:24"}
+{"current_steps": 194, "total_steps": 2148, "loss": 0.7001, "lr": 8.976744186046511e-06, "epoch": 0.2706347361078819, "percentage": 9.03, "elapsed_time": "0:26:00", "remaining_time": "4:21:56"}
+{"current_steps": 195, "total_steps": 2148, "loss": 0.7146, "lr": 9.023255813953489e-06, "epoch": 0.2720297605208091, "percentage": 9.08, "elapsed_time": "0:26:07", "remaining_time": "4:21:40"}
+{"current_steps": 196, "total_steps": 2148, "loss": 0.7804, "lr": 9.069767441860465e-06, "epoch": 0.27342478493373634, "percentage": 9.12, "elapsed_time": "0:26:15", "remaining_time": "4:21:27"}
+{"current_steps": 197, "total_steps": 2148, "loss": 0.709, "lr": 9.116279069767443e-06, "epoch": 0.27481980934666356, "percentage": 9.17, "elapsed_time": "0:26:21", "remaining_time": "4:21:01"}
+{"current_steps": 198, "total_steps": 2148, "loss": 0.7079, "lr": 9.162790697674419e-06, "epoch": 0.27621483375959077, "percentage": 9.22, "elapsed_time": "0:26:30", "remaining_time": "4:21:01"}
+{"current_steps": 199, "total_steps": 2148, "loss": 0.7115, "lr": 9.209302325581397e-06, "epoch": 0.27760985817251804, "percentage": 9.26, "elapsed_time": "0:26:36", "remaining_time": "4:20:37"}
+{"current_steps": 200, "total_steps": 2148, "loss": 0.7238, "lr": 9.255813953488373e-06, "epoch": 0.27900488258544526, "percentage": 9.31, "elapsed_time": "0:26:43", "remaining_time": "4:20:17"}
+{"current_steps": 201, "total_steps": 2148, "loss": 0.7214, "lr": 9.30232558139535e-06, "epoch": 0.2803999069983725, "percentage": 9.36, "elapsed_time": "0:26:50", "remaining_time": "4:19:58"}
+{"current_steps": 202, "total_steps": 2148, "loss": 0.7474, "lr": 9.348837209302326e-06, "epoch": 0.2817949314112997, "percentage": 9.4, "elapsed_time": "0:26:59", "remaining_time": "4:20:00"}
+{"current_steps": 203, "total_steps": 2148, "loss": 0.7681, "lr": 9.395348837209302e-06, "epoch": 0.2831899558242269, "percentage": 9.45, "elapsed_time": "0:27:08", "remaining_time": "4:19:59"}
+{"current_steps": 204, "total_steps": 2148, "loss": 0.727, "lr": 9.44186046511628e-06, "epoch": 0.2845849802371542, "percentage": 9.5, "elapsed_time": "0:27:16", "remaining_time": "4:19:57"}
+{"current_steps": 205, "total_steps": 2148, "loss": 0.7719, "lr": 9.488372093023258e-06, "epoch": 0.2859800046500814, "percentage": 9.54, "elapsed_time": "0:27:25", "remaining_time": "4:19:52"}
+{"current_steps": 206, "total_steps": 2148, "loss": 0.7441, "lr": 9.534883720930234e-06, "epoch": 0.2873750290630086, "percentage": 9.59, "elapsed_time": "0:27:36", "remaining_time": "4:20:14"}
+{"current_steps": 207, "total_steps": 2148, "loss": 0.6955, "lr": 9.58139534883721e-06, "epoch": 0.2887700534759358, "percentage": 9.64, "elapsed_time": "0:27:44", "remaining_time": "4:20:03"}
+{"current_steps": 208, "total_steps": 2148, "loss": 0.7716, "lr": 9.627906976744188e-06, "epoch": 0.29016507788886303, "percentage": 9.68, "elapsed_time": "0:27:50", "remaining_time": "4:19:44"}
+{"current_steps": 209, "total_steps": 2148, "loss": 0.7583, "lr": 9.674418604651164e-06, "epoch": 0.2915601023017903, "percentage": 9.73, "elapsed_time": "0:27:59", "remaining_time": "4:19:38"}
+{"current_steps": 210, "total_steps": 2148, "loss": 0.7681, "lr": 9.72093023255814e-06, "epoch": 0.2929551267147175, "percentage": 9.78, "elapsed_time": "0:28:08", "remaining_time": "4:19:42"}
+{"current_steps": 211, "total_steps": 2148, "loss": 0.7208, "lr": 9.767441860465117e-06, "epoch": 0.29435015112764473, "percentage": 9.82, "elapsed_time": "0:28:14", "remaining_time": "4:19:17"}
+{"current_steps": 212, "total_steps": 2148, "loss": 0.7693, "lr": 9.813953488372093e-06, "epoch": 0.29574517554057195, "percentage": 9.87, "elapsed_time": "0:28:20", "remaining_time": "4:18:51"}
+{"current_steps": 213, "total_steps": 2148, "loss": 0.7405, "lr": 9.860465116279071e-06, "epoch": 0.29714019995349916, "percentage": 9.92, "elapsed_time": "0:28:27", "remaining_time": "4:18:33"}
+{"current_steps": 214, "total_steps": 2148, "loss": 0.7111, "lr": 9.906976744186047e-06, "epoch": 0.29853522436642643, "percentage": 9.96, "elapsed_time": "0:28:34", "remaining_time": "4:18:11"}
+{"current_steps": 215, "total_steps": 2148, "loss": 0.7665, "lr": 9.953488372093025e-06, "epoch": 0.29993024877935365, "percentage": 10.01, "elapsed_time": "0:28:41", "remaining_time": "4:17:57"}
+{"current_steps": 216, "total_steps": 2148, "loss": 0.7454, "lr": 1e-05, "epoch": 0.30132527319228086, "percentage": 10.06, "elapsed_time": "0:28:50", "remaining_time": "4:17:57"}
+{"current_steps": 217, "total_steps": 2148, "loss": 0.7878, "lr": 9.999993396473114e-06, "epoch": 0.3027202976052081, "percentage": 10.1, "elapsed_time": "0:28:58", "remaining_time": "4:17:48"}
+{"current_steps": 218, "total_steps": 2148, "loss": 0.6551, "lr": 9.999973585909898e-06, "epoch": 0.3041153220181353, "percentage": 10.15, "elapsed_time": "0:29:04", "remaining_time": "4:17:26"}
+{"current_steps": 219, "total_steps": 2148, "loss": 0.7145, "lr": 9.99994056836268e-06, "epoch": 0.30551034643106256, "percentage": 10.2, "elapsed_time": "0:29:12", "remaining_time": "4:17:16"}
+{"current_steps": 220, "total_steps": 2148, "loss": 0.7239, "lr": 9.999894343918674e-06, "epoch": 0.3069053708439898, "percentage": 10.24, "elapsed_time": "0:29:22", "remaining_time": "4:17:25"}
+{"current_steps": 221, "total_steps": 2148, "loss": 0.7756, "lr": 9.999834912699974e-06, "epoch": 0.308300395256917, "percentage": 10.29, "elapsed_time": "0:29:39", "remaining_time": "4:18:35"}
+{"current_steps": 222, "total_steps": 2148, "loss": 0.7255, "lr": 9.999762274863567e-06, "epoch": 0.3096954196698442, "percentage": 10.34, "elapsed_time": "0:29:50", "remaining_time": "4:18:50"}
+{"current_steps": 223, "total_steps": 2148, "loss": 0.7809, "lr": 9.999676430601318e-06, "epoch": 0.3110904440827714, "percentage": 10.38, "elapsed_time": "0:30:02", "remaining_time": "4:19:18"}
+{"current_steps": 224, "total_steps": 2148, "loss": 0.702, "lr": 9.999577380139976e-06, "epoch": 0.3124854684956987, "percentage": 10.43, "elapsed_time": "0:30:08", "remaining_time": "4:18:55"}
+{"current_steps": 225, "total_steps": 2148, "loss": 0.7084, "lr": 9.999465123741172e-06, "epoch": 0.3138804929086259, "percentage": 10.47, "elapsed_time": "0:30:15", "remaining_time": "4:18:36"}
+{"current_steps": 226, "total_steps": 2148, "loss": 0.7033, "lr": 9.999339661701424e-06, "epoch": 0.3152755173215531, "percentage": 10.52, "elapsed_time": "0:30:22", "remaining_time": "4:18:18"}
+{"current_steps": 227, "total_steps": 2148, "loss": 0.7521, "lr": 9.99920099435213e-06, "epoch": 0.31667054173448034, "percentage": 10.57, "elapsed_time": "0:30:28", "remaining_time": "4:17:53"}
+{"current_steps": 228, "total_steps": 2148, "loss": 0.7706, "lr": 9.999049122059565e-06, "epoch": 0.31806556614740755, "percentage": 10.61, "elapsed_time": "0:30:40", "remaining_time": "4:18:19"}
+{"current_steps": 229, "total_steps": 2148, "loss": 0.7236, "lr": 9.998884045224886e-06, "epoch": 0.3194605905603348, "percentage": 10.66, "elapsed_time": "0:30:47", "remaining_time": "4:18:05"}
+{"current_steps": 230, "total_steps": 2148, "loss": 0.6958, "lr": 9.998705764284132e-06, "epoch": 0.32085561497326204, "percentage": 10.71, "elapsed_time": "0:30:55", "remaining_time": "4:17:51"}
+{"current_steps": 231, "total_steps": 2148, "loss": 0.7827, "lr": 9.998514279708212e-06, "epoch": 0.32225063938618925, "percentage": 10.75, "elapsed_time": "0:31:06", "remaining_time": "4:18:08"}
+{"current_steps": 232, "total_steps": 2148, "loss": 0.7762, "lr": 9.998309592002914e-06, "epoch": 0.32364566379911647, "percentage": 10.8, "elapsed_time": "0:31:12", "remaining_time": "4:17:44"}
+{"current_steps": 233, "total_steps": 2148, "loss": 0.6852, "lr": 9.99809170170891e-06, "epoch": 0.32504068821204374, "percentage": 10.85, "elapsed_time": "0:31:26", "remaining_time": "4:18:22"}
+{"current_steps": 234, "total_steps": 2148, "loss": 0.7778, "lr": 9.997860609401732e-06, "epoch": 0.32643571262497095, "percentage": 10.89, "elapsed_time": "0:31:33", "remaining_time": "4:18:05"}
+{"current_steps": 235, "total_steps": 2148, "loss": 0.7799, "lr": 9.99761631569179e-06, "epoch": 0.32783073703789817, "percentage": 10.94, "elapsed_time": "0:31:41", "remaining_time": "4:18:02"}
+{"current_steps": 236, "total_steps": 2148, "loss": 0.801, "lr": 9.997358821224365e-06, "epoch": 0.3292257614508254, "percentage": 10.99, "elapsed_time": "0:31:49", "remaining_time": "4:17:52"}
+{"current_steps": 237, "total_steps": 2148, "loss": 0.7303, "lr": 9.997088126679607e-06, "epoch": 0.3306207858637526, "percentage": 11.03, "elapsed_time": "0:31:56", "remaining_time": "4:17:30"}
+{"current_steps": 238, "total_steps": 2148, "loss": 0.7332, "lr": 9.996804232772528e-06, "epoch": 0.33201581027667987, "percentage": 11.08, "elapsed_time": "0:32:03", "remaining_time": "4:17:15"}
+{"current_steps": 239, "total_steps": 2148, "loss": 0.7565, "lr": 9.996507140253012e-06, "epoch": 0.3334108346896071, "percentage": 11.13, "elapsed_time": "0:32:11", "remaining_time": "4:17:09"}
+{"current_steps": 240, "total_steps": 2148, "loss": 0.7403, "lr": 9.9961968499058e-06, "epoch": 0.3348058591025343, "percentage": 11.17, "elapsed_time": "0:32:18", "remaining_time": "4:16:52"}
+{"current_steps": 241, "total_steps": 2148, "loss": 0.7672, "lr": 9.9958733625505e-06, "epoch": 0.3362008835154615, "percentage": 11.22, "elapsed_time": "0:32:25", "remaining_time": "4:16:34"}
+{"current_steps": 242, "total_steps": 2148, "loss": 0.7083, "lr": 9.995536679041568e-06, "epoch": 0.3375959079283887, "percentage": 11.27, "elapsed_time": "0:32:31", "remaining_time": "4:16:10"}
+{"current_steps": 243, "total_steps": 2148, "loss": 0.759, "lr": 9.99518680026833e-06, "epoch": 0.338990932341316, "percentage": 11.31, "elapsed_time": "0:32:43", "remaining_time": "4:16:31"}
+{"current_steps": 244, "total_steps": 2148, "loss": 0.7431, "lr": 9.994823727154957e-06, "epoch": 0.3403859567542432, "percentage": 11.36, "elapsed_time": "0:32:50", "remaining_time": "4:16:12"}
+{"current_steps": 245, "total_steps": 2148, "loss": 0.7658, "lr": 9.994447460660473e-06, "epoch": 0.34178098116717043, "percentage": 11.41, "elapsed_time": "0:32:57", "remaining_time": "4:16:00"}
+{"current_steps": 246, "total_steps": 2148, "loss": 0.7533, "lr": 9.994058001778754e-06, "epoch": 0.34317600558009764, "percentage": 11.45, "elapsed_time": "0:33:08", "remaining_time": "4:16:14"}
+{"current_steps": 247, "total_steps": 2148, "loss": 0.7256, "lr": 9.99365535153852e-06, "epoch": 0.34457102999302486, "percentage": 11.5, "elapsed_time": "0:33:14", "remaining_time": "4:15:53"}
+{"current_steps": 248, "total_steps": 2148, "loss": 0.7366, "lr": 9.993239511003338e-06, "epoch": 0.34596605440595213, "percentage": 11.55, "elapsed_time": "0:33:22", "remaining_time": "4:15:40"}
+{"current_steps": 249, "total_steps": 2148, "loss": 0.7347, "lr": 9.992810481271611e-06, "epoch": 0.34736107881887934, "percentage": 11.59, "elapsed_time": "0:33:28", "remaining_time": "4:15:21"}
+{"current_steps": 250, "total_steps": 2148, "loss": 0.7511, "lr": 9.992368263476585e-06, "epoch": 0.34875610323180656, "percentage": 11.64, "elapsed_time": "0:33:35", "remaining_time": "4:15:04"}
+{"current_steps": 251, "total_steps": 2148, "loss": 0.7604, "lr": 9.991912858786335e-06, "epoch": 0.3501511276447338, "percentage": 11.69, "elapsed_time": "0:33:42", "remaining_time": "4:14:45"}
+{"current_steps": 252, "total_steps": 2148, "loss": 0.7632, "lr": 9.991444268403776e-06, "epoch": 0.351546152057661, "percentage": 11.73, "elapsed_time": "0:33:53", "remaining_time": "4:14:59"}
+{"current_steps": 253, "total_steps": 2148, "loss": 0.8016, "lr": 9.990962493566645e-06, "epoch": 0.35294117647058826, "percentage": 11.78, "elapsed_time": "0:34:00", "remaining_time": "4:14:43"}
+{"current_steps": 254, "total_steps": 2148, "loss": 0.7021, "lr": 9.99046753554751e-06, "epoch": 0.3543362008835155, "percentage": 11.82, "elapsed_time": "0:34:06", "remaining_time": "4:14:20"}
+{"current_steps": 255, "total_steps": 2148, "loss": 0.7367, "lr": 9.989959395653756e-06, "epoch": 0.3557312252964427, "percentage": 11.87, "elapsed_time": "0:34:12", "remaining_time": "4:13:58"}
+{"current_steps": 256, "total_steps": 2148, "loss": 0.6324, "lr": 9.989438075227588e-06, "epoch": 0.3571262497093699, "percentage": 11.92, "elapsed_time": "0:34:21", "remaining_time": "4:13:52"}
+{"current_steps": 257, "total_steps": 2148, "loss": 0.6831, "lr": 9.988903575646032e-06, "epoch": 0.3585212741222971, "percentage": 11.96, "elapsed_time": "0:34:27", "remaining_time": "4:13:31"}
+{"current_steps": 258, "total_steps": 2148, "loss": 0.7296, "lr": 9.988355898320917e-06, "epoch": 0.3599162985352244, "percentage": 12.01, "elapsed_time": "0:34:37", "remaining_time": "4:13:39"}
+{"current_steps": 259, "total_steps": 2148, "loss": 0.7603, "lr": 9.987795044698885e-06, "epoch": 0.3613113229481516, "percentage": 12.06, "elapsed_time": "0:34:48", "remaining_time": "4:13:51"}
+{"current_steps": 260, "total_steps": 2148, "loss": 0.7424, "lr": 9.98722101626138e-06, "epoch": 0.3627063473610788, "percentage": 12.1, "elapsed_time": "0:34:55", "remaining_time": "4:13:35"}
+{"current_steps": 261, "total_steps": 2148, "loss": 0.7005, "lr": 9.986633814524648e-06, "epoch": 0.36410137177400603, "percentage": 12.15, "elapsed_time": "0:35:03", "remaining_time": "4:13:29"}
+{"current_steps": 262, "total_steps": 2148, "loss": 0.7353, "lr": 9.986033441039731e-06, "epoch": 0.36549639618693325, "percentage": 12.2, "elapsed_time": "0:35:10", "remaining_time": "4:13:10"}
+{"current_steps": 263, "total_steps": 2148, "loss": 0.7928, "lr": 9.985419897392459e-06, "epoch": 0.3668914205998605, "percentage": 12.24, "elapsed_time": "0:35:19", "remaining_time": "4:13:09"}
+{"current_steps": 264, "total_steps": 2148, "loss": 0.6918, "lr": 9.984793185203456e-06, "epoch": 0.36828644501278773, "percentage": 12.29, "elapsed_time": "0:35:26", "remaining_time": "4:12:55"}
+{"current_steps": 265, "total_steps": 2148, "loss": 0.7581, "lr": 9.984153306128124e-06, "epoch": 0.36968146942571495, "percentage": 12.34, "elapsed_time": "0:35:32", "remaining_time": "4:12:35"}
+{"current_steps": 266, "total_steps": 2148, "loss": 0.7616, "lr": 9.983500261856646e-06, "epoch": 0.37107649383864216, "percentage": 12.38, "elapsed_time": "0:35:39", "remaining_time": "4:12:20"}
+{"current_steps": 267, "total_steps": 2148, "loss": 0.6927, "lr": 9.982834054113982e-06, "epoch": 0.3724715182515694, "percentage": 12.43, "elapsed_time": "0:35:47", "remaining_time": "4:12:10"}
+{"current_steps": 268, "total_steps": 2148, "loss": 0.7197, "lr": 9.98215468465986e-06, "epoch": 0.37386654266449665, "percentage": 12.48, "elapsed_time": "0:35:55", "remaining_time": "4:11:57"}
+{"current_steps": 269, "total_steps": 2148, "loss": 0.7377, "lr": 9.981462155288773e-06, "epoch": 0.37526156707742386, "percentage": 12.52, "elapsed_time": "0:36:02", "remaining_time": "4:11:44"}
+{"current_steps": 270, "total_steps": 2148, "loss": 0.6838, "lr": 9.980756467829977e-06, "epoch": 0.3766565914903511, "percentage": 12.57, "elapsed_time": "0:36:08", "remaining_time": "4:11:23"}
+{"current_steps": 271, "total_steps": 2148, "loss": 0.7265, "lr": 9.98003762414748e-06, "epoch": 0.3780516159032783, "percentage": 12.62, "elapsed_time": "0:36:14", "remaining_time": "4:11:04"}
+{"current_steps": 272, "total_steps": 2148, "loss": 0.6886, "lr": 9.979305626140046e-06, "epoch": 0.3794466403162055, "percentage": 12.66, "elapsed_time": "0:36:22", "remaining_time": "4:10:49"}
+{"current_steps": 273, "total_steps": 2148, "loss": 0.7118, "lr": 9.978560475741181e-06, "epoch": 0.3808416647291328, "percentage": 12.71, "elapsed_time": "0:36:28", "remaining_time": "4:10:28"}
+{"current_steps": 274, "total_steps": 2148, "loss": 0.7936, "lr": 9.977802174919134e-06, "epoch": 0.38223668914206, "percentage": 12.76, "elapsed_time": "0:36:35", "remaining_time": "4:10:13"}
+{"current_steps": 275, "total_steps": 2148, "loss": 0.6947, "lr": 9.977030725676887e-06, "epoch": 0.3836317135549872, "percentage": 12.8, "elapsed_time": "0:36:42", "remaining_time": "4:09:58"}
+{"current_steps": 276, "total_steps": 2148, "loss": 0.7561, "lr": 9.976246130052157e-06, "epoch": 0.3850267379679144, "percentage": 12.85, "elapsed_time": "0:36:55", "remaining_time": "4:10:27"}
+{"current_steps": 277, "total_steps": 2148, "loss": 0.8082, "lr": 9.97544839011738e-06, "epoch": 0.38642176238084164, "percentage": 12.9, "elapsed_time": "0:37:07", "remaining_time": "4:10:47"}
+{"current_steps": 278, "total_steps": 2148, "loss": 0.6938, "lr": 9.974637507979721e-06, "epoch": 0.3878167867937689, "percentage": 12.94, "elapsed_time": "0:37:16", "remaining_time": "4:10:44"}
+{"current_steps": 279, "total_steps": 2148, "loss": 0.7031, "lr": 9.973813485781045e-06, "epoch": 0.3892118112066961, "percentage": 12.99, "elapsed_time": "0:37:25", "remaining_time": "4:10:42"}
+{"current_steps": 280, "total_steps": 2148, "loss": 0.6992, "lr": 9.972976325697938e-06, "epoch": 0.39060683561962334, "percentage": 13.04, "elapsed_time": "0:37:33", "remaining_time": "4:10:31"}
+{"current_steps": 281, "total_steps": 2148, "loss": 0.7237, "lr": 9.972126029941685e-06, "epoch": 0.39200186003255055, "percentage": 13.08, "elapsed_time": "0:37:40", "remaining_time": "4:10:21"}
+{"current_steps": 282, "total_steps": 2148, "loss": 0.7034, "lr": 9.97126260075826e-06, "epoch": 0.3933968844454778, "percentage": 13.13, "elapsed_time": "0:37:50", "remaining_time": "4:10:27"}
+{"current_steps": 283, "total_steps": 2148, "loss": 0.7811, "lr": 9.97038604042834e-06, "epoch": 0.39479190885840504, "percentage": 13.18, "elapsed_time": "0:37:58", "remaining_time": "4:10:16"}
+{"current_steps": 284, "total_steps": 2148, "loss": 0.7176, "lr": 9.969496351267278e-06, "epoch": 0.39618693327133225, "percentage": 13.22, "elapsed_time": "0:38:14", "remaining_time": "4:11:02"}
+{"current_steps": 285, "total_steps": 2148, "loss": 0.7466, "lr": 9.96859353562511e-06, "epoch": 0.39758195768425947, "percentage": 13.27, "elapsed_time": "0:38:22", "remaining_time": "4:10:49"}
+{"current_steps": 286, "total_steps": 2148, "loss": 0.7166, "lr": 9.967677595886542e-06, "epoch": 0.3989769820971867, "percentage": 13.31, "elapsed_time": "0:38:29", "remaining_time": "4:10:33"}
+{"current_steps": 287, "total_steps": 2148, "loss": 0.743, "lr": 9.96674853447095e-06, "epoch": 0.40037200651011395, "percentage": 13.36, "elapsed_time": "0:38:37", "remaining_time": "4:10:27"}
+{"current_steps": 288, "total_steps": 2148, "loss": 0.6622, "lr": 9.96580635383236e-06, "epoch": 0.40176703092304117, "percentage": 13.41, "elapsed_time": "0:38:43", "remaining_time": "4:10:08"}
+{"current_steps": 289, "total_steps": 2148, "loss": 0.6307, "lr": 9.964851056459465e-06, "epoch": 0.4031620553359684, "percentage": 13.45, "elapsed_time": "0:38:52", "remaining_time": "4:10:01"}
+{"current_steps": 290, "total_steps": 2148, "loss": 0.7169, "lr": 9.963882644875594e-06, "epoch": 0.4045570797488956, "percentage": 13.5, "elapsed_time": "0:38:59", "remaining_time": "4:09:47"}
+{"current_steps": 291, "total_steps": 2148, "loss": 0.7076, "lr": 9.96290112163872e-06, "epoch": 0.4059521041618228, "percentage": 13.55, "elapsed_time": "0:39:05", "remaining_time": "4:09:29"}
+{"current_steps": 292, "total_steps": 2148, "loss": 0.7351, "lr": 9.961906489341452e-06, "epoch": 0.4073471285747501, "percentage": 13.59, "elapsed_time": "0:39:15", "remaining_time": "4:09:29"}
+{"current_steps": 293, "total_steps": 2148, "loss": 0.7711, "lr": 9.960898750611019e-06, "epoch": 0.4087421529876773, "percentage": 13.64, "elapsed_time": "0:39:21", "remaining_time": "4:09:08"}
+{"current_steps": 294, "total_steps": 2148, "loss": 0.8339, "lr": 9.959877908109274e-06, "epoch": 0.4101371774006045, "percentage": 13.69, "elapsed_time": "0:39:28", "remaining_time": "4:08:54"}
+{"current_steps": 295, "total_steps": 2148, "loss": 0.681, "lr": 9.958843964532683e-06, "epoch": 0.41153220181353173, "percentage": 13.73, "elapsed_time": "0:39:34", "remaining_time": "4:08:33"}
+{"current_steps": 296, "total_steps": 2148, "loss": 0.7126, "lr": 9.957796922612314e-06, "epoch": 0.41292722622645894, "percentage": 13.78, "elapsed_time": "0:39:41", "remaining_time": "4:08:17"}
+{"current_steps": 297, "total_steps": 2148, "loss": 0.771, "lr": 9.956736785113833e-06, "epoch": 0.4143222506393862, "percentage": 13.83, "elapsed_time": "0:39:55", "remaining_time": "4:08:49"}
+{"current_steps": 298, "total_steps": 2148, "loss": 0.6509, "lr": 9.955663554837503e-06, "epoch": 0.41571727505231343, "percentage": 13.87, "elapsed_time": "0:40:01", "remaining_time": "4:08:31"}
+{"current_steps": 299, "total_steps": 2148, "loss": 0.7163, "lr": 9.954577234618162e-06, "epoch": 0.41711229946524064, "percentage": 13.92, "elapsed_time": "0:40:14", "remaining_time": "4:08:53"}
+{"current_steps": 300, "total_steps": 2148, "loss": 0.7352, "lr": 9.953477827325229e-06, "epoch": 0.41850732387816786, "percentage": 13.97, "elapsed_time": "0:40:21", "remaining_time": "4:08:39"}
+{"current_steps": 301, "total_steps": 2148, "loss": 0.7667, "lr": 9.952365335862693e-06, "epoch": 0.4199023482910951, "percentage": 14.01, "elapsed_time": "0:40:29", "remaining_time": "4:08:25"}
+{"current_steps": 302, "total_steps": 2148, "loss": 0.787, "lr": 9.951239763169097e-06, "epoch": 0.42129737270402234, "percentage": 14.06, "elapsed_time": "0:40:36", "remaining_time": "4:08:13"}
+{"current_steps": 303, "total_steps": 2148, "loss": 0.7207, "lr": 9.950101112217543e-06, "epoch": 0.42269239711694956, "percentage": 14.11, "elapsed_time": "0:40:46", "remaining_time": "4:08:15"}
+{"current_steps": 304, "total_steps": 2148, "loss": 0.6965, "lr": 9.948949386015677e-06, "epoch": 0.4240874215298768, "percentage": 14.15, "elapsed_time": "0:40:53", "remaining_time": "4:08:03"}
+{"current_steps": 305, "total_steps": 2148, "loss": 0.697, "lr": 9.947784587605678e-06, "epoch": 0.425482445942804, "percentage": 14.2, "elapsed_time": "0:41:02", "remaining_time": "4:07:59"}
+{"current_steps": 306, "total_steps": 2148, "loss": 0.7564, "lr": 9.946606720064257e-06, "epoch": 0.4268774703557312, "percentage": 14.25, "elapsed_time": "0:41:08", "remaining_time": "4:07:40"}
+{"current_steps": 307, "total_steps": 2148, "loss": 0.7356, "lr": 9.945415786502649e-06, "epoch": 0.4282724947686585, "percentage": 14.29, "elapsed_time": "0:41:14", "remaining_time": "4:07:19"}
+{"current_steps": 308, "total_steps": 2148, "loss": 0.7234, "lr": 9.944211790066597e-06, "epoch": 0.4296675191815857, "percentage": 14.34, "elapsed_time": "0:41:21", "remaining_time": "4:07:04"}
+{"current_steps": 309, "total_steps": 2148, "loss": 0.7773, "lr": 9.94299473393635e-06, "epoch": 0.4310625435945129, "percentage": 14.39, "elapsed_time": "0:41:28", "remaining_time": "4:06:50"}
+{"current_steps": 310, "total_steps": 2148, "loss": 0.7175, "lr": 9.941764621326655e-06, "epoch": 0.4324575680074401, "percentage": 14.43, "elapsed_time": "0:41:35", "remaining_time": "4:06:34"}
+{"current_steps": 311, "total_steps": 2148, "loss": 0.7542, "lr": 9.94052145548674e-06, "epoch": 0.43385259242036733, "percentage": 14.48, "elapsed_time": "0:41:42", "remaining_time": "4:06:23"}
+{"current_steps": 312, "total_steps": 2148, "loss": 0.7845, "lr": 9.939265239700321e-06, "epoch": 0.4352476168332946, "percentage": 14.53, "elapsed_time": "0:41:51", "remaining_time": "4:06:19"}
+{"current_steps": 313, "total_steps": 2148, "loss": 0.7494, "lr": 9.93799597728558e-06, "epoch": 0.4366426412462218, "percentage": 14.57, "elapsed_time": "0:41:57", "remaining_time": "4:06:00"}
+{"current_steps": 314, "total_steps": 2148, "loss": 0.7801, "lr": 9.936713671595158e-06, "epoch": 0.43803766565914903, "percentage": 14.62, "elapsed_time": "0:42:06", "remaining_time": "4:05:56"}
+{"current_steps": 315, "total_steps": 2148, "loss": 0.7055, "lr": 9.935418326016153e-06, "epoch": 0.43943269007207625, "percentage": 14.66, "elapsed_time": "0:42:16", "remaining_time": "4:06:02"}
+{"current_steps": 316, "total_steps": 2148, "loss": 0.7864, "lr": 9.934109943970103e-06, "epoch": 0.44082771448500346, "percentage": 14.71, "elapsed_time": "0:42:26", "remaining_time": "4:06:04"}
+{"current_steps": 317, "total_steps": 2148, "loss": 0.7558, "lr": 9.932788528912983e-06, "epoch": 0.44222273889793073, "percentage": 14.76, "elapsed_time": "0:42:36", "remaining_time": "4:06:08"}
+{"current_steps": 318, "total_steps": 2148, "loss": 0.7549, "lr": 9.931454084335192e-06, "epoch": 0.44361776331085795, "percentage": 14.8, "elapsed_time": "0:42:46", "remaining_time": "4:06:09"}
+{"current_steps": 319, "total_steps": 2148, "loss": 0.7004, "lr": 9.930106613761549e-06, "epoch": 0.44501278772378516, "percentage": 14.85, "elapsed_time": "0:42:53", "remaining_time": "4:05:52"}
+{"current_steps": 320, "total_steps": 2148, "loss": 0.7335, "lr": 9.928746120751275e-06, "epoch": 0.4464078121367124, "percentage": 14.9, "elapsed_time": "0:43:03", "remaining_time": "4:05:57"}
+{"current_steps": 321, "total_steps": 2148, "loss": 0.7691, "lr": 9.927372608897992e-06, "epoch": 0.4478028365496396, "percentage": 14.94, "elapsed_time": "0:43:11", "remaining_time": "4:05:51"}
+{"current_steps": 322, "total_steps": 2148, "loss": 0.6896, "lr": 9.925986081829708e-06, "epoch": 0.44919786096256686, "percentage": 14.99, "elapsed_time": "0:43:18", "remaining_time": "4:05:36"}
+{"current_steps": 323, "total_steps": 2148, "loss": 0.7663, "lr": 9.924586543208812e-06, "epoch": 0.4505928853754941, "percentage": 15.04, "elapsed_time": "0:43:26", "remaining_time": "4:05:24"}
+{"current_steps": 324, "total_steps": 2148, "loss": 0.7102, "lr": 9.923173996732058e-06, "epoch": 0.4519879097884213, "percentage": 15.08, "elapsed_time": "0:43:33", "remaining_time": "4:05:13"}
+{"current_steps": 325, "total_steps": 2148, "loss": 0.7588, "lr": 9.921748446130564e-06, "epoch": 0.4533829342013485, "percentage": 15.13, "elapsed_time": "0:43:43", "remaining_time": "4:05:13"}
+{"current_steps": 326, "total_steps": 2148, "loss": 0.7111, "lr": 9.920309895169793e-06, "epoch": 0.4547779586142757, "percentage": 15.18, "elapsed_time": "0:43:49", "remaining_time": "4:04:56"}
+{"current_steps": 327, "total_steps": 2148, "loss": 0.7137, "lr": 9.91885834764955e-06, "epoch": 0.456172983027203, "percentage": 15.22, "elapsed_time": "0:43:58", "remaining_time": "4:04:51"}
+{"current_steps": 328, "total_steps": 2148, "loss": 0.7683, "lr": 9.917393807403965e-06, "epoch": 0.4575680074401302, "percentage": 15.27, "elapsed_time": "0:44:04", "remaining_time": "4:04:33"}
+{"current_steps": 329, "total_steps": 2148, "loss": 0.7056, "lr": 9.915916278301496e-06, "epoch": 0.4589630318530574, "percentage": 15.32, "elapsed_time": "0:44:13", "remaining_time": "4:04:32"}
+{"current_steps": 330, "total_steps": 2148, "loss": 0.6896, "lr": 9.9144257642449e-06, "epoch": 0.46035805626598464, "percentage": 15.36, "elapsed_time": "0:44:20", "remaining_time": "4:04:18"}
+{"current_steps": 331, "total_steps": 2148, "loss": 0.6562, "lr": 9.91292226917124e-06, "epoch": 0.46175308067891185, "percentage": 15.41, "elapsed_time": "0:44:27", "remaining_time": "4:04:04"}
+{"current_steps": 332, "total_steps": 2148, "loss": 0.7529, "lr": 9.91140579705186e-06, "epoch": 0.4631481050918391, "percentage": 15.46, "elapsed_time": "0:44:34", "remaining_time": "4:03:46"}
+{"current_steps": 333, "total_steps": 2148, "loss": 0.7675, "lr": 9.909876351892388e-06, "epoch": 0.46454312950476634, "percentage": 15.5, "elapsed_time": "0:44:41", "remaining_time": "4:03:33"}
+{"current_steps": 334, "total_steps": 2148, "loss": 0.7209, "lr": 9.908333937732718e-06, "epoch": 0.46593815391769355, "percentage": 15.55, "elapsed_time": "0:44:48", "remaining_time": "4:03:19"}
+{"current_steps": 335, "total_steps": 2148, "loss": 0.71, "lr": 9.906778558647e-06, "epoch": 0.46733317833062077, "percentage": 15.6, "elapsed_time": "0:44:57", "remaining_time": "4:03:19"}
+{"current_steps": 336, "total_steps": 2148, "loss": 0.7266, "lr": 9.905210218743626e-06, "epoch": 0.46872820274354804, "percentage": 15.64, "elapsed_time": "0:45:03", "remaining_time": "4:03:00"}
+{"current_steps": 337, "total_steps": 2148, "loss": 0.7023, "lr": 9.903628922165227e-06, "epoch": 0.47012322715647525, "percentage": 15.69, "elapsed_time": "0:45:09", "remaining_time": "4:02:42"}
+{"current_steps": 338, "total_steps": 2148, "loss": 0.7472, "lr": 9.902034673088656e-06, "epoch": 0.47151825156940247, "percentage": 15.74, "elapsed_time": "0:45:16", "remaining_time": "4:02:29"}
+{"current_steps": 339, "total_steps": 2148, "loss": 0.6827, "lr": 9.90042747572498e-06, "epoch": 0.4729132759823297, "percentage": 15.78, "elapsed_time": "0:45:23", "remaining_time": "4:02:13"}
+{"current_steps": 340, "total_steps": 2148, "loss": 0.7682, "lr": 9.898807334319471e-06, "epoch": 0.4743083003952569, "percentage": 15.83, "elapsed_time": "0:45:33", "remaining_time": "4:02:18"}
+{"current_steps": 341, "total_steps": 2148, "loss": 0.7217, "lr": 9.897174253151583e-06, "epoch": 0.47570332480818417, "percentage": 15.88, "elapsed_time": "0:45:44", "remaining_time": "4:02:21"}
+{"current_steps": 342, "total_steps": 2148, "loss": 0.7414, "lr": 9.895528236534957e-06, "epoch": 0.4770983492211114, "percentage": 15.92, "elapsed_time": "0:45:50", "remaining_time": "4:02:04"}
+{"current_steps": 343, "total_steps": 2148, "loss": 0.7902, "lr": 9.893869288817397e-06, "epoch": 0.4784933736340386, "percentage": 15.97, "elapsed_time": "0:46:00", "remaining_time": "4:02:07"}
+{"current_steps": 344, "total_steps": 2148, "loss": 0.7133, "lr": 9.89219741438087e-06, "epoch": 0.4798883980469658, "percentage": 16.01, "elapsed_time": "0:46:06", "remaining_time": "4:01:49"}
+{"current_steps": 345, "total_steps": 2148, "loss": 0.7315, "lr": 9.890512617641474e-06, "epoch": 0.48128342245989303, "percentage": 16.06, "elapsed_time": "0:46:15", "remaining_time": "4:01:47"}
+{"current_steps": 346, "total_steps": 2148, "loss": 0.7551, "lr": 9.888814903049458e-06, "epoch": 0.4826784468728203, "percentage": 16.11, "elapsed_time": "0:46:26", "remaining_time": "4:01:50"}
+{"current_steps": 347, "total_steps": 2148, "loss": 0.7701, "lr": 9.88710427508918e-06, "epoch": 0.4840734712857475, "percentage": 16.15, "elapsed_time": "0:46:35", "remaining_time": "4:01:49"}
+{"current_steps": 348, "total_steps": 2148, "loss": 0.6928, "lr": 9.885380738279111e-06, "epoch": 0.48546849569867473, "percentage": 16.2, "elapsed_time": "0:46:44", "remaining_time": "4:01:45"}
+{"current_steps": 349, "total_steps": 2148, "loss": 0.7264, "lr": 9.883644297171821e-06, "epoch": 0.48686352011160194, "percentage": 16.25, "elapsed_time": "0:46:51", "remaining_time": "4:01:34"}
+{"current_steps": 350, "total_steps": 2148, "loss": 0.7533, "lr": 9.881894956353963e-06, "epoch": 0.48825854452452916, "percentage": 16.29, "elapsed_time": "0:47:01", "remaining_time": "4:01:34"}
+{"current_steps": 351, "total_steps": 2148, "loss": 0.7176, "lr": 9.880132720446265e-06, "epoch": 0.48965356893745643, "percentage": 16.34, "elapsed_time": "0:47:10", "remaining_time": "4:01:33"}
+{"current_steps": 352, "total_steps": 2148, "loss": 0.7732, "lr": 9.878357594103516e-06, "epoch": 0.49104859335038364, "percentage": 16.39, "elapsed_time": "0:47:17", "remaining_time": "4:01:17"}
+{"current_steps": 353, "total_steps": 2148, "loss": 0.7192, "lr": 9.876569582014554e-06, "epoch": 0.49244361776331086, "percentage": 16.43, "elapsed_time": "0:47:23", "remaining_time": "4:00:59"}
+{"current_steps": 354, "total_steps": 2148, "loss": 0.7583, "lr": 9.874768688902252e-06, "epoch": 0.4938386421762381, "percentage": 16.48, "elapsed_time": "0:47:30", "remaining_time": "4:00:46"}
+{"current_steps": 355, "total_steps": 2148, "loss": 0.7543, "lr": 9.87295491952351e-06, "epoch": 0.4952336665891653, "percentage": 16.53, "elapsed_time": "0:47:37", "remaining_time": "4:00:33"}
+{"current_steps": 356, "total_steps": 2148, "loss": 0.7254, "lr": 9.871128278669238e-06, "epoch": 0.49662869100209256, "percentage": 16.57, "elapsed_time": "0:47:44", "remaining_time": "4:00:18"}
+{"current_steps": 357, "total_steps": 2148, "loss": 0.7201, "lr": 9.869288771164344e-06, "epoch": 0.4980237154150198, "percentage": 16.62, "elapsed_time": "0:47:52", "remaining_time": "4:00:09"}
+{"current_steps": 358, "total_steps": 2148, "loss": 0.7316, "lr": 9.867436401867723e-06, "epoch": 0.499418739827947, "percentage": 16.67, "elapsed_time": "0:48:00", "remaining_time": "4:00:03"}
+{"current_steps": 359, "total_steps": 2148, "loss": 0.7659, "lr": 9.865571175672245e-06, "epoch": 0.5008137642408742, "percentage": 16.71, "elapsed_time": "0:48:08", "remaining_time": "3:59:54"}
+{"current_steps": 360, "total_steps": 2148, "loss": 0.7098, "lr": 9.863693097504733e-06, "epoch": 0.5022087886538015, "percentage": 16.76, "elapsed_time": "0:48:15", "remaining_time": "3:59:39"}
+{"current_steps": 361, "total_steps": 2148, "loss": 0.7726, "lr": 9.86180217232597e-06, "epoch": 0.5036038130667286, "percentage": 16.81, "elapsed_time": "0:48:21", "remaining_time": "3:59:24"}
+{"current_steps": 362, "total_steps": 2148, "loss": 0.7916, "lr": 9.859898405130661e-06, "epoch": 0.5049988374796559, "percentage": 16.85, "elapsed_time": "0:48:28", "remaining_time": "3:59:09"}
+{"current_steps": 363, "total_steps": 2148, "loss": 0.7118, "lr": 9.85798180094744e-06, "epoch": 0.5063938618925832, "percentage": 16.9, "elapsed_time": "0:48:38", "remaining_time": "3:59:08"}
+{"current_steps": 364, "total_steps": 2148, "loss": 0.8073, "lr": 9.856052364838846e-06, "epoch": 0.5077888863055103, "percentage": 16.95, "elapsed_time": "0:48:50", "remaining_time": "3:59:21"}
+{"current_steps": 365, "total_steps": 2148, "loss": 0.6749, "lr": 9.854110101901308e-06, "epoch": 0.5091839107184376, "percentage": 16.99, "elapsed_time": "0:49:01", "remaining_time": "3:59:30"}
+{"current_steps": 366, "total_steps": 2148, "loss": 0.6283, "lr": 9.852155017265146e-06, "epoch": 0.5105789351313648, "percentage": 17.04, "elapsed_time": "0:49:08", "remaining_time": "3:59:15"}
+{"current_steps": 367, "total_steps": 2148, "loss": 0.721, "lr": 9.850187116094538e-06, "epoch": 0.511973959544292, "percentage": 17.09, "elapsed_time": "0:49:15", "remaining_time": "3:59:02"}
+{"current_steps": 368, "total_steps": 2148, "loss": 0.7634, "lr": 9.848206403587521e-06, "epoch": 0.5133689839572193, "percentage": 17.13, "elapsed_time": "0:49:26", "remaining_time": "3:59:09"}
+{"current_steps": 369, "total_steps": 2148, "loss": 0.7562, "lr": 9.84621288497597e-06, "epoch": 0.5147640083701465, "percentage": 17.18, "elapsed_time": "0:49:36", "remaining_time": "3:59:10"}
+{"current_steps": 370, "total_steps": 2148, "loss": 0.7495, "lr": 9.844206565525585e-06, "epoch": 0.5161590327830737, "percentage": 17.23, "elapsed_time": "0:49:44", "remaining_time": "3:59:00"}
+{"current_steps": 371, "total_steps": 2148, "loss": 0.731, "lr": 9.842187450535881e-06, "epoch": 0.5175540571960009, "percentage": 17.27, "elapsed_time": "0:49:52", "remaining_time": "3:58:51"}
+{"current_steps": 372, "total_steps": 2148, "loss": 0.7457, "lr": 9.840155545340169e-06, "epoch": 0.5189490816089282, "percentage": 17.32, "elapsed_time": "0:49:58", "remaining_time": "3:58:33"}
+{"current_steps": 373, "total_steps": 2148, "loss": 0.7103, "lr": 9.838110855305548e-06, "epoch": 0.5203441060218554, "percentage": 17.36, "elapsed_time": "0:50:08", "remaining_time": "3:58:35"}
+{"current_steps": 374, "total_steps": 2148, "loss": 0.6503, "lr": 9.836053385832881e-06, "epoch": 0.5217391304347826, "percentage": 17.41, "elapsed_time": "0:50:18", "remaining_time": "3:58:35"}
+{"current_steps": 375, "total_steps": 2148, "loss": 0.7281, "lr": 9.833983142356792e-06, "epoch": 0.5231341548477099, "percentage": 17.46, "elapsed_time": "0:50:24", "remaining_time": "3:58:20"}
+{"current_steps": 376, "total_steps": 2148, "loss": 0.7009, "lr": 9.831900130345645e-06, "epoch": 0.524529179260637, "percentage": 17.5, "elapsed_time": "0:50:32", "remaining_time": "3:58:11"}
+{"current_steps": 377, "total_steps": 2148, "loss": 0.6762, "lr": 9.829804355301527e-06, "epoch": 0.5259242036735643, "percentage": 17.55, "elapsed_time": "0:50:38", "remaining_time": "3:57:55"}
+{"current_steps": 378, "total_steps": 2148, "loss": 0.6789, "lr": 9.827695822760245e-06, "epoch": 0.5273192280864916, "percentage": 17.6, "elapsed_time": "0:50:46", "remaining_time": "3:57:46"}
+{"current_steps": 379, "total_steps": 2148, "loss": 0.7384, "lr": 9.825574538291293e-06, "epoch": 0.5287142524994187, "percentage": 17.64, "elapsed_time": "0:50:53", "remaining_time": "3:57:33"}
+{"current_steps": 380, "total_steps": 2148, "loss": 0.6774, "lr": 9.823440507497863e-06, "epoch": 0.530109276912346, "percentage": 17.69, "elapsed_time": "0:51:00", "remaining_time": "3:57:18"}
+{"current_steps": 381, "total_steps": 2148, "loss": 0.6933, "lr": 9.821293736016802e-06, "epoch": 0.5315043013252732, "percentage": 17.74, "elapsed_time": "0:51:08", "remaining_time": "3:57:12"}
+{"current_steps": 382, "total_steps": 2148, "loss": 0.7449, "lr": 9.819134229518617e-06, "epoch": 0.5328993257382004, "percentage": 17.78, "elapsed_time": "0:51:17", "remaining_time": "3:57:08"}
+{"current_steps": 383, "total_steps": 2148, "loss": 0.7213, "lr": 9.81696199370745e-06, "epoch": 0.5342943501511277, "percentage": 17.83, "elapsed_time": "0:51:24", "remaining_time": "3:56:56"}
+{"current_steps": 384, "total_steps": 2148, "loss": 0.7453, "lr": 9.814777034321069e-06, "epoch": 0.5356893745640549, "percentage": 17.88, "elapsed_time": "0:51:33", "remaining_time": "3:56:49"}
+{"current_steps": 385, "total_steps": 2148, "loss": 0.7702, "lr": 9.812579357130848e-06, "epoch": 0.5370843989769821, "percentage": 17.92, "elapsed_time": "0:51:40", "remaining_time": "3:56:36"}
+{"current_steps": 386, "total_steps": 2148, "loss": 0.7129, "lr": 9.810368967941757e-06, "epoch": 0.5384794233899093, "percentage": 17.97, "elapsed_time": "0:51:46", "remaining_time": "3:56:20"}
+{"current_steps": 387, "total_steps": 2148, "loss": 0.7192, "lr": 9.808145872592341e-06, "epoch": 0.5398744478028366, "percentage": 18.02, "elapsed_time": "0:51:52", "remaining_time": "3:56:03"}
+{"current_steps": 388, "total_steps": 2148, "loss": 0.689, "lr": 9.80591007695471e-06, "epoch": 0.5412694722157638, "percentage": 18.06, "elapsed_time": "0:52:02", "remaining_time": "3:56:06"}
+{"current_steps": 389, "total_steps": 2148, "loss": 0.7041, "lr": 9.803661586934514e-06, "epoch": 0.542664496628691, "percentage": 18.11, "elapsed_time": "0:52:10", "remaining_time": "3:55:55"}
+{"current_steps": 390, "total_steps": 2148, "loss": 0.7558, "lr": 9.801400408470943e-06, "epoch": 0.5440595210416183, "percentage": 18.16, "elapsed_time": "0:52:19", "remaining_time": "3:55:49"}
+{"current_steps": 391, "total_steps": 2148, "loss": 0.7542, "lr": 9.799126547536695e-06, "epoch": 0.5454545454545454, "percentage": 18.2, "elapsed_time": "0:52:27", "remaining_time": "3:55:41"}
+{"current_steps": 392, "total_steps": 2148, "loss": 0.7036, "lr": 9.796840010137972e-06, "epoch": 0.5468495698674727, "percentage": 18.25, "elapsed_time": "0:52:33", "remaining_time": "3:55:26"}
+{"current_steps": 393, "total_steps": 2148, "loss": 0.8377, "lr": 9.79454080231446e-06, "epoch": 0.5482445942804, "percentage": 18.3, "elapsed_time": "0:52:42", "remaining_time": "3:55:23"}
+{"current_steps": 394, "total_steps": 2148, "loss": 0.6456, "lr": 9.79222893013931e-06, "epoch": 0.5496396186933271, "percentage": 18.34, "elapsed_time": "0:52:49", "remaining_time": "3:55:09"}
+{"current_steps": 395, "total_steps": 2148, "loss": 0.7739, "lr": 9.789904399719124e-06, "epoch": 0.5510346431062544, "percentage": 18.39, "elapsed_time": "0:52:56", "remaining_time": "3:54:56"}
+{"current_steps": 396, "total_steps": 2148, "loss": 0.6688, "lr": 9.787567217193944e-06, "epoch": 0.5524296675191815, "percentage": 18.44, "elapsed_time": "0:53:03", "remaining_time": "3:54:45"}
+{"current_steps": 397, "total_steps": 2148, "loss": 0.7427, "lr": 9.785217388737232e-06, "epoch": 0.5538246919321088, "percentage": 18.48, "elapsed_time": "0:53:10", "remaining_time": "3:54:33"}
+{"current_steps": 398, "total_steps": 2148, "loss": 0.6962, "lr": 9.782854920555844e-06, "epoch": 0.5552197163450361, "percentage": 18.53, "elapsed_time": "0:53:18", "remaining_time": "3:54:24"}
+{"current_steps": 399, "total_steps": 2148, "loss": 0.7644, "lr": 9.780479818890032e-06, "epoch": 0.5566147407579632, "percentage": 18.58, "elapsed_time": "0:53:27", "remaining_time": "3:54:20"}
+{"current_steps": 400, "total_steps": 2148, "loss": 0.7137, "lr": 9.778092090013416e-06, "epoch": 0.5580097651708905, "percentage": 18.62, "elapsed_time": "0:53:34", "remaining_time": "3:54:05"}
+{"current_steps": 401, "total_steps": 2148, "loss": 0.6742, "lr": 9.775691740232966e-06, "epoch": 0.5594047895838177, "percentage": 18.67, "elapsed_time": "0:55:17", "remaining_time": "4:00:51"}
+{"current_steps": 402, "total_steps": 2148, "loss": 0.7475, "lr": 9.773278775888995e-06, "epoch": 0.560799813996745, "percentage": 18.72, "elapsed_time": "0:55:28", "remaining_time": "4:00:55"}
+{"current_steps": 403, "total_steps": 2148, "loss": 0.7055, "lr": 9.77085320335513e-06, "epoch": 0.5621948384096722, "percentage": 18.76, "elapsed_time": "0:55:36", "remaining_time": "4:00:45"}
+{"current_steps": 404, "total_steps": 2148, "loss": 0.7101, "lr": 9.768415029038304e-06, "epoch": 0.5635898628225994, "percentage": 18.81, "elapsed_time": "0:55:43", "remaining_time": "4:00:34"}
+{"current_steps": 405, "total_steps": 2148, "loss": 0.6877, "lr": 9.76596425937874e-06, "epoch": 0.5649848872355266, "percentage": 18.85, "elapsed_time": "0:55:55", "remaining_time": "4:00:40"}
+{"current_steps": 406, "total_steps": 2148, "loss": 0.6662, "lr": 9.763500900849926e-06, "epoch": 0.5663799116484538, "percentage": 18.9, "elapsed_time": "0:56:01", "remaining_time": "4:00:24"}
+{"current_steps": 407, "total_steps": 2148, "loss": 0.7489, "lr": 9.761024959958605e-06, "epoch": 0.5677749360613811, "percentage": 18.95, "elapsed_time": "0:56:11", "remaining_time": "4:00:23"}
+{"current_steps": 408, "total_steps": 2148, "loss": 0.7565, "lr": 9.75853644324475e-06, "epoch": 0.5691699604743083, "percentage": 18.99, "elapsed_time": "0:56:20", "remaining_time": "4:00:15"}
+{"current_steps": 409, "total_steps": 2148, "loss": 0.7399, "lr": 9.756035357281559e-06, "epoch": 0.5705649848872355, "percentage": 19.04, "elapsed_time": "0:56:29", "remaining_time": "4:00:10"}
+{"current_steps": 410, "total_steps": 2148, "loss": 0.7035, "lr": 9.753521708675426e-06, "epoch": 0.5719600093001628, "percentage": 19.09, "elapsed_time": "0:56:37", "remaining_time": "4:00:04"}
+{"current_steps": 411, "total_steps": 2148, "loss": 0.7104, "lr": 9.75099550406593e-06, "epoch": 0.5733550337130899, "percentage": 19.13, "elapsed_time": "0:56:45", "remaining_time": "3:59:51"}
+{"current_steps": 412, "total_steps": 2148, "loss": 0.755, "lr": 9.748456750125817e-06, "epoch": 0.5747500581260172, "percentage": 19.18, "elapsed_time": "0:56:51", "remaining_time": "3:59:36"}
+{"current_steps": 413, "total_steps": 2148, "loss": 0.7055, "lr": 9.745905453560976e-06, "epoch": 0.5761450825389445, "percentage": 19.23, "elapsed_time": "0:57:00", "remaining_time": "3:59:30"}
+{"current_steps": 414, "total_steps": 2148, "loss": 0.7452, "lr": 9.74334162111043e-06, "epoch": 0.5775401069518716, "percentage": 19.27, "elapsed_time": "0:57:07", "remaining_time": "3:59:17"}
+{"current_steps": 415, "total_steps": 2148, "loss": 0.7233, "lr": 9.740765259546312e-06, "epoch": 0.5789351313647989, "percentage": 19.32, "elapsed_time": "0:57:16", "remaining_time": "3:59:12"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:725a4f13dfab58ff48947467cae251916c69da74b76765b647f2e09915028895
+size 8376