Training in progress, epoch 0

.gitattributes

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

added_tokens.json

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

config.json

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

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9594daf2a8aeebda470b10f2a3a42f9ae06a2e46d549ab18dcdc4d10abc2283
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fea1840910a8b6883f1cff81f8d512bbc9cf36c55ed53e7a920c2c331a97bfcf
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57ca05cbbea9cbdeecfff35eddc5fd99d7ca73a0622e409e79ad464e86b3af44
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d89dfa1863f6a3b78b5ce7639411d6cdb9c73bd29c59c7461ae1cb764b018e95
+size 1089994880

model.safetensors.index.json

@@ -0,0 +1,346 @@
+{
+  "metadata": {
+    "total_size": 15231233024
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00003-of-00004.safetensors"
+  }
+}

special_tokens_map.json

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

tokenizer.json

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

tokenizer_config.json

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

trainer_log.jsonl

@@ -0,0 +1,484 @@
+{"current_steps": 1, "total_steps": 2405, "loss": 1.0238, "lr": 3.319502074688797e-07, "epoch": 0.0020752269779507134, "percentage": 0.04, "elapsed_time": "0:01:03", "remaining_time": "1 day, 18:29:50"}
+{"current_steps": 2, "total_steps": 2405, "loss": 1.019, "lr": 6.639004149377594e-07, "epoch": 0.004150453955901427, "percentage": 0.08, "elapsed_time": "0:01:57", "remaining_time": "1 day, 15:16:11"}
+{"current_steps": 3, "total_steps": 2405, "loss": 1.024, "lr": 9.95850622406639e-07, "epoch": 0.0062256809338521405, "percentage": 0.12, "elapsed_time": "0:02:52", "remaining_time": "1 day, 14:17:23"}
+{"current_steps": 4, "total_steps": 2405, "loss": 1.024, "lr": 1.3278008298755188e-06, "epoch": 0.008300907911802853, "percentage": 0.17, "elapsed_time": "0:03:45", "remaining_time": "1 day, 13:40:04"}
+{"current_steps": 5, "total_steps": 2405, "loss": 1.0102, "lr": 1.6597510373443984e-06, "epoch": 0.010376134889753566, "percentage": 0.21, "elapsed_time": "0:04:39", "remaining_time": "1 day, 13:17:37"}
+{"current_steps": 6, "total_steps": 2405, "loss": 1.0118, "lr": 1.991701244813278e-06, "epoch": 0.012451361867704281, "percentage": 0.25, "elapsed_time": "0:05:33", "remaining_time": "1 day, 13:02:26"}
+{"current_steps": 7, "total_steps": 2405, "loss": 0.9758, "lr": 2.323651452282158e-06, "epoch": 0.014526588845654994, "percentage": 0.29, "elapsed_time": "0:06:27", "remaining_time": "1 day, 12:51:55"}
+{"current_steps": 8, "total_steps": 2405, "loss": 0.9705, "lr": 2.6556016597510377e-06, "epoch": 0.016601815823605707, "percentage": 0.33, "elapsed_time": "0:07:21", "remaining_time": "1 day, 12:43:37"}
+{"current_steps": 9, "total_steps": 2405, "loss": 0.9208, "lr": 2.9875518672199173e-06, "epoch": 0.01867704280155642, "percentage": 0.37, "elapsed_time": "0:08:16", "remaining_time": "1 day, 12:41:39"}
+{"current_steps": 10, "total_steps": 2405, "loss": 0.9158, "lr": 3.319502074688797e-06, "epoch": 0.020752269779507133, "percentage": 0.42, "elapsed_time": "0:09:10", "remaining_time": "1 day, 12:37:11"}
+{"current_steps": 11, "total_steps": 2405, "loss": 0.9069, "lr": 3.6514522821576765e-06, "epoch": 0.022827496757457846, "percentage": 0.46, "elapsed_time": "0:10:04", "remaining_time": "1 day, 12:33:27"}
+{"current_steps": 12, "total_steps": 2405, "loss": 0.9053, "lr": 3.983402489626556e-06, "epoch": 0.024902723735408562, "percentage": 0.5, "elapsed_time": "0:10:58", "remaining_time": "1 day, 12:28:39"}
+{"current_steps": 13, "total_steps": 2405, "loss": 0.9004, "lr": 4.315352697095436e-06, "epoch": 0.026977950713359275, "percentage": 0.54, "elapsed_time": "0:11:52", "remaining_time": "1 day, 12:24:27"}
+{"current_steps": 14, "total_steps": 2405, "loss": 0.8938, "lr": 4.647302904564316e-06, "epoch": 0.029053177691309988, "percentage": 0.58, "elapsed_time": "0:12:46", "remaining_time": "1 day, 12:21:42"}
+{"current_steps": 15, "total_steps": 2405, "loss": 0.8786, "lr": 4.979253112033195e-06, "epoch": 0.0311284046692607, "percentage": 0.62, "elapsed_time": "0:13:40", "remaining_time": "1 day, 12:18:14"}
+{"current_steps": 16, "total_steps": 2405, "loss": 0.8314, "lr": 5.311203319502075e-06, "epoch": 0.033203631647211414, "percentage": 0.67, "elapsed_time": "0:14:34", "remaining_time": "1 day, 12:16:29"}
+{"current_steps": 17, "total_steps": 2405, "loss": 0.837, "lr": 5.6431535269709545e-06, "epoch": 0.03527885862516213, "percentage": 0.71, "elapsed_time": "0:15:29", "remaining_time": "1 day, 12:15:34"}
+{"current_steps": 18, "total_steps": 2405, "loss": 0.8125, "lr": 5.9751037344398345e-06, "epoch": 0.03735408560311284, "percentage": 0.75, "elapsed_time": "0:16:23", "remaining_time": "1 day, 12:13:51"}
+{"current_steps": 19, "total_steps": 2405, "loss": 0.7923, "lr": 6.307053941908714e-06, "epoch": 0.03942931258106355, "percentage": 0.79, "elapsed_time": "0:17:17", "remaining_time": "1 day, 12:12:13"}
+{"current_steps": 20, "total_steps": 2405, "loss": 0.7964, "lr": 6.639004149377594e-06, "epoch": 0.041504539559014265, "percentage": 0.83, "elapsed_time": "0:18:12", "remaining_time": "1 day, 12:11:35"}
+{"current_steps": 21, "total_steps": 2405, "loss": 0.7789, "lr": 6.970954356846473e-06, "epoch": 0.04357976653696498, "percentage": 0.87, "elapsed_time": "0:19:06", "remaining_time": "1 day, 12:09:39"}
+{"current_steps": 22, "total_steps": 2405, "loss": 0.7711, "lr": 7.302904564315353e-06, "epoch": 0.04565499351491569, "percentage": 0.91, "elapsed_time": "0:20:01", "remaining_time": "1 day, 12:08:39"}
+{"current_steps": 23, "total_steps": 2405, "loss": 0.7673, "lr": 7.634854771784232e-06, "epoch": 0.047730220492866404, "percentage": 0.96, "elapsed_time": "0:20:55", "remaining_time": "1 day, 12:07:13"}
+{"current_steps": 24, "total_steps": 2405, "loss": 0.7608, "lr": 7.966804979253112e-06, "epoch": 0.049805447470817124, "percentage": 1.0, "elapsed_time": "0:21:51", "remaining_time": "1 day, 12:08:55"}
+{"current_steps": 25, "total_steps": 2405, "loss": 0.7516, "lr": 8.298755186721992e-06, "epoch": 0.05188067444876784, "percentage": 1.04, "elapsed_time": "0:22:46", "remaining_time": "1 day, 12:08:07"}
+{"current_steps": 26, "total_steps": 2405, "loss": 0.739, "lr": 8.630705394190872e-06, "epoch": 0.05395590142671855, "percentage": 1.08, "elapsed_time": "0:23:40", "remaining_time": "1 day, 12:05:30"}
+{"current_steps": 27, "total_steps": 2405, "loss": 0.7265, "lr": 8.962655601659752e-06, "epoch": 0.05603112840466926, "percentage": 1.12, "elapsed_time": "0:24:33", "remaining_time": "1 day, 12:03:17"}
+{"current_steps": 28, "total_steps": 2405, "loss": 0.7358, "lr": 9.294605809128632e-06, "epoch": 0.058106355382619976, "percentage": 1.16, "elapsed_time": "0:25:28", "remaining_time": "1 day, 12:02:10"}
+{"current_steps": 29, "total_steps": 2405, "loss": 0.7358, "lr": 9.62655601659751e-06, "epoch": 0.06018158236057069, "percentage": 1.21, "elapsed_time": "0:26:22", "remaining_time": "1 day, 12:00:47"}
+{"current_steps": 30, "total_steps": 2405, "loss": 0.715, "lr": 9.95850622406639e-06, "epoch": 0.0622568093385214, "percentage": 1.25, "elapsed_time": "0:27:16", "remaining_time": "1 day, 11:59:40"}
+{"current_steps": 31, "total_steps": 2405, "loss": 0.7107, "lr": 1.0290456431535269e-05, "epoch": 0.06433203631647211, "percentage": 1.29, "elapsed_time": "0:28:11", "remaining_time": "1 day, 11:58:42"}
+{"current_steps": 32, "total_steps": 2405, "loss": 0.7031, "lr": 1.062240663900415e-05, "epoch": 0.06640726329442283, "percentage": 1.33, "elapsed_time": "0:29:05", "remaining_time": "1 day, 11:57:36"}
+{"current_steps": 33, "total_steps": 2405, "loss": 0.7099, "lr": 1.095435684647303e-05, "epoch": 0.06848249027237355, "percentage": 1.37, "elapsed_time": "0:30:00", "remaining_time": "1 day, 11:56:30"}
+{"current_steps": 34, "total_steps": 2405, "loss": 0.7006, "lr": 1.1286307053941909e-05, "epoch": 0.07055771725032425, "percentage": 1.41, "elapsed_time": "0:30:53", "remaining_time": "1 day, 11:54:41"}
+{"current_steps": 35, "total_steps": 2405, "loss": 0.7008, "lr": 1.161825726141079e-05, "epoch": 0.07263294422827497, "percentage": 1.46, "elapsed_time": "0:31:48", "remaining_time": "1 day, 11:53:24"}
+{"current_steps": 36, "total_steps": 2405, "loss": 0.6931, "lr": 1.1950207468879669e-05, "epoch": 0.07470817120622568, "percentage": 1.5, "elapsed_time": "0:32:42", "remaining_time": "1 day, 11:52:38"}
+{"current_steps": 37, "total_steps": 2405, "loss": 0.6887, "lr": 1.2282157676348549e-05, "epoch": 0.0767833981841764, "percentage": 1.54, "elapsed_time": "0:33:37", "remaining_time": "1 day, 11:52:01"}
+{"current_steps": 38, "total_steps": 2405, "loss": 0.6966, "lr": 1.2614107883817427e-05, "epoch": 0.0788586251621271, "percentage": 1.58, "elapsed_time": "0:34:31", "remaining_time": "1 day, 11:50:36"}
+{"current_steps": 39, "total_steps": 2405, "loss": 0.6824, "lr": 1.2946058091286309e-05, "epoch": 0.08093385214007782, "percentage": 1.62, "elapsed_time": "0:35:25", "remaining_time": "1 day, 11:49:00"}
+{"current_steps": 40, "total_steps": 2405, "loss": 0.6824, "lr": 1.3278008298755187e-05, "epoch": 0.08300907911802853, "percentage": 1.66, "elapsed_time": "0:36:19", "remaining_time": "1 day, 11:47:51"}
+{"current_steps": 41, "total_steps": 2405, "loss": 0.6774, "lr": 1.3609958506224067e-05, "epoch": 0.08508430609597925, "percentage": 1.7, "elapsed_time": "0:37:13", "remaining_time": "1 day, 11:46:11"}
+{"current_steps": 42, "total_steps": 2405, "loss": 0.6756, "lr": 1.3941908713692946e-05, "epoch": 0.08715953307392996, "percentage": 1.75, "elapsed_time": "0:38:06", "remaining_time": "1 day, 11:43:57"}
+{"current_steps": 43, "total_steps": 2405, "loss": 0.6683, "lr": 1.4273858921161828e-05, "epoch": 0.08923476005188068, "percentage": 1.79, "elapsed_time": "0:39:00", "remaining_time": "1 day, 11:42:25"}
+{"current_steps": 44, "total_steps": 2405, "loss": 0.6796, "lr": 1.4605809128630706e-05, "epoch": 0.09130998702983138, "percentage": 1.83, "elapsed_time": "0:39:53", "remaining_time": "1 day, 11:40:52"}
+{"current_steps": 45, "total_steps": 2405, "loss": 0.6674, "lr": 1.4937759336099586e-05, "epoch": 0.0933852140077821, "percentage": 1.87, "elapsed_time": "0:40:47", "remaining_time": "1 day, 11:39:12"}
+{"current_steps": 46, "total_steps": 2405, "loss": 0.6694, "lr": 1.5269709543568464e-05, "epoch": 0.09546044098573281, "percentage": 1.91, "elapsed_time": "0:41:41", "remaining_time": "1 day, 11:37:38"}
+{"current_steps": 47, "total_steps": 2405, "loss": 0.6561, "lr": 1.5601659751037346e-05, "epoch": 0.09753566796368353, "percentage": 1.95, "elapsed_time": "0:42:35", "remaining_time": "1 day, 11:37:11"}
+{"current_steps": 48, "total_steps": 2405, "loss": 0.6511, "lr": 1.5933609958506224e-05, "epoch": 0.09961089494163425, "percentage": 2.0, "elapsed_time": "0:43:29", "remaining_time": "1 day, 11:35:44"}
+{"current_steps": 49, "total_steps": 2405, "loss": 0.6537, "lr": 1.6265560165975106e-05, "epoch": 0.10168612191958495, "percentage": 2.04, "elapsed_time": "0:44:23", "remaining_time": "1 day, 11:34:34"}
+{"current_steps": 50, "total_steps": 2405, "loss": 0.6558, "lr": 1.6597510373443984e-05, "epoch": 0.10376134889753567, "percentage": 2.08, "elapsed_time": "0:45:17", "remaining_time": "1 day, 11:33:04"}
+{"current_steps": 51, "total_steps": 2405, "loss": 0.6547, "lr": 1.6929460580912863e-05, "epoch": 0.10583657587548638, "percentage": 2.12, "elapsed_time": "0:46:10", "remaining_time": "1 day, 11:31:36"}
+{"current_steps": 52, "total_steps": 2405, "loss": 0.651, "lr": 1.7261410788381744e-05, "epoch": 0.1079118028534371, "percentage": 2.16, "elapsed_time": "0:47:05", "remaining_time": "1 day, 11:30:57"}
+{"current_steps": 53, "total_steps": 2405, "loss": 0.6511, "lr": 1.7593360995850623e-05, "epoch": 0.1099870298313878, "percentage": 2.2, "elapsed_time": "0:48:00", "remaining_time": "1 day, 11:30:29"}
+{"current_steps": 54, "total_steps": 2405, "loss": 0.6454, "lr": 1.7925311203319504e-05, "epoch": 0.11206225680933853, "percentage": 2.25, "elapsed_time": "0:48:54", "remaining_time": "1 day, 11:29:34"}
+{"current_steps": 55, "total_steps": 2405, "loss": 0.6526, "lr": 1.8257261410788383e-05, "epoch": 0.11413748378728923, "percentage": 2.29, "elapsed_time": "0:49:49", "remaining_time": "1 day, 11:29:10"}
+{"current_steps": 56, "total_steps": 2405, "loss": 0.6502, "lr": 1.8589211618257264e-05, "epoch": 0.11621271076523995, "percentage": 2.33, "elapsed_time": "0:50:43", "remaining_time": "1 day, 11:27:52"}
+{"current_steps": 57, "total_steps": 2405, "loss": 0.6526, "lr": 1.8921161825726143e-05, "epoch": 0.11828793774319066, "percentage": 2.37, "elapsed_time": "0:51:37", "remaining_time": "1 day, 11:26:15"}
+{"current_steps": 58, "total_steps": 2405, "loss": 0.6441, "lr": 1.925311203319502e-05, "epoch": 0.12036316472114138, "percentage": 2.41, "elapsed_time": "0:52:31", "remaining_time": "1 day, 11:25:25"}
+{"current_steps": 59, "total_steps": 2405, "loss": 0.6518, "lr": 1.9585062240663903e-05, "epoch": 0.12243839169909208, "percentage": 2.45, "elapsed_time": "0:53:25", "remaining_time": "1 day, 11:24:22"}
+{"current_steps": 60, "total_steps": 2405, "loss": 0.6524, "lr": 1.991701244813278e-05, "epoch": 0.1245136186770428, "percentage": 2.49, "elapsed_time": "0:54:19", "remaining_time": "1 day, 11:23:19"}
+{"current_steps": 61, "total_steps": 2405, "loss": 0.6419, "lr": 2.024896265560166e-05, "epoch": 0.1265888456549935, "percentage": 2.54, "elapsed_time": "0:55:13", "remaining_time": "1 day, 11:22:04"}
+{"current_steps": 62, "total_steps": 2405, "loss": 0.6552, "lr": 2.0580912863070538e-05, "epoch": 0.12866407263294422, "percentage": 2.58, "elapsed_time": "0:56:07", "remaining_time": "1 day, 11:21:10"}
+{"current_steps": 63, "total_steps": 2405, "loss": 0.6426, "lr": 2.0912863070539423e-05, "epoch": 0.13073929961089495, "percentage": 2.62, "elapsed_time": "0:57:01", "remaining_time": "1 day, 11:19:57"}
+{"current_steps": 64, "total_steps": 2405, "loss": 0.6395, "lr": 2.12448132780083e-05, "epoch": 0.13281452658884565, "percentage": 2.66, "elapsed_time": "0:57:55", "remaining_time": "1 day, 11:18:35"}
+{"current_steps": 65, "total_steps": 2405, "loss": 0.6364, "lr": 2.157676348547718e-05, "epoch": 0.13488975356679636, "percentage": 2.7, "elapsed_time": "0:58:49", "remaining_time": "1 day, 11:17:26"}
+{"current_steps": 66, "total_steps": 2405, "loss": 0.6358, "lr": 2.190871369294606e-05, "epoch": 0.1369649805447471, "percentage": 2.74, "elapsed_time": "0:59:42", "remaining_time": "1 day, 11:16:17"}
+{"current_steps": 67, "total_steps": 2405, "loss": 0.6379, "lr": 2.224066390041494e-05, "epoch": 0.1390402075226978, "percentage": 2.79, "elapsed_time": "1:00:37", "remaining_time": "1 day, 11:15:18"}
+{"current_steps": 68, "total_steps": 2405, "loss": 0.6344, "lr": 2.2572614107883818e-05, "epoch": 0.1411154345006485, "percentage": 2.83, "elapsed_time": "1:01:30", "remaining_time": "1 day, 11:14:03"}
+{"current_steps": 69, "total_steps": 2405, "loss": 0.6345, "lr": 2.2904564315352696e-05, "epoch": 0.1431906614785992, "percentage": 2.87, "elapsed_time": "1:02:25", "remaining_time": "1 day, 11:13:17"}
+{"current_steps": 70, "total_steps": 2405, "loss": 0.6351, "lr": 2.323651452282158e-05, "epoch": 0.14526588845654995, "percentage": 2.91, "elapsed_time": "1:03:18", "remaining_time": "1 day, 11:11:59"}
+{"current_steps": 71, "total_steps": 2405, "loss": 0.6294, "lr": 2.356846473029046e-05, "epoch": 0.14734111543450065, "percentage": 2.95, "elapsed_time": "1:04:13", "remaining_time": "1 day, 11:11:10"}
+{"current_steps": 72, "total_steps": 2405, "loss": 0.6311, "lr": 2.3900414937759338e-05, "epoch": 0.14941634241245136, "percentage": 2.99, "elapsed_time": "1:05:07", "remaining_time": "1 day, 11:10:15"}
+{"current_steps": 73, "total_steps": 2405, "loss": 0.634, "lr": 2.4232365145228216e-05, "epoch": 0.15149156939040206, "percentage": 3.04, "elapsed_time": "1:06:01", "remaining_time": "1 day, 11:08:55"}
+{"current_steps": 74, "total_steps": 2405, "loss": 0.6371, "lr": 2.4564315352697098e-05, "epoch": 0.1535667963683528, "percentage": 3.08, "elapsed_time": "1:06:58", "remaining_time": "1 day, 11:09:42"}
+{"current_steps": 75, "total_steps": 2405, "loss": 0.6319, "lr": 2.4896265560165977e-05, "epoch": 0.1556420233463035, "percentage": 3.12, "elapsed_time": "1:07:54", "remaining_time": "1 day, 11:09:54"}
+{"current_steps": 76, "total_steps": 2405, "loss": 0.6309, "lr": 2.5228215767634855e-05, "epoch": 0.1577172503242542, "percentage": 3.16, "elapsed_time": "1:08:50", "remaining_time": "1 day, 11:09:25"}
+{"current_steps": 77, "total_steps": 2405, "loss": 0.6325, "lr": 2.5560165975103733e-05, "epoch": 0.15979247730220492, "percentage": 3.2, "elapsed_time": "1:09:45", "remaining_time": "1 day, 11:09:07"}
+{"current_steps": 78, "total_steps": 2405, "loss": 0.6347, "lr": 2.5892116182572618e-05, "epoch": 0.16186770428015565, "percentage": 3.24, "elapsed_time": "1:10:39", "remaining_time": "1 day, 11:08:01"}
+{"current_steps": 79, "total_steps": 2405, "loss": 0.6294, "lr": 2.6224066390041497e-05, "epoch": 0.16394293125810636, "percentage": 3.28, "elapsed_time": "1:11:32", "remaining_time": "1 day, 11:06:34"}
+{"current_steps": 80, "total_steps": 2405, "loss": 0.6294, "lr": 2.6556016597510375e-05, "epoch": 0.16601815823605706, "percentage": 3.33, "elapsed_time": "1:12:26", "remaining_time": "1 day, 11:05:10"}
+{"current_steps": 81, "total_steps": 2405, "loss": 0.6279, "lr": 2.6887966804979257e-05, "epoch": 0.1680933852140078, "percentage": 3.37, "elapsed_time": "1:13:19", "remaining_time": "1 day, 11:03:59"}
+{"current_steps": 82, "total_steps": 2405, "loss": 0.6308, "lr": 2.7219917012448135e-05, "epoch": 0.1701686121919585, "percentage": 3.41, "elapsed_time": "1:14:15", "remaining_time": "1 day, 11:03:31"}
+{"current_steps": 83, "total_steps": 2405, "loss": 0.6285, "lr": 2.7551867219917013e-05, "epoch": 0.1722438391699092, "percentage": 3.45, "elapsed_time": "1:15:08", "remaining_time": "1 day, 11:02:18"}
+{"current_steps": 84, "total_steps": 2405, "loss": 0.6244, "lr": 2.788381742738589e-05, "epoch": 0.1743190661478599, "percentage": 3.49, "elapsed_time": "1:16:03", "remaining_time": "1 day, 11:01:32"}
+{"current_steps": 85, "total_steps": 2405, "loss": 0.6163, "lr": 2.8215767634854777e-05, "epoch": 0.17639429312581065, "percentage": 3.53, "elapsed_time": "1:16:58", "remaining_time": "1 day, 11:00:46"}
+{"current_steps": 86, "total_steps": 2405, "loss": 0.6268, "lr": 2.8547717842323655e-05, "epoch": 0.17846952010376135, "percentage": 3.58, "elapsed_time": "1:17:52", "remaining_time": "1 day, 10:59:58"}
+{"current_steps": 87, "total_steps": 2405, "loss": 0.6232, "lr": 2.8879668049792533e-05, "epoch": 0.18054474708171206, "percentage": 3.62, "elapsed_time": "1:18:47", "remaining_time": "1 day, 10:59:10"}
+{"current_steps": 88, "total_steps": 2405, "loss": 0.619, "lr": 2.9211618257261412e-05, "epoch": 0.18261997405966277, "percentage": 3.66, "elapsed_time": "1:19:41", "remaining_time": "1 day, 10:58:24"}
+{"current_steps": 89, "total_steps": 2405, "loss": 0.6264, "lr": 2.9543568464730293e-05, "epoch": 0.1846952010376135, "percentage": 3.7, "elapsed_time": "1:20:36", "remaining_time": "1 day, 10:57:25"}
+{"current_steps": 90, "total_steps": 2405, "loss": 0.6166, "lr": 2.9875518672199172e-05, "epoch": 0.1867704280155642, "percentage": 3.74, "elapsed_time": "1:21:29", "remaining_time": "1 day, 10:56:06"}
+{"current_steps": 91, "total_steps": 2405, "loss": 0.619, "lr": 3.020746887966805e-05, "epoch": 0.1888456549935149, "percentage": 3.78, "elapsed_time": "1:22:23", "remaining_time": "1 day, 10:54:58"}
+{"current_steps": 92, "total_steps": 2405, "loss": 0.6155, "lr": 3.053941908713693e-05, "epoch": 0.19092088197146562, "percentage": 3.83, "elapsed_time": "1:23:18", "remaining_time": "1 day, 10:54:30"}
+{"current_steps": 93, "total_steps": 2405, "loss": 0.619, "lr": 3.087136929460581e-05, "epoch": 0.19299610894941635, "percentage": 3.87, "elapsed_time": "1:24:12", "remaining_time": "1 day, 10:53:19"}
+{"current_steps": 94, "total_steps": 2405, "loss": 0.6184, "lr": 3.120331950207469e-05, "epoch": 0.19507133592736706, "percentage": 3.91, "elapsed_time": "1:25:06", "remaining_time": "1 day, 10:52:21"}
+{"current_steps": 95, "total_steps": 2405, "loss": 0.6179, "lr": 3.153526970954357e-05, "epoch": 0.19714656290531776, "percentage": 3.95, "elapsed_time": "1:26:00", "remaining_time": "1 day, 10:51:25"}
+{"current_steps": 96, "total_steps": 2405, "loss": 0.6213, "lr": 3.186721991701245e-05, "epoch": 0.1992217898832685, "percentage": 3.99, "elapsed_time": "1:26:54", "remaining_time": "1 day, 10:50:30"}
+{"current_steps": 97, "total_steps": 2405, "loss": 0.6227, "lr": 3.219917012448133e-05, "epoch": 0.2012970168612192, "percentage": 4.03, "elapsed_time": "1:27:49", "remaining_time": "1 day, 10:49:37"}
+{"current_steps": 98, "total_steps": 2405, "loss": 0.6144, "lr": 3.253112033195021e-05, "epoch": 0.2033722438391699, "percentage": 4.07, "elapsed_time": "1:28:43", "remaining_time": "1 day, 10:48:49"}
+{"current_steps": 99, "total_steps": 2405, "loss": 0.6164, "lr": 3.286307053941909e-05, "epoch": 0.20544747081712061, "percentage": 4.12, "elapsed_time": "1:29:38", "remaining_time": "1 day, 10:47:52"}
+{"current_steps": 100, "total_steps": 2405, "loss": 0.6014, "lr": 3.319502074688797e-05, "epoch": 0.20752269779507135, "percentage": 4.16, "elapsed_time": "1:30:32", "remaining_time": "1 day, 10:47:05"}
+{"current_steps": 101, "total_steps": 2405, "loss": 0.626, "lr": 3.352697095435685e-05, "epoch": 0.20959792477302205, "percentage": 4.2, "elapsed_time": "1:31:26", "remaining_time": "1 day, 10:46:03"}
+{"current_steps": 102, "total_steps": 2405, "loss": 0.6216, "lr": 3.3858921161825725e-05, "epoch": 0.21167315175097276, "percentage": 4.24, "elapsed_time": "1:32:20", "remaining_time": "1 day, 10:45:06"}
+{"current_steps": 103, "total_steps": 2405, "loss": 0.6093, "lr": 3.419087136929461e-05, "epoch": 0.21374837872892347, "percentage": 4.28, "elapsed_time": "1:33:15", "remaining_time": "1 day, 10:44:17"}
+{"current_steps": 104, "total_steps": 2405, "loss": 0.6274, "lr": 3.452282157676349e-05, "epoch": 0.2158236057068742, "percentage": 4.32, "elapsed_time": "1:34:09", "remaining_time": "1 day, 10:43:16"}
+{"current_steps": 105, "total_steps": 2405, "loss": 0.6196, "lr": 3.485477178423237e-05, "epoch": 0.2178988326848249, "percentage": 4.37, "elapsed_time": "1:35:04", "remaining_time": "1 day, 10:42:26"}
+{"current_steps": 106, "total_steps": 2405, "loss": 0.6285, "lr": 3.5186721991701245e-05, "epoch": 0.2199740596627756, "percentage": 4.41, "elapsed_time": "1:35:58", "remaining_time": "1 day, 10:41:28"}
+{"current_steps": 107, "total_steps": 2405, "loss": 0.6308, "lr": 3.551867219917013e-05, "epoch": 0.22204928664072632, "percentage": 4.45, "elapsed_time": "1:36:52", "remaining_time": "1 day, 10:40:35"}
+{"current_steps": 108, "total_steps": 2405, "loss": 0.6179, "lr": 3.585062240663901e-05, "epoch": 0.22412451361867705, "percentage": 4.49, "elapsed_time": "1:37:46", "remaining_time": "1 day, 10:39:38"}
+{"current_steps": 109, "total_steps": 2405, "loss": 0.6214, "lr": 3.6182572614107884e-05, "epoch": 0.22619974059662776, "percentage": 4.53, "elapsed_time": "1:38:41", "remaining_time": "1 day, 10:38:51"}
+{"current_steps": 110, "total_steps": 2405, "loss": 0.6149, "lr": 3.6514522821576766e-05, "epoch": 0.22827496757457846, "percentage": 4.57, "elapsed_time": "1:39:35", "remaining_time": "1 day, 10:37:58"}
+{"current_steps": 111, "total_steps": 2405, "loss": 0.6185, "lr": 3.684647302904565e-05, "epoch": 0.23035019455252917, "percentage": 4.62, "elapsed_time": "1:40:29", "remaining_time": "1 day, 10:36:58"}
+{"current_steps": 112, "total_steps": 2405, "loss": 0.614, "lr": 3.717842323651453e-05, "epoch": 0.2324254215304799, "percentage": 4.66, "elapsed_time": "1:41:24", "remaining_time": "1 day, 10:36:12"}
+{"current_steps": 113, "total_steps": 2405, "loss": 0.6145, "lr": 3.7510373443983404e-05, "epoch": 0.2345006485084306, "percentage": 4.7, "elapsed_time": "1:42:19", "remaining_time": "1 day, 10:35:20"}
+{"current_steps": 114, "total_steps": 2405, "loss": 0.6141, "lr": 3.7842323651452286e-05, "epoch": 0.23657587548638132, "percentage": 4.74, "elapsed_time": "1:43:13", "remaining_time": "1 day, 10:34:34"}
+{"current_steps": 115, "total_steps": 2405, "loss": 0.6082, "lr": 3.817427385892117e-05, "epoch": 0.23865110246433205, "percentage": 4.78, "elapsed_time": "1:44:08", "remaining_time": "1 day, 10:33:36"}
+{"current_steps": 116, "total_steps": 2405, "loss": 0.61, "lr": 3.850622406639004e-05, "epoch": 0.24072632944228275, "percentage": 4.82, "elapsed_time": "1:45:01", "remaining_time": "1 day, 10:32:28"}
+{"current_steps": 117, "total_steps": 2405, "loss": 0.6119, "lr": 3.8838174273858924e-05, "epoch": 0.24280155642023346, "percentage": 4.86, "elapsed_time": "1:45:55", "remaining_time": "1 day, 10:31:25"}
+{"current_steps": 118, "total_steps": 2405, "loss": 0.6141, "lr": 3.9170124481327806e-05, "epoch": 0.24487678339818417, "percentage": 4.91, "elapsed_time": "1:46:49", "remaining_time": "1 day, 10:30:20"}
+{"current_steps": 119, "total_steps": 2405, "loss": 0.62, "lr": 3.950207468879669e-05, "epoch": 0.2469520103761349, "percentage": 4.95, "elapsed_time": "1:47:42", "remaining_time": "1 day, 10:29:06"}
+{"current_steps": 120, "total_steps": 2405, "loss": 0.619, "lr": 3.983402489626556e-05, "epoch": 0.2490272373540856, "percentage": 4.99, "elapsed_time": "1:48:37", "remaining_time": "1 day, 10:28:15"}
+{"current_steps": 121, "total_steps": 2405, "loss": 0.6046, "lr": 4.0165975103734444e-05, "epoch": 0.2511024643320363, "percentage": 5.03, "elapsed_time": "1:49:30", "remaining_time": "1 day, 10:27:12"}
+{"current_steps": 122, "total_steps": 2405, "loss": 0.6065, "lr": 4.049792531120332e-05, "epoch": 0.253177691309987, "percentage": 5.07, "elapsed_time": "1:50:24", "remaining_time": "1 day, 10:25:56"}
+{"current_steps": 123, "total_steps": 2405, "loss": 0.6013, "lr": 4.08298755186722e-05, "epoch": 0.2552529182879377, "percentage": 5.11, "elapsed_time": "1:51:17", "remaining_time": "1 day, 10:24:48"}
+{"current_steps": 124, "total_steps": 2405, "loss": 0.5968, "lr": 4.1161825726141076e-05, "epoch": 0.25732814526588843, "percentage": 5.16, "elapsed_time": "1:52:11", "remaining_time": "1 day, 10:23:51"}
+{"current_steps": 125, "total_steps": 2405, "loss": 0.6045, "lr": 4.1493775933609964e-05, "epoch": 0.2594033722438392, "percentage": 5.2, "elapsed_time": "1:53:05", "remaining_time": "1 day, 10:22:52"}
+{"current_steps": 126, "total_steps": 2405, "loss": 0.6036, "lr": 4.1825726141078846e-05, "epoch": 0.2614785992217899, "percentage": 5.24, "elapsed_time": "1:53:58", "remaining_time": "1 day, 10:21:38"}
+{"current_steps": 127, "total_steps": 2405, "loss": 0.6086, "lr": 4.215767634854772e-05, "epoch": 0.2635538261997406, "percentage": 5.28, "elapsed_time": "1:54:52", "remaining_time": "1 day, 10:20:27"}
+{"current_steps": 128, "total_steps": 2405, "loss": 0.6045, "lr": 4.24896265560166e-05, "epoch": 0.2656290531776913, "percentage": 5.32, "elapsed_time": "1:55:46", "remaining_time": "1 day, 10:19:22"}
+{"current_steps": 129, "total_steps": 2405, "loss": 0.6029, "lr": 4.282157676348548e-05, "epoch": 0.267704280155642, "percentage": 5.36, "elapsed_time": "1:56:40", "remaining_time": "1 day, 10:18:23"}
+{"current_steps": 130, "total_steps": 2405, "loss": 0.604, "lr": 4.315352697095436e-05, "epoch": 0.2697795071335927, "percentage": 5.41, "elapsed_time": "1:57:34", "remaining_time": "1 day, 10:17:30"}
+{"current_steps": 131, "total_steps": 2405, "loss": 0.6019, "lr": 4.3485477178423234e-05, "epoch": 0.27185473411154343, "percentage": 5.45, "elapsed_time": "1:58:28", "remaining_time": "1 day, 10:16:39"}
+{"current_steps": 132, "total_steps": 2405, "loss": 0.6017, "lr": 4.381742738589212e-05, "epoch": 0.2739299610894942, "percentage": 5.49, "elapsed_time": "1:59:22", "remaining_time": "1 day, 10:15:28"}
+{"current_steps": 133, "total_steps": 2405, "loss": 0.6116, "lr": 4.4149377593361004e-05, "epoch": 0.2760051880674449, "percentage": 5.53, "elapsed_time": "2:00:15", "remaining_time": "1 day, 10:14:24"}
+{"current_steps": 134, "total_steps": 2405, "loss": 0.6094, "lr": 4.448132780082988e-05, "epoch": 0.2780804150453956, "percentage": 5.57, "elapsed_time": "2:01:10", "remaining_time": "1 day, 10:13:33"}
+{"current_steps": 135, "total_steps": 2405, "loss": 0.6102, "lr": 4.481327800829876e-05, "epoch": 0.2801556420233463, "percentage": 5.61, "elapsed_time": "2:02:04", "remaining_time": "1 day, 10:12:37"}
+{"current_steps": 136, "total_steps": 2405, "loss": 0.6058, "lr": 4.5145228215767636e-05, "epoch": 0.282230869001297, "percentage": 5.65, "elapsed_time": "2:02:58", "remaining_time": "1 day, 10:11:39"}
+{"current_steps": 137, "total_steps": 2405, "loss": 0.6017, "lr": 4.547717842323652e-05, "epoch": 0.2843060959792477, "percentage": 5.7, "elapsed_time": "2:03:52", "remaining_time": "1 day, 10:10:48"}
+{"current_steps": 138, "total_steps": 2405, "loss": 0.6002, "lr": 4.580912863070539e-05, "epoch": 0.2863813229571984, "percentage": 5.74, "elapsed_time": "2:04:47", "remaining_time": "1 day, 10:10:01"}
+{"current_steps": 139, "total_steps": 2405, "loss": 0.6033, "lr": 4.6141078838174274e-05, "epoch": 0.28845654993514913, "percentage": 5.78, "elapsed_time": "2:05:41", "remaining_time": "1 day, 10:09:09"}
+{"current_steps": 140, "total_steps": 2405, "loss": 0.599, "lr": 4.647302904564316e-05, "epoch": 0.2905317769130999, "percentage": 5.82, "elapsed_time": "2:06:36", "remaining_time": "1 day, 10:08:13"}
+{"current_steps": 141, "total_steps": 2405, "loss": 0.6021, "lr": 4.680497925311204e-05, "epoch": 0.2926070038910506, "percentage": 5.86, "elapsed_time": "2:07:30", "remaining_time": "1 day, 10:07:22"}
+{"current_steps": 142, "total_steps": 2405, "loss": 0.6067, "lr": 4.713692946058092e-05, "epoch": 0.2946822308690013, "percentage": 5.9, "elapsed_time": "2:08:24", "remaining_time": "1 day, 10:06:25"}
+{"current_steps": 143, "total_steps": 2405, "loss": 0.5953, "lr": 4.7468879668049795e-05, "epoch": 0.296757457846952, "percentage": 5.95, "elapsed_time": "2:09:18", "remaining_time": "1 day, 10:05:28"}
+{"current_steps": 144, "total_steps": 2405, "loss": 0.6033, "lr": 4.7800829875518676e-05, "epoch": 0.2988326848249027, "percentage": 5.99, "elapsed_time": "2:10:13", "remaining_time": "1 day, 10:04:38"}
+{"current_steps": 145, "total_steps": 2405, "loss": 0.5965, "lr": 4.813278008298755e-05, "epoch": 0.3009079118028534, "percentage": 6.03, "elapsed_time": "2:11:07", "remaining_time": "1 day, 10:03:37"}
+{"current_steps": 146, "total_steps": 2405, "loss": 0.5964, "lr": 4.846473029045643e-05, "epoch": 0.30298313878080413, "percentage": 6.07, "elapsed_time": "2:12:02", "remaining_time": "1 day, 10:03:02"}
+{"current_steps": 147, "total_steps": 2405, "loss": 0.5936, "lr": 4.879668049792532e-05, "epoch": 0.3050583657587549, "percentage": 6.11, "elapsed_time": "2:12:57", "remaining_time": "1 day, 10:02:21"}
+{"current_steps": 148, "total_steps": 2405, "loss": 0.6053, "lr": 4.9128630705394196e-05, "epoch": 0.3071335927367056, "percentage": 6.15, "elapsed_time": "2:13:52", "remaining_time": "1 day, 10:01:34"}
+{"current_steps": 149, "total_steps": 2405, "loss": 0.598, "lr": 4.946058091286308e-05, "epoch": 0.3092088197146563, "percentage": 6.2, "elapsed_time": "2:14:46", "remaining_time": "1 day, 10:00:38"}
+{"current_steps": 150, "total_steps": 2405, "loss": 0.5943, "lr": 4.979253112033195e-05, "epoch": 0.311284046692607, "percentage": 6.24, "elapsed_time": "2:15:39", "remaining_time": "1 day, 9:59:30"}
+{"current_steps": 151, "total_steps": 2405, "loss": 0.6031, "lr": 5.0124481327800835e-05, "epoch": 0.3133592736705577, "percentage": 6.28, "elapsed_time": "2:16:33", "remaining_time": "1 day, 9:58:22"}
+{"current_steps": 152, "total_steps": 2405, "loss": 0.589, "lr": 5.045643153526971e-05, "epoch": 0.3154345006485084, "percentage": 6.32, "elapsed_time": "2:17:26", "remaining_time": "1 day, 9:57:18"}
+{"current_steps": 153, "total_steps": 2405, "loss": 0.5909, "lr": 5.078838174273859e-05, "epoch": 0.3175097276264591, "percentage": 6.36, "elapsed_time": "2:18:20", "remaining_time": "1 day, 9:56:16"}
+{"current_steps": 154, "total_steps": 2405, "loss": 0.5991, "lr": 5.1120331950207466e-05, "epoch": 0.31958495460440983, "percentage": 6.4, "elapsed_time": "2:19:14", "remaining_time": "1 day, 9:55:22"}
+{"current_steps": 155, "total_steps": 2405, "loss": 0.592, "lr": 5.1452282157676355e-05, "epoch": 0.3216601815823606, "percentage": 6.44, "elapsed_time": "2:20:09", "remaining_time": "1 day, 9:54:34"}
+{"current_steps": 156, "total_steps": 2405, "loss": 0.6053, "lr": 5.1784232365145237e-05, "epoch": 0.3237354085603113, "percentage": 6.49, "elapsed_time": "2:21:03", "remaining_time": "1 day, 9:53:42"}
+{"current_steps": 157, "total_steps": 2405, "loss": 0.596, "lr": 5.211618257261411e-05, "epoch": 0.325810635538262, "percentage": 6.53, "elapsed_time": "2:21:58", "remaining_time": "1 day, 9:52:50"}
+{"current_steps": 158, "total_steps": 2405, "loss": 0.5914, "lr": 5.244813278008299e-05, "epoch": 0.3278858625162127, "percentage": 6.57, "elapsed_time": "2:22:52", "remaining_time": "1 day, 9:51:58"}
+{"current_steps": 159, "total_steps": 2405, "loss": 0.5994, "lr": 5.278008298755187e-05, "epoch": 0.3299610894941634, "percentage": 6.61, "elapsed_time": "2:23:46", "remaining_time": "1 day, 9:51:02"}
+{"current_steps": 160, "total_steps": 2405, "loss": 0.5939, "lr": 5.311203319502075e-05, "epoch": 0.3320363164721141, "percentage": 6.65, "elapsed_time": "2:24:40", "remaining_time": "1 day, 9:50:00"}
+{"current_steps": 161, "total_steps": 2405, "loss": 0.591, "lr": 5.3443983402489625e-05, "epoch": 0.33411154345006483, "percentage": 6.69, "elapsed_time": "2:25:34", "remaining_time": "1 day, 9:48:57"}
+{"current_steps": 162, "total_steps": 2405, "loss": 0.5974, "lr": 5.377593360995851e-05, "epoch": 0.3361867704280156, "percentage": 6.74, "elapsed_time": "2:26:28", "remaining_time": "1 day, 9:48:08"}
+{"current_steps": 163, "total_steps": 2405, "loss": 0.6035, "lr": 5.4107883817427395e-05, "epoch": 0.3382619974059663, "percentage": 6.78, "elapsed_time": "2:27:23", "remaining_time": "1 day, 9:47:16"}
+{"current_steps": 164, "total_steps": 2405, "loss": 0.607, "lr": 5.443983402489627e-05, "epoch": 0.340337224383917, "percentage": 6.82, "elapsed_time": "2:28:17", "remaining_time": "1 day, 9:46:19"}
+{"current_steps": 165, "total_steps": 2405, "loss": 0.5986, "lr": 5.477178423236515e-05, "epoch": 0.3424124513618677, "percentage": 6.86, "elapsed_time": "2:29:12", "remaining_time": "1 day, 9:45:31"}
+{"current_steps": 166, "total_steps": 2405, "loss": 0.6051, "lr": 5.510373443983403e-05, "epoch": 0.3444876783398184, "percentage": 6.9, "elapsed_time": "2:30:06", "remaining_time": "1 day, 9:44:45"}
+{"current_steps": 167, "total_steps": 2405, "loss": 0.6054, "lr": 5.543568464730291e-05, "epoch": 0.3465629053177691, "percentage": 6.94, "elapsed_time": "2:31:01", "remaining_time": "1 day, 9:43:48"}
+{"current_steps": 168, "total_steps": 2405, "loss": 0.5928, "lr": 5.576763485477178e-05, "epoch": 0.3486381322957198, "percentage": 6.99, "elapsed_time": "2:31:55", "remaining_time": "1 day, 9:42:57"}
+{"current_steps": 169, "total_steps": 2405, "loss": 0.6137, "lr": 5.6099585062240665e-05, "epoch": 0.35071335927367053, "percentage": 7.03, "elapsed_time": "2:32:49", "remaining_time": "1 day, 9:42:04"}
+{"current_steps": 170, "total_steps": 2405, "loss": 0.5941, "lr": 5.6431535269709554e-05, "epoch": 0.3527885862516213, "percentage": 7.07, "elapsed_time": "2:33:45", "remaining_time": "1 day, 9:41:26"}
+{"current_steps": 171, "total_steps": 2405, "loss": 0.6079, "lr": 5.676348547717843e-05, "epoch": 0.354863813229572, "percentage": 7.11, "elapsed_time": "2:34:40", "remaining_time": "1 day, 9:40:41"}
+{"current_steps": 172, "total_steps": 2405, "loss": 0.6007, "lr": 5.709543568464731e-05, "epoch": 0.3569390402075227, "percentage": 7.15, "elapsed_time": "2:35:35", "remaining_time": "1 day, 9:39:52"}
+{"current_steps": 173, "total_steps": 2405, "loss": 0.593, "lr": 5.7427385892116185e-05, "epoch": 0.3590142671854734, "percentage": 7.19, "elapsed_time": "2:36:29", "remaining_time": "1 day, 9:38:57"}
+{"current_steps": 174, "total_steps": 2405, "loss": 0.5897, "lr": 5.775933609958507e-05, "epoch": 0.3610894941634241, "percentage": 7.23, "elapsed_time": "2:37:23", "remaining_time": "1 day, 9:38:02"}
+{"current_steps": 175, "total_steps": 2405, "loss": 0.5994, "lr": 5.809128630705394e-05, "epoch": 0.3631647211413748, "percentage": 7.28, "elapsed_time": "2:38:17", "remaining_time": "1 day, 9:37:08"}
+{"current_steps": 176, "total_steps": 2405, "loss": 0.593, "lr": 5.8423236514522824e-05, "epoch": 0.36523994811932553, "percentage": 7.32, "elapsed_time": "2:39:11", "remaining_time": "1 day, 9:36:08"}
+{"current_steps": 177, "total_steps": 2405, "loss": 0.5935, "lr": 5.875518672199171e-05, "epoch": 0.3673151750972763, "percentage": 7.36, "elapsed_time": "2:40:06", "remaining_time": "1 day, 9:35:16"}
+{"current_steps": 178, "total_steps": 2405, "loss": 0.603, "lr": 5.908713692946059e-05, "epoch": 0.369390402075227, "percentage": 7.4, "elapsed_time": "2:41:00", "remaining_time": "1 day, 9:34:23"}
+{"current_steps": 179, "total_steps": 2405, "loss": 0.6002, "lr": 5.941908713692947e-05, "epoch": 0.3714656290531777, "percentage": 7.44, "elapsed_time": "2:41:55", "remaining_time": "1 day, 9:33:40"}
+{"current_steps": 180, "total_steps": 2405, "loss": 0.5961, "lr": 5.9751037344398344e-05, "epoch": 0.3735408560311284, "percentage": 7.48, "elapsed_time": "2:42:49", "remaining_time": "1 day, 9:32:36"}
+{"current_steps": 181, "total_steps": 2405, "loss": 0.5979, "lr": 6.0082987551867225e-05, "epoch": 0.3756160830090791, "percentage": 7.53, "elapsed_time": "2:43:42", "remaining_time": "1 day, 9:31:31"}
+{"current_steps": 182, "total_steps": 2405, "loss": 0.5922, "lr": 6.04149377593361e-05, "epoch": 0.3776913099870298, "percentage": 7.57, "elapsed_time": "2:44:35", "remaining_time": "1 day, 9:30:26"}
+{"current_steps": 183, "total_steps": 2405, "loss": 0.5952, "lr": 6.074688796680498e-05, "epoch": 0.37976653696498053, "percentage": 7.61, "elapsed_time": "2:45:30", "remaining_time": "1 day, 9:29:33"}
+{"current_steps": 184, "total_steps": 2405, "loss": 0.5957, "lr": 6.107883817427386e-05, "epoch": 0.38184176394293123, "percentage": 7.65, "elapsed_time": "2:46:24", "remaining_time": "1 day, 9:28:34"}
+{"current_steps": 185, "total_steps": 2405, "loss": 0.5972, "lr": 6.141078838174275e-05, "epoch": 0.383916990920882, "percentage": 7.69, "elapsed_time": "2:47:18", "remaining_time": "1 day, 9:27:37"}
+{"current_steps": 186, "total_steps": 2405, "loss": 0.5944, "lr": 6.174273858921162e-05, "epoch": 0.3859922178988327, "percentage": 7.73, "elapsed_time": "2:48:11", "remaining_time": "1 day, 9:26:35"}
+{"current_steps": 187, "total_steps": 2405, "loss": 0.5862, "lr": 6.20746887966805e-05, "epoch": 0.3880674448767834, "percentage": 7.78, "elapsed_time": "2:49:05", "remaining_time": "1 day, 9:25:36"}
+{"current_steps": 188, "total_steps": 2405, "loss": 0.5858, "lr": 6.240663900414938e-05, "epoch": 0.3901426718547341, "percentage": 7.82, "elapsed_time": "2:49:59", "remaining_time": "1 day, 9:24:35"}
+{"current_steps": 189, "total_steps": 2405, "loss": 0.59, "lr": 6.273858921161827e-05, "epoch": 0.3922178988326848, "percentage": 7.86, "elapsed_time": "2:50:53", "remaining_time": "1 day, 9:23:39"}
+{"current_steps": 190, "total_steps": 2405, "loss": 0.5944, "lr": 6.307053941908713e-05, "epoch": 0.3942931258106355, "percentage": 7.9, "elapsed_time": "2:51:47", "remaining_time": "1 day, 9:22:46"}
+{"current_steps": 191, "total_steps": 2405, "loss": 0.5931, "lr": 6.340248962655602e-05, "epoch": 0.39636835278858623, "percentage": 7.94, "elapsed_time": "2:52:42", "remaining_time": "1 day, 9:21:57"}
+{"current_steps": 192, "total_steps": 2405, "loss": 0.5888, "lr": 6.37344398340249e-05, "epoch": 0.398443579766537, "percentage": 7.98, "elapsed_time": "2:53:36", "remaining_time": "1 day, 9:21:02"}
+{"current_steps": 193, "total_steps": 2405, "loss": 0.5854, "lr": 6.406639004149378e-05, "epoch": 0.4005188067444877, "percentage": 8.02, "elapsed_time": "2:54:30", "remaining_time": "1 day, 9:20:04"}
+{"current_steps": 194, "total_steps": 2405, "loss": 0.593, "lr": 6.439834024896266e-05, "epoch": 0.4025940337224384, "percentage": 8.07, "elapsed_time": "2:55:24", "remaining_time": "1 day, 9:19:06"}
+{"current_steps": 195, "total_steps": 2405, "loss": 0.5909, "lr": 6.473029045643154e-05, "epoch": 0.4046692607003891, "percentage": 8.11, "elapsed_time": "2:56:19", "remaining_time": "1 day, 9:18:19"}
+{"current_steps": 196, "total_steps": 2405, "loss": 0.5882, "lr": 6.506224066390042e-05, "epoch": 0.4067444876783398, "percentage": 8.15, "elapsed_time": "2:57:13", "remaining_time": "1 day, 9:17:22"}
+{"current_steps": 197, "total_steps": 2405, "loss": 0.5964, "lr": 6.539419087136929e-05, "epoch": 0.4088197146562905, "percentage": 8.19, "elapsed_time": "2:58:07", "remaining_time": "1 day, 9:16:21"}
+{"current_steps": 198, "total_steps": 2405, "loss": 0.5946, "lr": 6.572614107883817e-05, "epoch": 0.41089494163424123, "percentage": 8.23, "elapsed_time": "2:59:00", "remaining_time": "1 day, 9:15:23"}
+{"current_steps": 199, "total_steps": 2405, "loss": 0.5852, "lr": 6.605809128630706e-05, "epoch": 0.41297016861219193, "percentage": 8.27, "elapsed_time": "2:59:54", "remaining_time": "1 day, 9:14:26"}
+{"current_steps": 200, "total_steps": 2405, "loss": 0.5889, "lr": 6.639004149377594e-05, "epoch": 0.4150453955901427, "percentage": 8.32, "elapsed_time": "3:00:48", "remaining_time": "1 day, 9:13:29"}
+{"current_steps": 201, "total_steps": 2405, "loss": 0.5794, "lr": 6.672199170124482e-05, "epoch": 0.4171206225680934, "percentage": 8.36, "elapsed_time": "3:01:42", "remaining_time": "1 day, 9:12:30"}
+{"current_steps": 202, "total_steps": 2405, "loss": 0.5798, "lr": 6.70539419087137e-05, "epoch": 0.4191958495460441, "percentage": 8.4, "elapsed_time": "3:02:36", "remaining_time": "1 day, 9:11:27"}
+{"current_steps": 203, "total_steps": 2405, "loss": 0.5841, "lr": 6.738589211618258e-05, "epoch": 0.4212710765239948, "percentage": 8.44, "elapsed_time": "3:03:30", "remaining_time": "1 day, 9:10:29"}
+{"current_steps": 204, "total_steps": 2405, "loss": 0.5837, "lr": 6.771784232365145e-05, "epoch": 0.4233463035019455, "percentage": 8.48, "elapsed_time": "3:04:23", "remaining_time": "1 day, 9:09:24"}
+{"current_steps": 205, "total_steps": 2405, "loss": 0.5828, "lr": 6.804979253112033e-05, "epoch": 0.4254215304798962, "percentage": 8.52, "elapsed_time": "3:05:17", "remaining_time": "1 day, 9:08:29"}
+{"current_steps": 206, "total_steps": 2405, "loss": 0.5872, "lr": 6.838174273858921e-05, "epoch": 0.42749675745784693, "percentage": 8.57, "elapsed_time": "3:06:11", "remaining_time": "1 day, 9:07:29"}
+{"current_steps": 207, "total_steps": 2405, "loss": 0.5931, "lr": 6.87136929460581e-05, "epoch": 0.42957198443579764, "percentage": 8.61, "elapsed_time": "3:07:05", "remaining_time": "1 day, 9:06:39"}
+{"current_steps": 208, "total_steps": 2405, "loss": 0.5887, "lr": 6.904564315352698e-05, "epoch": 0.4316472114137484, "percentage": 8.65, "elapsed_time": "3:08:00", "remaining_time": "1 day, 9:05:49"}
+{"current_steps": 209, "total_steps": 2405, "loss": 0.5894, "lr": 6.937759336099586e-05, "epoch": 0.4337224383916991, "percentage": 8.69, "elapsed_time": "3:08:54", "remaining_time": "1 day, 9:04:57"}
+{"current_steps": 210, "total_steps": 2405, "loss": 0.5811, "lr": 6.970954356846474e-05, "epoch": 0.4357976653696498, "percentage": 8.73, "elapsed_time": "3:09:49", "remaining_time": "1 day, 9:04:07"}
+{"current_steps": 211, "total_steps": 2405, "loss": 0.5858, "lr": 7.004149377593361e-05, "epoch": 0.4378728923476005, "percentage": 8.77, "elapsed_time": "3:10:43", "remaining_time": "1 day, 9:03:14"}
+{"current_steps": 212, "total_steps": 2405, "loss": 0.586, "lr": 7.037344398340249e-05, "epoch": 0.4399481193255512, "percentage": 8.81, "elapsed_time": "3:11:37", "remaining_time": "1 day, 9:02:18"}
+{"current_steps": 213, "total_steps": 2405, "loss": 0.589, "lr": 7.070539419087137e-05, "epoch": 0.44202334630350193, "percentage": 8.86, "elapsed_time": "3:12:32", "remaining_time": "1 day, 9:01:23"}
+{"current_steps": 214, "total_steps": 2405, "loss": 0.5818, "lr": 7.103734439834025e-05, "epoch": 0.44409857328145264, "percentage": 8.9, "elapsed_time": "3:13:26", "remaining_time": "1 day, 9:00:28"}
+{"current_steps": 215, "total_steps": 2405, "loss": 0.5815, "lr": 7.136929460580914e-05, "epoch": 0.4461738002594034, "percentage": 8.94, "elapsed_time": "3:14:21", "remaining_time": "1 day, 8:59:41"}
+{"current_steps": 216, "total_steps": 2405, "loss": 0.584, "lr": 7.170124481327802e-05, "epoch": 0.4482490272373541, "percentage": 8.98, "elapsed_time": "3:15:15", "remaining_time": "1 day, 8:58:48"}
+{"current_steps": 217, "total_steps": 2405, "loss": 0.584, "lr": 7.20331950207469e-05, "epoch": 0.4503242542153048, "percentage": 9.02, "elapsed_time": "3:16:09", "remaining_time": "1 day, 8:57:55"}
+{"current_steps": 218, "total_steps": 2405, "loss": 0.5817, "lr": 7.236514522821577e-05, "epoch": 0.4523994811932555, "percentage": 9.06, "elapsed_time": "3:17:06", "remaining_time": "1 day, 8:57:24"}
+{"current_steps": 219, "total_steps": 2405, "loss": 0.5792, "lr": 7.269709543568465e-05, "epoch": 0.4544747081712062, "percentage": 9.11, "elapsed_time": "3:18:01", "remaining_time": "1 day, 8:56:39"}
+{"current_steps": 220, "total_steps": 2405, "loss": 0.5829, "lr": 7.302904564315353e-05, "epoch": 0.4565499351491569, "percentage": 9.15, "elapsed_time": "3:18:56", "remaining_time": "1 day, 8:55:49"}
+{"current_steps": 221, "total_steps": 2405, "loss": 0.5872, "lr": 7.336099585062241e-05, "epoch": 0.45862516212710763, "percentage": 9.19, "elapsed_time": "3:19:51", "remaining_time": "1 day, 8:55:07"}
+{"current_steps": 222, "total_steps": 2405, "loss": 0.5792, "lr": 7.36929460580913e-05, "epoch": 0.46070038910505834, "percentage": 9.23, "elapsed_time": "3:20:46", "remaining_time": "1 day, 8:54:12"}
+{"current_steps": 223, "total_steps": 2405, "loss": 0.5793, "lr": 7.402489626556018e-05, "epoch": 0.4627756160830091, "percentage": 9.27, "elapsed_time": "3:21:40", "remaining_time": "1 day, 8:53:22"}
+{"current_steps": 224, "total_steps": 2405, "loss": 0.572, "lr": 7.435684647302906e-05, "epoch": 0.4648508430609598, "percentage": 9.31, "elapsed_time": "3:22:35", "remaining_time": "1 day, 8:52:30"}
+{"current_steps": 225, "total_steps": 2405, "loss": 0.583, "lr": 7.468879668049793e-05, "epoch": 0.4669260700389105, "percentage": 9.36, "elapsed_time": "3:23:29", "remaining_time": "1 day, 8:51:31"}
+{"current_steps": 226, "total_steps": 2405, "loss": 0.5847, "lr": 7.502074688796681e-05, "epoch": 0.4690012970168612, "percentage": 9.4, "elapsed_time": "3:24:22", "remaining_time": "1 day, 8:50:34"}
+{"current_steps": 227, "total_steps": 2405, "loss": 0.5826, "lr": 7.535269709543569e-05, "epoch": 0.4710765239948119, "percentage": 9.44, "elapsed_time": "3:25:18", "remaining_time": "1 day, 8:49:48"}
+{"current_steps": 228, "total_steps": 2405, "loss": 0.5793, "lr": 7.568464730290457e-05, "epoch": 0.47315175097276263, "percentage": 9.48, "elapsed_time": "3:26:12", "remaining_time": "1 day, 8:48:56"}
+{"current_steps": 229, "total_steps": 2405, "loss": 0.5744, "lr": 7.601659751037345e-05, "epoch": 0.47522697795071334, "percentage": 9.52, "elapsed_time": "3:27:08", "remaining_time": "1 day, 8:48:17"}
+{"current_steps": 230, "total_steps": 2405, "loss": 0.582, "lr": 7.634854771784233e-05, "epoch": 0.4773022049286641, "percentage": 9.56, "elapsed_time": "3:28:03", "remaining_time": "1 day, 8:47:30"}
+{"current_steps": 231, "total_steps": 2405, "loss": 0.5866, "lr": 7.668049792531122e-05, "epoch": 0.4793774319066148, "percentage": 9.6, "elapsed_time": "3:28:57", "remaining_time": "1 day, 8:46:36"}
+{"current_steps": 232, "total_steps": 2405, "loss": 0.5788, "lr": 7.701244813278008e-05, "epoch": 0.4814526588845655, "percentage": 9.65, "elapsed_time": "3:29:53", "remaining_time": "1 day, 8:45:53"}
+{"current_steps": 233, "total_steps": 2405, "loss": 0.5856, "lr": 7.734439834024897e-05, "epoch": 0.4835278858625162, "percentage": 9.69, "elapsed_time": "3:30:48", "remaining_time": "1 day, 8:45:06"}
+{"current_steps": 234, "total_steps": 2405, "loss": 0.5842, "lr": 7.767634854771785e-05, "epoch": 0.4856031128404669, "percentage": 9.73, "elapsed_time": "3:31:43", "remaining_time": "1 day, 8:44:21"}
+{"current_steps": 235, "total_steps": 2405, "loss": 0.5826, "lr": 7.800829875518673e-05, "epoch": 0.4876783398184176, "percentage": 9.77, "elapsed_time": "3:32:37", "remaining_time": "1 day, 8:43:27"}
+{"current_steps": 236, "total_steps": 2405, "loss": 0.5764, "lr": 7.834024896265561e-05, "epoch": 0.48975356679636833, "percentage": 9.81, "elapsed_time": "3:33:33", "remaining_time": "1 day, 8:42:44"}
+{"current_steps": 237, "total_steps": 2405, "loss": 0.5813, "lr": 7.867219917012448e-05, "epoch": 0.49182879377431904, "percentage": 9.85, "elapsed_time": "3:34:27", "remaining_time": "1 day, 8:41:51"}
+{"current_steps": 238, "total_steps": 2405, "loss": 0.5778, "lr": 7.900414937759337e-05, "epoch": 0.4939040207522698, "percentage": 9.9, "elapsed_time": "3:35:22", "remaining_time": "1 day, 8:41:02"}
+{"current_steps": 239, "total_steps": 2405, "loss": 0.5766, "lr": 7.933609958506224e-05, "epoch": 0.4959792477302205, "percentage": 9.94, "elapsed_time": "3:36:17", "remaining_time": "1 day, 8:40:13"}
+{"current_steps": 240, "total_steps": 2405, "loss": 0.5753, "lr": 7.966804979253112e-05, "epoch": 0.4980544747081712, "percentage": 9.98, "elapsed_time": "3:37:12", "remaining_time": "1 day, 8:39:25"}
+{"current_steps": 241, "total_steps": 2405, "loss": 0.5847, "lr": 8e-05, "epoch": 0.5001297016861219, "percentage": 10.02, "elapsed_time": "3:38:07", "remaining_time": "1 day, 8:38:38"}
+{"current_steps": 242, "total_steps": 2405, "loss": 0.5767, "lr": 7.999995784829445e-05, "epoch": 0.5022049286640726, "percentage": 10.06, "elapsed_time": "3:39:02", "remaining_time": "1 day, 8:37:44"}
+{"current_steps": 243, "total_steps": 2405, "loss": 0.581, "lr": 7.999983139326661e-05, "epoch": 0.5042801556420233, "percentage": 10.1, "elapsed_time": "3:39:56", "remaining_time": "1 day, 8:36:48"}
+{"current_steps": 244, "total_steps": 2405, "loss": 0.5875, "lr": 7.999962063518301e-05, "epoch": 0.506355382619974, "percentage": 10.15, "elapsed_time": "3:40:50", "remaining_time": "1 day, 8:35:53"}
+{"current_steps": 245, "total_steps": 2405, "loss": 0.5811, "lr": 7.999932557448784e-05, "epoch": 0.5084306095979247, "percentage": 10.19, "elapsed_time": "3:41:44", "remaining_time": "1 day, 8:34:55"}
+{"current_steps": 246, "total_steps": 2405, "loss": 0.579, "lr": 7.999894621180295e-05, "epoch": 0.5105058365758754, "percentage": 10.23, "elapsed_time": "3:42:38", "remaining_time": "1 day, 8:34:04"}
+{"current_steps": 247, "total_steps": 2405, "loss": 0.5781, "lr": 7.99984825479279e-05, "epoch": 0.5125810635538262, "percentage": 10.27, "elapsed_time": "3:43:33", "remaining_time": "1 day, 8:33:12"}
+{"current_steps": 248, "total_steps": 2405, "loss": 0.5776, "lr": 7.99979345838399e-05, "epoch": 0.5146562905317769, "percentage": 10.31, "elapsed_time": "3:44:27", "remaining_time": "1 day, 8:32:16"}
+{"current_steps": 249, "total_steps": 2405, "loss": 0.5803, "lr": 7.99973023206938e-05, "epoch": 0.5167315175097277, "percentage": 10.35, "elapsed_time": "3:45:21", "remaining_time": "1 day, 8:31:20"}
+{"current_steps": 250, "total_steps": 2405, "loss": 0.5727, "lr": 7.99965857598222e-05, "epoch": 0.5188067444876784, "percentage": 10.4, "elapsed_time": "3:46:15", "remaining_time": "1 day, 8:30:22"}
+{"current_steps": 251, "total_steps": 2405, "loss": 0.5856, "lr": 7.999578490273526e-05, "epoch": 0.5208819714656291, "percentage": 10.44, "elapsed_time": "3:47:09", "remaining_time": "1 day, 8:29:28"}
+{"current_steps": 252, "total_steps": 2405, "loss": 0.5797, "lr": 7.99948997511209e-05, "epoch": 0.5229571984435798, "percentage": 10.48, "elapsed_time": "3:48:03", "remaining_time": "1 day, 8:28:31"}
+{"current_steps": 253, "total_steps": 2405, "loss": 0.569, "lr": 7.999393030684461e-05, "epoch": 0.5250324254215305, "percentage": 10.52, "elapsed_time": "3:48:58", "remaining_time": "1 day, 8:27:35"}
+{"current_steps": 254, "total_steps": 2405, "loss": 0.5804, "lr": 7.999287657194963e-05, "epoch": 0.5271076523994812, "percentage": 10.56, "elapsed_time": "3:49:52", "remaining_time": "1 day, 8:26:39"}
+{"current_steps": 255, "total_steps": 2405, "loss": 0.5746, "lr": 7.999173854865673e-05, "epoch": 0.5291828793774319, "percentage": 10.6, "elapsed_time": "3:50:46", "remaining_time": "1 day, 8:25:44"}
+{"current_steps": 256, "total_steps": 2405, "loss": 0.5767, "lr": 7.999051623936446e-05, "epoch": 0.5312581063553826, "percentage": 10.64, "elapsed_time": "3:51:40", "remaining_time": "1 day, 8:24:44"}
+{"current_steps": 257, "total_steps": 2405, "loss": 0.5687, "lr": 7.998920964664889e-05, "epoch": 0.5333333333333333, "percentage": 10.69, "elapsed_time": "3:52:34", "remaining_time": "1 day, 8:23:55"}
+{"current_steps": 258, "total_steps": 2405, "loss": 0.5784, "lr": 7.998781877326378e-05, "epoch": 0.535408560311284, "percentage": 10.73, "elapsed_time": "3:53:29", "remaining_time": "1 day, 8:23:00"}
+{"current_steps": 259, "total_steps": 2405, "loss": 0.5707, "lr": 7.998634362214055e-05, "epoch": 0.5374837872892347, "percentage": 10.77, "elapsed_time": "3:54:23", "remaining_time": "1 day, 8:22:08"}
+{"current_steps": 260, "total_steps": 2405, "loss": 0.5712, "lr": 7.998478419638818e-05, "epoch": 0.5395590142671854, "percentage": 10.81, "elapsed_time": "3:55:17", "remaining_time": "1 day, 8:21:13"}
+{"current_steps": 261, "total_steps": 2405, "loss": 0.5754, "lr": 7.998314049929328e-05, "epoch": 0.5416342412451361, "percentage": 10.85, "elapsed_time": "3:56:12", "remaining_time": "1 day, 8:20:22"}
+{"current_steps": 262, "total_steps": 2405, "loss": 0.5703, "lr": 7.998141253432012e-05, "epoch": 0.5437094682230869, "percentage": 10.89, "elapsed_time": "3:57:06", "remaining_time": "1 day, 8:19:24"}
+{"current_steps": 263, "total_steps": 2405, "loss": 0.5707, "lr": 7.99796003051105e-05, "epoch": 0.5457846952010376, "percentage": 10.94, "elapsed_time": "3:58:00", "remaining_time": "1 day, 8:18:30"}
+{"current_steps": 264, "total_steps": 2405, "loss": 0.5698, "lr": 7.997770381548385e-05, "epoch": 0.5478599221789884, "percentage": 10.98, "elapsed_time": "3:58:55", "remaining_time": "1 day, 8:17:36"}
+{"current_steps": 265, "total_steps": 2405, "loss": 0.5702, "lr": 7.99757230694372e-05, "epoch": 0.5499351491569391, "percentage": 11.02, "elapsed_time": "3:59:50", "remaining_time": "1 day, 8:16:46"}
+{"current_steps": 266, "total_steps": 2405, "loss": 0.571, "lr": 7.997365807114514e-05, "epoch": 0.5520103761348898, "percentage": 11.06, "elapsed_time": "4:00:44", "remaining_time": "1 day, 8:15:54"}
+{"current_steps": 267, "total_steps": 2405, "loss": 0.5656, "lr": 7.997150882495984e-05, "epoch": 0.5540856031128405, "percentage": 11.1, "elapsed_time": "4:01:38", "remaining_time": "1 day, 8:14:56"}
+{"current_steps": 268, "total_steps": 2405, "loss": 0.5689, "lr": 7.996927533541097e-05, "epoch": 0.5561608300907912, "percentage": 11.14, "elapsed_time": "4:02:32", "remaining_time": "1 day, 8:14:02"}
+{"current_steps": 269, "total_steps": 2405, "loss": 0.5747, "lr": 7.996695760720585e-05, "epoch": 0.5582360570687419, "percentage": 11.19, "elapsed_time": "4:03:26", "remaining_time": "1 day, 8:13:05"}
+{"current_steps": 270, "total_steps": 2405, "loss": 0.5781, "lr": 7.996455564522927e-05, "epoch": 0.5603112840466926, "percentage": 11.23, "elapsed_time": "4:04:21", "remaining_time": "1 day, 8:12:12"}
+{"current_steps": 271, "total_steps": 2405, "loss": 0.5744, "lr": 7.996206945454359e-05, "epoch": 0.5623865110246433, "percentage": 11.27, "elapsed_time": "4:05:15", "remaining_time": "1 day, 8:11:21"}
+{"current_steps": 272, "total_steps": 2405, "loss": 0.5691, "lr": 7.995949904038864e-05, "epoch": 0.564461738002594, "percentage": 11.31, "elapsed_time": "4:06:10", "remaining_time": "1 day, 8:10:29"}
+{"current_steps": 273, "total_steps": 2405, "loss": 0.5687, "lr": 7.99568444081818e-05, "epoch": 0.5665369649805447, "percentage": 11.35, "elapsed_time": "4:07:04", "remaining_time": "1 day, 8:09:30"}
+{"current_steps": 274, "total_steps": 2405, "loss": 0.5732, "lr": 7.995410556351793e-05, "epoch": 0.5686121919584954, "percentage": 11.39, "elapsed_time": "4:07:58", "remaining_time": "1 day, 8:08:35"}
+{"current_steps": 275, "total_steps": 2405, "loss": 0.5776, "lr": 7.995128251216937e-05, "epoch": 0.5706874189364461, "percentage": 11.43, "elapsed_time": "4:08:52", "remaining_time": "1 day, 8:07:39"}
+{"current_steps": 276, "total_steps": 2405, "loss": 0.5652, "lr": 7.994837526008597e-05, "epoch": 0.5727626459143969, "percentage": 11.48, "elapsed_time": "4:09:46", "remaining_time": "1 day, 8:06:40"}
+{"current_steps": 277, "total_steps": 2405, "loss": 0.5662, "lr": 7.9945383813395e-05, "epoch": 0.5748378728923476, "percentage": 11.52, "elapsed_time": "4:10:41", "remaining_time": "1 day, 8:05:51"}
+{"current_steps": 278, "total_steps": 2405, "loss": 0.5661, "lr": 7.994230817840116e-05, "epoch": 0.5769130998702983, "percentage": 11.56, "elapsed_time": "4:11:35", "remaining_time": "1 day, 8:04:57"}
+{"current_steps": 279, "total_steps": 2405, "loss": 0.5778, "lr": 7.993914836158666e-05, "epoch": 0.5789883268482491, "percentage": 11.6, "elapsed_time": "4:12:30", "remaining_time": "1 day, 8:04:04"}
+{"current_steps": 280, "total_steps": 2405, "loss": 0.5718, "lr": 7.993590436961106e-05, "epoch": 0.5810635538261998, "percentage": 11.64, "elapsed_time": "4:13:24", "remaining_time": "1 day, 8:03:11"}
+{"current_steps": 281, "total_steps": 2405, "loss": 0.5693, "lr": 7.993257620931132e-05, "epoch": 0.5831387808041505, "percentage": 11.68, "elapsed_time": "4:14:18", "remaining_time": "1 day, 8:02:16"}
+{"current_steps": 282, "total_steps": 2405, "loss": 0.5614, "lr": 7.992916388770187e-05, "epoch": 0.5852140077821012, "percentage": 11.73, "elapsed_time": "4:15:13", "remaining_time": "1 day, 8:01:26"}
+{"current_steps": 283, "total_steps": 2405, "loss": 0.577, "lr": 7.992566741197446e-05, "epoch": 0.5872892347600519, "percentage": 11.77, "elapsed_time": "4:16:08", "remaining_time": "1 day, 8:00:35"}
+{"current_steps": 284, "total_steps": 2405, "loss": 0.5647, "lr": 7.99220867894982e-05, "epoch": 0.5893644617380026, "percentage": 11.81, "elapsed_time": "4:17:03", "remaining_time": "1 day, 7:59:47"}
+{"current_steps": 285, "total_steps": 2405, "loss": 0.5699, "lr": 7.991842202781954e-05, "epoch": 0.5914396887159533, "percentage": 11.85, "elapsed_time": "4:17:57", "remaining_time": "1 day, 7:58:53"}
+{"current_steps": 286, "total_steps": 2405, "loss": 0.5685, "lr": 7.991467313466232e-05, "epoch": 0.593514915693904, "percentage": 11.89, "elapsed_time": "4:18:52", "remaining_time": "1 day, 7:57:58"}
+{"current_steps": 287, "total_steps": 2405, "loss": 0.5623, "lr": 7.991084011792762e-05, "epoch": 0.5955901426718547, "percentage": 11.93, "elapsed_time": "4:19:46", "remaining_time": "1 day, 7:57:03"}
+{"current_steps": 288, "total_steps": 2405, "loss": 0.5682, "lr": 7.990692298569385e-05, "epoch": 0.5976653696498054, "percentage": 11.98, "elapsed_time": "4:20:40", "remaining_time": "1 day, 7:56:09"}
+{"current_steps": 289, "total_steps": 2405, "loss": 0.5682, "lr": 7.99029217462167e-05, "epoch": 0.5997405966277561, "percentage": 12.02, "elapsed_time": "4:21:36", "remaining_time": "1 day, 7:55:23"}
+{"current_steps": 290, "total_steps": 2405, "loss": 0.572, "lr": 7.989883640792915e-05, "epoch": 0.6018158236057068, "percentage": 12.06, "elapsed_time": "4:22:31", "remaining_time": "1 day, 7:54:36"}
+{"current_steps": 291, "total_steps": 2405, "loss": 0.557, "lr": 7.989466697944137e-05, "epoch": 0.6038910505836576, "percentage": 12.1, "elapsed_time": "4:23:26", "remaining_time": "1 day, 7:53:46"}
+{"current_steps": 292, "total_steps": 2405, "loss": 0.5617, "lr": 7.98904134695408e-05, "epoch": 0.6059662775616083, "percentage": 12.14, "elapsed_time": "4:24:20", "remaining_time": "1 day, 7:52:51"}
+{"current_steps": 293, "total_steps": 2405, "loss": 0.5685, "lr": 7.988607588719208e-05, "epoch": 0.608041504539559, "percentage": 12.18, "elapsed_time": "4:25:14", "remaining_time": "1 day, 7:51:53"}
+{"current_steps": 294, "total_steps": 2405, "loss": 0.5684, "lr": 7.9881654241537e-05, "epoch": 0.6101167315175098, "percentage": 12.22, "elapsed_time": "4:26:08", "remaining_time": "1 day, 7:51:01"}
+{"current_steps": 295, "total_steps": 2405, "loss": 0.5628, "lr": 7.98771485418946e-05, "epoch": 0.6121919584954605, "percentage": 12.27, "elapsed_time": "4:27:03", "remaining_time": "1 day, 7:50:10"}
+{"current_steps": 296, "total_steps": 2405, "loss": 0.5661, "lr": 7.9872558797761e-05, "epoch": 0.6142671854734112, "percentage": 12.31, "elapsed_time": "4:27:57", "remaining_time": "1 day, 7:49:15"}
+{"current_steps": 297, "total_steps": 2405, "loss": 0.5684, "lr": 7.98678850188095e-05, "epoch": 0.6163424124513619, "percentage": 12.35, "elapsed_time": "4:28:52", "remaining_time": "1 day, 7:48:21"}
+{"current_steps": 298, "total_steps": 2405, "loss": 0.5741, "lr": 7.986312721489045e-05, "epoch": 0.6184176394293126, "percentage": 12.39, "elapsed_time": "4:29:46", "remaining_time": "1 day, 7:47:24"}
+{"current_steps": 299, "total_steps": 2405, "loss": 0.5651, "lr": 7.985828539603137e-05, "epoch": 0.6204928664072633, "percentage": 12.43, "elapsed_time": "4:30:39", "remaining_time": "1 day, 7:46:25"}
+{"current_steps": 300, "total_steps": 2405, "loss": 0.5625, "lr": 7.985335957243678e-05, "epoch": 0.622568093385214, "percentage": 12.47, "elapsed_time": "4:31:33", "remaining_time": "1 day, 7:45:27"}
+{"current_steps": 301, "total_steps": 2405, "loss": 0.5625, "lr": 7.984834975448827e-05, "epoch": 0.6246433203631647, "percentage": 12.52, "elapsed_time": "4:32:27", "remaining_time": "1 day, 7:44:30"}
+{"current_steps": 302, "total_steps": 2405, "loss": 0.5697, "lr": 7.984325595274447e-05, "epoch": 0.6267185473411154, "percentage": 12.56, "elapsed_time": "4:33:22", "remaining_time": "1 day, 7:43:39"}
+{"current_steps": 303, "total_steps": 2405, "loss": 0.5679, "lr": 7.9838078177941e-05, "epoch": 0.6287937743190661, "percentage": 12.6, "elapsed_time": "4:34:15", "remaining_time": "1 day, 7:42:38"}
+{"current_steps": 304, "total_steps": 2405, "loss": 0.5637, "lr": 7.983281644099046e-05, "epoch": 0.6308690012970168, "percentage": 12.64, "elapsed_time": "4:35:09", "remaining_time": "1 day, 7:41:42"}
+{"current_steps": 305, "total_steps": 2405, "loss": 0.5592, "lr": 7.982747075298243e-05, "epoch": 0.6329442282749675, "percentage": 12.68, "elapsed_time": "4:36:04", "remaining_time": "1 day, 7:40:47"}
+{"current_steps": 306, "total_steps": 2405, "loss": 0.5605, "lr": 7.982204112518336e-05, "epoch": 0.6350194552529183, "percentage": 12.72, "elapsed_time": "4:36:58", "remaining_time": "1 day, 7:39:54"}
+{"current_steps": 307, "total_steps": 2405, "loss": 0.5642, "lr": 7.981652756903668e-05, "epoch": 0.637094682230869, "percentage": 12.77, "elapsed_time": "4:37:52", "remaining_time": "1 day, 7:38:57"}
+{"current_steps": 308, "total_steps": 2405, "loss": 0.5616, "lr": 7.981093009616268e-05, "epoch": 0.6391699092088197, "percentage": 12.81, "elapsed_time": "4:38:46", "remaining_time": "1 day, 7:38:04"}
+{"current_steps": 309, "total_steps": 2405, "loss": 0.5662, "lr": 7.980524871835853e-05, "epoch": 0.6412451361867705, "percentage": 12.85, "elapsed_time": "4:39:41", "remaining_time": "1 day, 7:37:09"}
+{"current_steps": 310, "total_steps": 2405, "loss": 0.5688, "lr": 7.979948344759817e-05, "epoch": 0.6433203631647212, "percentage": 12.89, "elapsed_time": "4:40:35", "remaining_time": "1 day, 7:36:12"}
+{"current_steps": 311, "total_steps": 2405, "loss": 0.5675, "lr": 7.979363429603243e-05, "epoch": 0.6453955901426719, "percentage": 12.93, "elapsed_time": "4:41:29", "remaining_time": "1 day, 7:35:17"}
+{"current_steps": 312, "total_steps": 2405, "loss": 0.5651, "lr": 7.978770127598891e-05, "epoch": 0.6474708171206226, "percentage": 12.97, "elapsed_time": "4:42:23", "remaining_time": "1 day, 7:34:19"}
+{"current_steps": 313, "total_steps": 2405, "loss": 0.5615, "lr": 7.978168439997192e-05, "epoch": 0.6495460440985733, "percentage": 13.01, "elapsed_time": "4:43:17", "remaining_time": "1 day, 7:33:26"}
+{"current_steps": 314, "total_steps": 2405, "loss": 0.5668, "lr": 7.977558368066256e-05, "epoch": 0.651621271076524, "percentage": 13.06, "elapsed_time": "4:44:11", "remaining_time": "1 day, 7:32:30"}
+{"current_steps": 315, "total_steps": 2405, "loss": 0.5762, "lr": 7.976939913091862e-05, "epoch": 0.6536964980544747, "percentage": 13.1, "elapsed_time": "4:45:04", "remaining_time": "1 day, 7:31:30"}
+{"current_steps": 316, "total_steps": 2405, "loss": 0.5682, "lr": 7.976313076377455e-05, "epoch": 0.6557717250324254, "percentage": 13.14, "elapsed_time": "4:45:58", "remaining_time": "1 day, 7:30:32"}
+{"current_steps": 317, "total_steps": 2405, "loss": 0.5703, "lr": 7.97567785924415e-05, "epoch": 0.6578469520103761, "percentage": 13.18, "elapsed_time": "4:46:54", "remaining_time": "1 day, 7:29:46"}
+{"current_steps": 318, "total_steps": 2405, "loss": 0.5599, "lr": 7.975034263030717e-05, "epoch": 0.6599221789883268, "percentage": 13.22, "elapsed_time": "4:47:48", "remaining_time": "1 day, 7:28:49"}
+{"current_steps": 319, "total_steps": 2405, "loss": 0.5663, "lr": 7.974382289093593e-05, "epoch": 0.6619974059662775, "percentage": 13.26, "elapsed_time": "4:48:42", "remaining_time": "1 day, 7:27:57"}
+{"current_steps": 320, "total_steps": 2405, "loss": 0.5677, "lr": 7.973721938806866e-05, "epoch": 0.6640726329442282, "percentage": 13.31, "elapsed_time": "4:49:37", "remaining_time": "1 day, 7:27:02"}
+{"current_steps": 321, "total_steps": 2405, "loss": 0.5649, "lr": 7.973053213562284e-05, "epoch": 0.666147859922179, "percentage": 13.35, "elapsed_time": "4:50:30", "remaining_time": "1 day, 7:26:04"}
+{"current_steps": 322, "total_steps": 2405, "loss": 0.5659, "lr": 7.97237611476924e-05, "epoch": 0.6682230869001297, "percentage": 13.39, "elapsed_time": "4:51:24", "remaining_time": "1 day, 7:25:06"}
+{"current_steps": 323, "total_steps": 2405, "loss": 0.5546, "lr": 7.971690643854778e-05, "epoch": 0.6702983138780804, "percentage": 13.43, "elapsed_time": "4:52:18", "remaining_time": "1 day, 7:24:10"}
+{"current_steps": 324, "total_steps": 2405, "loss": 0.5572, "lr": 7.970996802263588e-05, "epoch": 0.6723735408560312, "percentage": 13.47, "elapsed_time": "4:53:12", "remaining_time": "1 day, 7:23:14"}
+{"current_steps": 325, "total_steps": 2405, "loss": 0.5586, "lr": 7.970294591457996e-05, "epoch": 0.6744487678339819, "percentage": 13.51, "elapsed_time": "4:54:06", "remaining_time": "1 day, 7:22:20"}
+{"current_steps": 326, "total_steps": 2405, "loss": 0.5625, "lr": 7.969584012917975e-05, "epoch": 0.6765239948119326, "percentage": 13.56, "elapsed_time": "4:55:01", "remaining_time": "1 day, 7:21:25"}
+{"current_steps": 327, "total_steps": 2405, "loss": 0.5637, "lr": 7.96886506814113e-05, "epoch": 0.6785992217898833, "percentage": 13.6, "elapsed_time": "4:55:55", "remaining_time": "1 day, 7:20:34"}
+{"current_steps": 328, "total_steps": 2405, "loss": 0.5603, "lr": 7.968137758642697e-05, "epoch": 0.680674448767834, "percentage": 13.64, "elapsed_time": "4:56:49", "remaining_time": "1 day, 7:19:38"}
+{"current_steps": 329, "total_steps": 2405, "loss": 0.5596, "lr": 7.967402085955541e-05, "epoch": 0.6827496757457847, "percentage": 13.68, "elapsed_time": "4:57:44", "remaining_time": "1 day, 7:18:45"}
+{"current_steps": 330, "total_steps": 2405, "loss": 0.558, "lr": 7.966658051630161e-05, "epoch": 0.6848249027237354, "percentage": 13.72, "elapsed_time": "4:58:38", "remaining_time": "1 day, 7:17:52"}
+{"current_steps": 331, "total_steps": 2405, "loss": 0.5567, "lr": 7.965905657234665e-05, "epoch": 0.6869001297016861, "percentage": 13.76, "elapsed_time": "4:59:33", "remaining_time": "1 day, 7:17:00"}
+{"current_steps": 332, "total_steps": 2405, "loss": 0.5512, "lr": 7.965144904354795e-05, "epoch": 0.6889753566796368, "percentage": 13.8, "elapsed_time": "5:00:28", "remaining_time": "1 day, 7:16:08"}
+{"current_steps": 333, "total_steps": 2405, "loss": 0.5555, "lr": 7.964375794593897e-05, "epoch": 0.6910505836575875, "percentage": 13.85, "elapsed_time": "5:01:22", "remaining_time": "1 day, 7:15:15"}
+{"current_steps": 334, "total_steps": 2405, "loss": 0.5471, "lr": 7.963598329572942e-05, "epoch": 0.6931258106355382, "percentage": 13.89, "elapsed_time": "5:02:16", "remaining_time": "1 day, 7:14:19"}
+{"current_steps": 335, "total_steps": 2405, "loss": 0.5642, "lr": 7.962812510930496e-05, "epoch": 0.695201037613489, "percentage": 13.93, "elapsed_time": "5:03:12", "remaining_time": "1 day, 7:13:32"}
+{"current_steps": 336, "total_steps": 2405, "loss": 0.5618, "lr": 7.962018340322746e-05, "epoch": 0.6972762645914397, "percentage": 13.97, "elapsed_time": "5:04:06", "remaining_time": "1 day, 7:12:37"}
+{"current_steps": 337, "total_steps": 2405, "loss": 0.5531, "lr": 7.961215819423467e-05, "epoch": 0.6993514915693904, "percentage": 14.01, "elapsed_time": "5:05:02", "remaining_time": "1 day, 7:11:52"}
+{"current_steps": 338, "total_steps": 2405, "loss": 0.5497, "lr": 7.960404949924049e-05, "epoch": 0.7014267185473411, "percentage": 14.05, "elapsed_time": "5:05:58", "remaining_time": "1 day, 7:11:09"}
+{"current_steps": 339, "total_steps": 2405, "loss": 0.5559, "lr": 7.959585733533461e-05, "epoch": 0.7035019455252919, "percentage": 14.1, "elapsed_time": "5:06:53", "remaining_time": "1 day, 7:10:17"}
+{"current_steps": 340, "total_steps": 2405, "loss": 0.5642, "lr": 7.958758171978275e-05, "epoch": 0.7055771725032426, "percentage": 14.14, "elapsed_time": "5:07:47", "remaining_time": "1 day, 7:09:24"}
+{"current_steps": 341, "total_steps": 2405, "loss": 0.5621, "lr": 7.957922267002647e-05, "epoch": 0.7076523994811933, "percentage": 14.18, "elapsed_time": "5:08:42", "remaining_time": "1 day, 7:08:34"}
+{"current_steps": 342, "total_steps": 2405, "loss": 0.5563, "lr": 7.957078020368317e-05, "epoch": 0.709727626459144, "percentage": 14.22, "elapsed_time": "5:09:37", "remaining_time": "1 day, 7:07:42"}
+{"current_steps": 343, "total_steps": 2405, "loss": 0.556, "lr": 7.956225433854608e-05, "epoch": 0.7118028534370947, "percentage": 14.26, "elapsed_time": "5:10:32", "remaining_time": "1 day, 7:06:51"}
+{"current_steps": 344, "total_steps": 2405, "loss": 0.5631, "lr": 7.955364509258418e-05, "epoch": 0.7138780804150454, "percentage": 14.3, "elapsed_time": "5:11:27", "remaining_time": "1 day, 7:06:02"}
+{"current_steps": 345, "total_steps": 2405, "loss": 0.5591, "lr": 7.95449524839422e-05, "epoch": 0.7159533073929961, "percentage": 14.35, "elapsed_time": "5:12:22", "remaining_time": "1 day, 7:05:08"}
+{"current_steps": 346, "total_steps": 2405, "loss": 0.5601, "lr": 7.953617653094056e-05, "epoch": 0.7180285343709468, "percentage": 14.39, "elapsed_time": "5:13:15", "remaining_time": "1 day, 7:04:09"}
+{"current_steps": 347, "total_steps": 2405, "loss": 0.5688, "lr": 7.952731725207532e-05, "epoch": 0.7201037613488975, "percentage": 14.43, "elapsed_time": "5:14:10", "remaining_time": "1 day, 7:03:18"}
+{"current_steps": 348, "total_steps": 2405, "loss": 0.5519, "lr": 7.951837466601815e-05, "epoch": 0.7221789883268482, "percentage": 14.47, "elapsed_time": "5:15:06", "remaining_time": "1 day, 7:02:32"}
+{"current_steps": 349, "total_steps": 2405, "loss": 0.5557, "lr": 7.950934879161632e-05, "epoch": 0.7242542153047989, "percentage": 14.51, "elapsed_time": "5:16:00", "remaining_time": "1 day, 7:01:39"}
+{"current_steps": 350, "total_steps": 2405, "loss": 0.5612, "lr": 7.950023964789264e-05, "epoch": 0.7263294422827496, "percentage": 14.55, "elapsed_time": "5:16:55", "remaining_time": "1 day, 7:00:45"}
+{"current_steps": 351, "total_steps": 2405, "loss": 0.5544, "lr": 7.94910472540454e-05, "epoch": 0.7284046692607004, "percentage": 14.59, "elapsed_time": "5:17:49", "remaining_time": "1 day, 6:59:53"}
+{"current_steps": 352, "total_steps": 2405, "loss": 0.5577, "lr": 7.948177162944839e-05, "epoch": 0.7304798962386511, "percentage": 14.64, "elapsed_time": "5:18:43", "remaining_time": "1 day, 6:58:56"}
+{"current_steps": 353, "total_steps": 2405, "loss": 0.5598, "lr": 7.947241279365072e-05, "epoch": 0.7325551232166018, "percentage": 14.68, "elapsed_time": "5:19:39", "remaining_time": "1 day, 6:58:10"}
+{"current_steps": 354, "total_steps": 2405, "loss": 0.5557, "lr": 7.946297076637696e-05, "epoch": 0.7346303501945526, "percentage": 14.72, "elapsed_time": "5:20:33", "remaining_time": "1 day, 6:57:16"}
+{"current_steps": 355, "total_steps": 2405, "loss": 0.5523, "lr": 7.945344556752698e-05, "epoch": 0.7367055771725033, "percentage": 14.76, "elapsed_time": "5:21:28", "remaining_time": "1 day, 6:56:23"}
+{"current_steps": 356, "total_steps": 2405, "loss": 0.5586, "lr": 7.944383721717599e-05, "epoch": 0.738780804150454, "percentage": 14.8, "elapsed_time": "5:22:24", "remaining_time": "1 day, 6:55:38"}
+{"current_steps": 357, "total_steps": 2405, "loss": 0.5568, "lr": 7.943414573557435e-05, "epoch": 0.7408560311284047, "percentage": 14.84, "elapsed_time": "5:23:18", "remaining_time": "1 day, 6:54:46"}
+{"current_steps": 358, "total_steps": 2405, "loss": 0.5543, "lr": 7.942437114314771e-05, "epoch": 0.7429312581063554, "percentage": 14.89, "elapsed_time": "5:24:13", "remaining_time": "1 day, 6:53:50"}
+{"current_steps": 359, "total_steps": 2405, "loss": 0.5508, "lr": 7.941451346049687e-05, "epoch": 0.7450064850843061, "percentage": 14.93, "elapsed_time": "5:25:07", "remaining_time": "1 day, 6:52:58"}
+{"current_steps": 360, "total_steps": 2405, "loss": 0.5535, "lr": 7.940457270839772e-05, "epoch": 0.7470817120622568, "percentage": 14.97, "elapsed_time": "5:26:02", "remaining_time": "1 day, 6:52:06"}
+{"current_steps": 361, "total_steps": 2405, "loss": 0.5544, "lr": 7.939454890780126e-05, "epoch": 0.7491569390402075, "percentage": 15.01, "elapsed_time": "5:26:58", "remaining_time": "1 day, 6:51:23"}
+{"current_steps": 362, "total_steps": 2405, "loss": 0.5493, "lr": 7.938444207983346e-05, "epoch": 0.7512321660181582, "percentage": 15.05, "elapsed_time": "5:27:56", "remaining_time": "1 day, 6:50:45"}
+{"current_steps": 363, "total_steps": 2405, "loss": 0.5455, "lr": 7.937425224579537e-05, "epoch": 0.7533073929961089, "percentage": 15.09, "elapsed_time": "5:28:53", "remaining_time": "1 day, 6:50:05"}
+{"current_steps": 364, "total_steps": 2405, "loss": 0.5546, "lr": 7.936397942716293e-05, "epoch": 0.7553826199740596, "percentage": 15.14, "elapsed_time": "5:29:47", "remaining_time": "1 day, 6:49:11"}
+{"current_steps": 365, "total_steps": 2405, "loss": 0.5547, "lr": 7.935362364558695e-05, "epoch": 0.7574578469520103, "percentage": 15.18, "elapsed_time": "5:30:41", "remaining_time": "1 day, 6:48:16"}
+{"current_steps": 366, "total_steps": 2405, "loss": 0.5568, "lr": 7.934318492289315e-05, "epoch": 0.7595330739299611, "percentage": 15.22, "elapsed_time": "5:31:36", "remaining_time": "1 day, 6:47:22"}
+{"current_steps": 367, "total_steps": 2405, "loss": 0.554, "lr": 7.933266328108201e-05, "epoch": 0.7616083009079118, "percentage": 15.26, "elapsed_time": "5:32:30", "remaining_time": "1 day, 6:46:29"}
+{"current_steps": 368, "total_steps": 2405, "loss": 0.5528, "lr": 7.932205874232882e-05, "epoch": 0.7636835278858625, "percentage": 15.3, "elapsed_time": "5:33:24", "remaining_time": "1 day, 6:45:32"}
+{"current_steps": 369, "total_steps": 2405, "loss": 0.563, "lr": 7.93113713289835e-05, "epoch": 0.7657587548638133, "percentage": 15.34, "elapsed_time": "5:34:18", "remaining_time": "1 day, 6:44:33"}
+{"current_steps": 370, "total_steps": 2405, "loss": 0.5559, "lr": 7.930060106357072e-05, "epoch": 0.767833981841764, "percentage": 15.38, "elapsed_time": "5:35:11", "remaining_time": "1 day, 6:43:33"}
+{"current_steps": 371, "total_steps": 2405, "loss": 0.5452, "lr": 7.928974796878973e-05, "epoch": 0.7699092088197147, "percentage": 15.43, "elapsed_time": "5:36:05", "remaining_time": "1 day, 6:42:38"}
+{"current_steps": 372, "total_steps": 2405, "loss": 0.5558, "lr": 7.927881206751434e-05, "epoch": 0.7719844357976654, "percentage": 15.47, "elapsed_time": "5:37:00", "remaining_time": "1 day, 6:41:43"}
+{"current_steps": 373, "total_steps": 2405, "loss": 0.5583, "lr": 7.92677933827929e-05, "epoch": 0.7740596627756161, "percentage": 15.51, "elapsed_time": "5:37:55", "remaining_time": "1 day, 6:40:52"}
+{"current_steps": 374, "total_steps": 2405, "loss": 0.5541, "lr": 7.925669193784825e-05, "epoch": 0.7761348897535668, "percentage": 15.55, "elapsed_time": "5:38:52", "remaining_time": "1 day, 6:40:13"}
+{"current_steps": 375, "total_steps": 2405, "loss": 0.5552, "lr": 7.92455077560776e-05, "epoch": 0.7782101167315175, "percentage": 15.59, "elapsed_time": "5:39:46", "remaining_time": "1 day, 6:39:18"}
+{"current_steps": 376, "total_steps": 2405, "loss": 0.5581, "lr": 7.923424086105257e-05, "epoch": 0.7802853437094682, "percentage": 15.63, "elapsed_time": "5:40:41", "remaining_time": "1 day, 6:38:26"}
+{"current_steps": 377, "total_steps": 2405, "loss": 0.5538, "lr": 7.922289127651912e-05, "epoch": 0.7823605706874189, "percentage": 15.68, "elapsed_time": "5:41:35", "remaining_time": "1 day, 6:37:33"}
+{"current_steps": 378, "total_steps": 2405, "loss": 0.5561, "lr": 7.921145902639746e-05, "epoch": 0.7844357976653696, "percentage": 15.72, "elapsed_time": "5:42:29", "remaining_time": "1 day, 6:36:37"}
+{"current_steps": 379, "total_steps": 2405, "loss": 0.5515, "lr": 7.919994413478203e-05, "epoch": 0.7865110246433203, "percentage": 15.76, "elapsed_time": "5:43:24", "remaining_time": "1 day, 6:35:42"}
+{"current_steps": 380, "total_steps": 2405, "loss": 0.5593, "lr": 7.918834662594144e-05, "epoch": 0.788586251621271, "percentage": 15.8, "elapsed_time": "5:44:19", "remaining_time": "1 day, 6:34:51"}
+{"current_steps": 381, "total_steps": 2405, "loss": 0.5573, "lr": 7.917666652431843e-05, "epoch": 0.7906614785992218, "percentage": 15.84, "elapsed_time": "5:45:13", "remaining_time": "1 day, 6:33:55"}
+{"current_steps": 382, "total_steps": 2405, "loss": 0.5518, "lr": 7.916490385452983e-05, "epoch": 0.7927367055771725, "percentage": 15.88, "elapsed_time": "5:46:07", "remaining_time": "1 day, 6:33:01"}
+{"current_steps": 383, "total_steps": 2405, "loss": 0.5544, "lr": 7.915305864136644e-05, "epoch": 0.7948119325551232, "percentage": 15.93, "elapsed_time": "5:47:02", "remaining_time": "1 day, 6:32:07"}
+{"current_steps": 384, "total_steps": 2405, "loss": 0.5494, "lr": 7.914113090979308e-05, "epoch": 0.796887159533074, "percentage": 15.97, "elapsed_time": "5:47:56", "remaining_time": "1 day, 6:31:12"}
+{"current_steps": 385, "total_steps": 2405, "loss": 0.5534, "lr": 7.912912068494845e-05, "epoch": 0.7989623865110247, "percentage": 16.01, "elapsed_time": "5:48:50", "remaining_time": "1 day, 6:30:17"}
+{"current_steps": 386, "total_steps": 2405, "loss": 0.553, "lr": 7.911702799214512e-05, "epoch": 0.8010376134889754, "percentage": 16.05, "elapsed_time": "5:49:44", "remaining_time": "1 day, 6:29:22"}
+{"current_steps": 387, "total_steps": 2405, "loss": 0.5526, "lr": 7.910485285686948e-05, "epoch": 0.8031128404669261, "percentage": 16.09, "elapsed_time": "5:50:40", "remaining_time": "1 day, 6:28:33"}
+{"current_steps": 388, "total_steps": 2405, "loss": 0.5484, "lr": 7.909259530478167e-05, "epoch": 0.8051880674448768, "percentage": 16.13, "elapsed_time": "5:51:36", "remaining_time": "1 day, 6:27:46"}
+{"current_steps": 389, "total_steps": 2405, "loss": 0.5487, "lr": 7.908025536171552e-05, "epoch": 0.8072632944228275, "percentage": 16.17, "elapsed_time": "5:52:29", "remaining_time": "1 day, 6:26:49"}
+{"current_steps": 390, "total_steps": 2405, "loss": 0.551, "lr": 7.906783305367851e-05, "epoch": 0.8093385214007782, "percentage": 16.22, "elapsed_time": "5:53:24", "remaining_time": "1 day, 6:25:56"}
+{"current_steps": 391, "total_steps": 2405, "loss": 0.5509, "lr": 7.905532840685172e-05, "epoch": 0.8114137483787289, "percentage": 16.26, "elapsed_time": "5:54:18", "remaining_time": "1 day, 6:25:02"}
+{"current_steps": 392, "total_steps": 2405, "loss": 0.5519, "lr": 7.904274144758975e-05, "epoch": 0.8134889753566796, "percentage": 16.3, "elapsed_time": "5:55:13", "remaining_time": "1 day, 6:24:07"}
+{"current_steps": 393, "total_steps": 2405, "loss": 0.5434, "lr": 7.90300722024207e-05, "epoch": 0.8155642023346303, "percentage": 16.34, "elapsed_time": "5:56:08", "remaining_time": "1 day, 6:23:16"}
+{"current_steps": 394, "total_steps": 2405, "loss": 0.553, "lr": 7.901732069804606e-05, "epoch": 0.817639429312581, "percentage": 16.38, "elapsed_time": "5:57:02", "remaining_time": "1 day, 6:22:20"}
+{"current_steps": 395, "total_steps": 2405, "loss": 0.5525, "lr": 7.900448696134075e-05, "epoch": 0.8197146562905318, "percentage": 16.42, "elapsed_time": "5:57:56", "remaining_time": "1 day, 6:21:23"}
+{"current_steps": 396, "total_steps": 2405, "loss": 0.5571, "lr": 7.899157101935295e-05, "epoch": 0.8217898832684825, "percentage": 16.47, "elapsed_time": "5:58:49", "remaining_time": "1 day, 6:20:25"}
+{"current_steps": 397, "total_steps": 2405, "loss": 0.5497, "lr": 7.897857289930409e-05, "epoch": 0.8238651102464332, "percentage": 16.51, "elapsed_time": "5:59:44", "remaining_time": "1 day, 6:19:32"}
+{"current_steps": 398, "total_steps": 2405, "loss": 0.5547, "lr": 7.896549262858886e-05, "epoch": 0.8259403372243839, "percentage": 16.55, "elapsed_time": "6:00:38", "remaining_time": "1 day, 6:18:34"}
+{"current_steps": 399, "total_steps": 2405, "loss": 0.5617, "lr": 7.895233023477499e-05, "epoch": 0.8280155642023346, "percentage": 16.59, "elapsed_time": "6:01:32", "remaining_time": "1 day, 6:17:41"}
+{"current_steps": 400, "total_steps": 2405, "loss": 0.5502, "lr": 7.893908574560339e-05, "epoch": 0.8300907911802854, "percentage": 16.63, "elapsed_time": "6:02:26", "remaining_time": "1 day, 6:16:46"}
+{"current_steps": 401, "total_steps": 2405, "loss": 0.5533, "lr": 7.892575918898793e-05, "epoch": 0.8321660181582361, "percentage": 16.67, "elapsed_time": "6:03:21", "remaining_time": "1 day, 6:15:54"}
+{"current_steps": 402, "total_steps": 2405, "loss": 0.5587, "lr": 7.891235059301547e-05, "epoch": 0.8342412451361868, "percentage": 16.72, "elapsed_time": "6:04:16", "remaining_time": "1 day, 6:15:00"}
+{"current_steps": 403, "total_steps": 2405, "loss": 0.5487, "lr": 7.88988599859458e-05, "epoch": 0.8363164721141375, "percentage": 16.76, "elapsed_time": "6:05:11", "remaining_time": "1 day, 6:14:09"}
+{"current_steps": 404, "total_steps": 2405, "loss": 0.5455, "lr": 7.888528739621146e-05, "epoch": 0.8383916990920882, "percentage": 16.8, "elapsed_time": "6:06:05", "remaining_time": "1 day, 6:13:12"}
+{"current_steps": 405, "total_steps": 2405, "loss": 0.5508, "lr": 7.887163285241789e-05, "epoch": 0.8404669260700389, "percentage": 16.84, "elapsed_time": "6:06:59", "remaining_time": "1 day, 6:12:18"}
+{"current_steps": 406, "total_steps": 2405, "loss": 0.546, "lr": 7.885789638334318e-05, "epoch": 0.8425421530479896, "percentage": 16.88, "elapsed_time": "6:07:54", "remaining_time": "1 day, 6:11:26"}
+{"current_steps": 407, "total_steps": 2405, "loss": 0.5502, "lr": 7.884407801793813e-05, "epoch": 0.8446173800259403, "percentage": 16.92, "elapsed_time": "6:08:49", "remaining_time": "1 day, 6:10:33"}
+{"current_steps": 408, "total_steps": 2405, "loss": 0.5438, "lr": 7.883017778532612e-05, "epoch": 0.846692607003891, "percentage": 16.96, "elapsed_time": "6:09:43", "remaining_time": "1 day, 6:09:38"}
+{"current_steps": 409, "total_steps": 2405, "loss": 0.5509, "lr": 7.881619571480306e-05, "epoch": 0.8487678339818417, "percentage": 17.01, "elapsed_time": "6:10:37", "remaining_time": "1 day, 6:08:42"}
+{"current_steps": 410, "total_steps": 2405, "loss": 0.5428, "lr": 7.880213183583738e-05, "epoch": 0.8508430609597925, "percentage": 17.05, "elapsed_time": "6:11:32", "remaining_time": "1 day, 6:07:53"}
+{"current_steps": 411, "total_steps": 2405, "loss": 0.5531, "lr": 7.878798617806989e-05, "epoch": 0.8529182879377432, "percentage": 17.09, "elapsed_time": "6:12:27", "remaining_time": "1 day, 6:06:59"}
+{"current_steps": 412, "total_steps": 2405, "loss": 0.5487, "lr": 7.877375877131375e-05, "epoch": 0.8549935149156939, "percentage": 17.13, "elapsed_time": "6:13:21", "remaining_time": "1 day, 6:06:04"}
+{"current_steps": 413, "total_steps": 2405, "loss": 0.5525, "lr": 7.875944964555447e-05, "epoch": 0.8570687418936446, "percentage": 17.17, "elapsed_time": "6:14:16", "remaining_time": "1 day, 6:05:11"}
+{"current_steps": 414, "total_steps": 2405, "loss": 0.5478, "lr": 7.874505883094973e-05, "epoch": 0.8591439688715953, "percentage": 17.21, "elapsed_time": "6:15:10", "remaining_time": "1 day, 6:04:16"}
+{"current_steps": 415, "total_steps": 2405, "loss": 0.5423, "lr": 7.87305863578294e-05, "epoch": 0.8612191958495461, "percentage": 17.26, "elapsed_time": "6:16:04", "remaining_time": "1 day, 6:03:22"}
+{"current_steps": 416, "total_steps": 2405, "loss": 0.5469, "lr": 7.871603225669547e-05, "epoch": 0.8632944228274968, "percentage": 17.3, "elapsed_time": "6:16:59", "remaining_time": "1 day, 6:02:30"}
+{"current_steps": 417, "total_steps": 2405, "loss": 0.5453, "lr": 7.870139655822193e-05, "epoch": 0.8653696498054475, "percentage": 17.34, "elapsed_time": "6:17:54", "remaining_time": "1 day, 6:01:37"}
+{"current_steps": 418, "total_steps": 2405, "loss": 0.5455, "lr": 7.868667929325477e-05, "epoch": 0.8674448767833982, "percentage": 17.38, "elapsed_time": "6:18:48", "remaining_time": "1 day, 6:00:43"}
+{"current_steps": 419, "total_steps": 2405, "loss": 0.5479, "lr": 7.867188049281189e-05, "epoch": 0.8695201037613489, "percentage": 17.42, "elapsed_time": "6:19:43", "remaining_time": "1 day, 5:59:49"}
+{"current_steps": 420, "total_steps": 2405, "loss": 0.5441, "lr": 7.865700018808299e-05, "epoch": 0.8715953307392996, "percentage": 17.46, "elapsed_time": "6:20:37", "remaining_time": "1 day, 5:58:54"}
+{"current_steps": 421, "total_steps": 2405, "loss": 0.5482, "lr": 7.864203841042962e-05, "epoch": 0.8736705577172503, "percentage": 17.51, "elapsed_time": "6:21:31", "remaining_time": "1 day, 5:57:56"}
+{"current_steps": 422, "total_steps": 2405, "loss": 0.5402, "lr": 7.862699519138498e-05, "epoch": 0.875745784695201, "percentage": 17.55, "elapsed_time": "6:22:25", "remaining_time": "1 day, 5:57:00"}
+{"current_steps": 423, "total_steps": 2405, "loss": 0.551, "lr": 7.861187056265394e-05, "epoch": 0.8778210116731517, "percentage": 17.59, "elapsed_time": "6:23:19", "remaining_time": "1 day, 5:56:07"}
+{"current_steps": 424, "total_steps": 2405, "loss": 0.5482, "lr": 7.859666455611297e-05, "epoch": 0.8798962386511024, "percentage": 17.63, "elapsed_time": "6:24:14", "remaining_time": "1 day, 5:55:15"}
+{"current_steps": 425, "total_steps": 2405, "loss": 0.5462, "lr": 7.858137720380999e-05, "epoch": 0.8819714656290532, "percentage": 17.67, "elapsed_time": "6:25:08", "remaining_time": "1 day, 5:54:19"}
+{"current_steps": 426, "total_steps": 2405, "loss": 0.5426, "lr": 7.856600853796441e-05, "epoch": 0.8840466926070039, "percentage": 17.71, "elapsed_time": "6:26:02", "remaining_time": "1 day, 5:53:23"}
+{"current_steps": 427, "total_steps": 2405, "loss": 0.5457, "lr": 7.855055859096702e-05, "epoch": 0.8861219195849546, "percentage": 17.75, "elapsed_time": "6:26:57", "remaining_time": "1 day, 5:52:28"}
+{"current_steps": 428, "total_steps": 2405, "loss": 0.5454, "lr": 7.853502739537988e-05, "epoch": 0.8881971465629053, "percentage": 17.8, "elapsed_time": "6:27:51", "remaining_time": "1 day, 5:51:35"}
+{"current_steps": 429, "total_steps": 2405, "loss": 0.5524, "lr": 7.85194149839363e-05, "epoch": 0.890272373540856, "percentage": 17.84, "elapsed_time": "6:28:46", "remaining_time": "1 day, 5:50:41"}
+{"current_steps": 430, "total_steps": 2405, "loss": 0.5454, "lr": 7.850372138954081e-05, "epoch": 0.8923476005188068, "percentage": 17.88, "elapsed_time": "6:29:40", "remaining_time": "1 day, 5:49:48"}
+{"current_steps": 431, "total_steps": 2405, "loss": 0.5407, "lr": 7.848794664526896e-05, "epoch": 0.8944228274967575, "percentage": 17.92, "elapsed_time": "6:30:35", "remaining_time": "1 day, 5:48:53"}
+{"current_steps": 432, "total_steps": 2405, "loss": 0.5431, "lr": 7.84720907843674e-05, "epoch": 0.8964980544747082, "percentage": 17.96, "elapsed_time": "6:31:31", "remaining_time": "1 day, 5:48:09"}
+{"current_steps": 433, "total_steps": 2405, "loss": 0.5454, "lr": 7.845615384025367e-05, "epoch": 0.8985732814526589, "percentage": 18.0, "elapsed_time": "6:32:27", "remaining_time": "1 day, 5:47:20"}
+{"current_steps": 434, "total_steps": 2405, "loss": 0.5513, "lr": 7.844013584651627e-05, "epoch": 0.9006485084306096, "percentage": 18.05, "elapsed_time": "6:33:22", "remaining_time": "1 day, 5:46:30"}
+{"current_steps": 435, "total_steps": 2405, "loss": 0.5491, "lr": 7.842403683691448e-05, "epoch": 0.9027237354085603, "percentage": 18.09, "elapsed_time": "6:34:18", "remaining_time": "1 day, 5:45:43"}
+{"current_steps": 436, "total_steps": 2405, "loss": 0.5468, "lr": 7.840785684537834e-05, "epoch": 0.904798962386511, "percentage": 18.13, "elapsed_time": "6:35:14", "remaining_time": "1 day, 5:44:55"}
+{"current_steps": 437, "total_steps": 2405, "loss": 0.5486, "lr": 7.839159590600854e-05, "epoch": 0.9068741893644617, "percentage": 18.17, "elapsed_time": "6:36:10", "remaining_time": "1 day, 5:44:06"}
+{"current_steps": 438, "total_steps": 2405, "loss": 0.5448, "lr": 7.837525405307642e-05, "epoch": 0.9089494163424124, "percentage": 18.21, "elapsed_time": "6:37:05", "remaining_time": "1 day, 5:43:15"}
+{"current_steps": 439, "total_steps": 2405, "loss": 0.5552, "lr": 7.835883132102382e-05, "epoch": 0.9110246433203631, "percentage": 18.25, "elapsed_time": "6:37:59", "remaining_time": "1 day, 5:42:20"}
+{"current_steps": 440, "total_steps": 2405, "loss": 0.552, "lr": 7.834232774446306e-05, "epoch": 0.9130998702983139, "percentage": 18.3, "elapsed_time": "6:38:54", "remaining_time": "1 day, 5:41:27"}
+{"current_steps": 441, "total_steps": 2405, "loss": 0.5482, "lr": 7.83257433581768e-05, "epoch": 0.9151750972762646, "percentage": 18.34, "elapsed_time": "6:39:48", "remaining_time": "1 day, 5:40:33"}
+{"current_steps": 442, "total_steps": 2405, "loss": 0.5425, "lr": 7.83090781971181e-05, "epoch": 0.9172503242542153, "percentage": 18.38, "elapsed_time": "6:40:44", "remaining_time": "1 day, 5:39:47"}
+{"current_steps": 443, "total_steps": 2405, "loss": 0.5478, "lr": 7.829233229641017e-05, "epoch": 0.919325551232166, "percentage": 18.42, "elapsed_time": "6:41:39", "remaining_time": "1 day, 5:38:54"}
+{"current_steps": 444, "total_steps": 2405, "loss": 0.5423, "lr": 7.827550569134643e-05, "epoch": 0.9214007782101167, "percentage": 18.46, "elapsed_time": "6:42:33", "remaining_time": "1 day, 5:37:59"}
+{"current_steps": 445, "total_steps": 2405, "loss": 0.5519, "lr": 7.82585984173904e-05, "epoch": 0.9234760051880675, "percentage": 18.5, "elapsed_time": "6:43:28", "remaining_time": "1 day, 5:37:07"}
+{"current_steps": 446, "total_steps": 2405, "loss": 0.5428, "lr": 7.824161051017557e-05, "epoch": 0.9255512321660182, "percentage": 18.54, "elapsed_time": "6:44:24", "remaining_time": "1 day, 5:36:20"}
+{"current_steps": 447, "total_steps": 2405, "loss": 0.5435, "lr": 7.822454200550545e-05, "epoch": 0.9276264591439689, "percentage": 18.59, "elapsed_time": "6:45:21", "remaining_time": "1 day, 5:35:33"}
+{"current_steps": 448, "total_steps": 2405, "loss": 0.5381, "lr": 7.820739293935333e-05, "epoch": 0.9297016861219196, "percentage": 18.63, "elapsed_time": "6:46:16", "remaining_time": "1 day, 5:34:42"}
+{"current_steps": 449, "total_steps": 2405, "loss": 0.5451, "lr": 7.819016334786234e-05, "epoch": 0.9317769130998703, "percentage": 18.67, "elapsed_time": "6:47:11", "remaining_time": "1 day, 5:33:53"}
+{"current_steps": 450, "total_steps": 2405, "loss": 0.5467, "lr": 7.817285326734534e-05, "epoch": 0.933852140077821, "percentage": 18.71, "elapsed_time": "6:48:07", "remaining_time": "1 day, 5:33:04"}
+{"current_steps": 451, "total_steps": 2405, "loss": 0.5458, "lr": 7.815546273428474e-05, "epoch": 0.9359273670557717, "percentage": 18.75, "elapsed_time": "6:49:02", "remaining_time": "1 day, 5:32:10"}
+{"current_steps": 452, "total_steps": 2405, "loss": 0.5451, "lr": 7.813799178533263e-05, "epoch": 0.9380025940337224, "percentage": 18.79, "elapsed_time": "6:49:56", "remaining_time": "1 day, 5:31:17"}
+{"current_steps": 453, "total_steps": 2405, "loss": 0.5481, "lr": 7.81204404573105e-05, "epoch": 0.9400778210116731, "percentage": 18.84, "elapsed_time": "6:50:51", "remaining_time": "1 day, 5:30:22"}
+{"current_steps": 454, "total_steps": 2405, "loss": 0.5388, "lr": 7.810280878720929e-05, "epoch": 0.9421530479896238, "percentage": 18.88, "elapsed_time": "6:51:46", "remaining_time": "1 day, 5:29:30"}
+{"current_steps": 455, "total_steps": 2405, "loss": 0.5488, "lr": 7.808509681218921e-05, "epoch": 0.9442282749675746, "percentage": 18.92, "elapsed_time": "6:52:41", "remaining_time": "1 day, 5:28:41"}
+{"current_steps": 456, "total_steps": 2405, "loss": 0.5502, "lr": 7.806730456957979e-05, "epoch": 0.9463035019455253, "percentage": 18.96, "elapsed_time": "6:53:35", "remaining_time": "1 day, 5:27:44"}
+{"current_steps": 457, "total_steps": 2405, "loss": 0.5427, "lr": 7.804943209687969e-05, "epoch": 0.948378728923476, "percentage": 19.0, "elapsed_time": "6:54:30", "remaining_time": "1 day, 5:26:54"}
+{"current_steps": 458, "total_steps": 2405, "loss": 0.5467, "lr": 7.803147943175665e-05, "epoch": 0.9504539559014267, "percentage": 19.04, "elapsed_time": "6:55:25", "remaining_time": "1 day, 5:25:58"}
+{"current_steps": 459, "total_steps": 2405, "loss": 0.5398, "lr": 7.801344661204749e-05, "epoch": 0.9525291828793774, "percentage": 19.09, "elapsed_time": "6:56:20", "remaining_time": "1 day, 5:25:07"}
+{"current_steps": 460, "total_steps": 2405, "loss": 0.5498, "lr": 7.799533367575787e-05, "epoch": 0.9546044098573282, "percentage": 19.13, "elapsed_time": "6:57:15", "remaining_time": "1 day, 5:24:15"}
+{"current_steps": 461, "total_steps": 2405, "loss": 0.538, "lr": 7.797714066106236e-05, "epoch": 0.9566796368352789, "percentage": 19.17, "elapsed_time": "6:58:11", "remaining_time": "1 day, 5:23:28"}
+{"current_steps": 462, "total_steps": 2405, "loss": 0.5511, "lr": 7.795886760630432e-05, "epoch": 0.9587548638132296, "percentage": 19.21, "elapsed_time": "6:59:06", "remaining_time": "1 day, 5:22:35"}
+{"current_steps": 463, "total_steps": 2405, "loss": 0.5376, "lr": 7.794051454999573e-05, "epoch": 0.9608300907911803, "percentage": 19.25, "elapsed_time": "7:00:00", "remaining_time": "1 day, 5:21:41"}
+{"current_steps": 464, "total_steps": 2405, "loss": 0.5486, "lr": 7.792208153081723e-05, "epoch": 0.962905317769131, "percentage": 19.29, "elapsed_time": "7:00:55", "remaining_time": "1 day, 5:20:48"}
+{"current_steps": 465, "total_steps": 2405, "loss": 0.5469, "lr": 7.790356858761799e-05, "epoch": 0.9649805447470817, "percentage": 19.33, "elapsed_time": "7:01:50", "remaining_time": "1 day, 5:19:58"}
+{"current_steps": 466, "total_steps": 2405, "loss": 0.5368, "lr": 7.788497575941562e-05, "epoch": 0.9670557717250324, "percentage": 19.38, "elapsed_time": "7:02:45", "remaining_time": "1 day, 5:19:05"}
+{"current_steps": 467, "total_steps": 2405, "loss": 0.5462, "lr": 7.78663030853961e-05, "epoch": 0.9691309987029831, "percentage": 19.42, "elapsed_time": "7:03:40", "remaining_time": "1 day, 5:18:12"}
+{"current_steps": 468, "total_steps": 2405, "loss": 0.5392, "lr": 7.784755060491365e-05, "epoch": 0.9712062256809338, "percentage": 19.46, "elapsed_time": "7:04:34", "remaining_time": "1 day, 5:17:17"}
+{"current_steps": 469, "total_steps": 2405, "loss": 0.5468, "lr": 7.782871835749073e-05, "epoch": 0.9732814526588845, "percentage": 19.5, "elapsed_time": "7:05:29", "remaining_time": "1 day, 5:16:25"}
+{"current_steps": 470, "total_steps": 2405, "loss": 0.5502, "lr": 7.780980638281793e-05, "epoch": 0.9753566796368353, "percentage": 19.54, "elapsed_time": "7:06:23", "remaining_time": "1 day, 5:15:28"}
+{"current_steps": 471, "total_steps": 2405, "loss": 0.5453, "lr": 7.779081472075383e-05, "epoch": 0.977431906614786, "percentage": 19.58, "elapsed_time": "7:07:18", "remaining_time": "1 day, 5:14:34"}
+{"current_steps": 472, "total_steps": 2405, "loss": 0.5338, "lr": 7.7771743411325e-05, "epoch": 0.9795071335927367, "percentage": 19.63, "elapsed_time": "7:08:12", "remaining_time": "1 day, 5:13:41"}
+{"current_steps": 473, "total_steps": 2405, "loss": 0.5414, "lr": 7.77525924947258e-05, "epoch": 0.9815823605706874, "percentage": 19.67, "elapsed_time": "7:09:07", "remaining_time": "1 day, 5:12:49"}
+{"current_steps": 474, "total_steps": 2405, "loss": 0.5463, "lr": 7.773336201131849e-05, "epoch": 0.9836575875486381, "percentage": 19.71, "elapsed_time": "7:10:02", "remaining_time": "1 day, 5:11:57"}
+{"current_steps": 475, "total_steps": 2405, "loss": 0.543, "lr": 7.771405200163291e-05, "epoch": 0.9857328145265889, "percentage": 19.75, "elapsed_time": "7:10:58", "remaining_time": "1 day, 5:11:06"}
+{"current_steps": 476, "total_steps": 2405, "loss": 0.5371, "lr": 7.769466250636654e-05, "epoch": 0.9878080415045396, "percentage": 19.79, "elapsed_time": "7:11:52", "remaining_time": "1 day, 5:10:09"}
+{"current_steps": 477, "total_steps": 2405, "loss": 0.5313, "lr": 7.767519356638444e-05, "epoch": 0.9898832684824903, "percentage": 19.83, "elapsed_time": "7:12:46", "remaining_time": "1 day, 5:09:15"}
+{"current_steps": 478, "total_steps": 2405, "loss": 0.5374, "lr": 7.765564522271902e-05, "epoch": 0.991958495460441, "percentage": 19.88, "elapsed_time": "7:13:41", "remaining_time": "1 day, 5:08:22"}
+{"current_steps": 479, "total_steps": 2405, "loss": 0.5474, "lr": 7.76360175165701e-05, "epoch": 0.9940337224383917, "percentage": 19.92, "elapsed_time": "7:14:36", "remaining_time": "1 day, 5:07:30"}
+{"current_steps": 480, "total_steps": 2405, "loss": 0.5352, "lr": 7.761631048930476e-05, "epoch": 0.9961089494163424, "percentage": 19.96, "elapsed_time": "7:15:30", "remaining_time": "1 day, 5:06:36"}
+{"current_steps": 481, "total_steps": 2405, "loss": 0.5439, "lr": 7.75965241824572e-05, "epoch": 0.9981841763942931, "percentage": 20.0, "elapsed_time": "7:16:25", "remaining_time": "1 day, 5:05:41"}
+{"current_steps": 482, "total_steps": 2405, "loss": 0.6622, "lr": 7.757665863772879e-05, "epoch": 1.0005188067444877, "percentage": 20.04, "elapsed_time": "7:19:08", "remaining_time": "1 day, 5:12:00"}
+{"current_steps": 483, "total_steps": 2405, "loss": 0.5146, "lr": 7.755671389698783e-05, "epoch": 1.0025940337224384, "percentage": 20.08, "elapsed_time": "7:20:01", "remaining_time": "1 day, 5:10:59"}
+{"current_steps": 484, "total_steps": 2405, "loss": 0.5111, "lr": 7.753669000226957e-05, "epoch": 1.004669260700389, "percentage": 20.12, "elapsed_time": "7:20:55", "remaining_time": "1 day, 5:10:03"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f01b43956433cd0b7ab1cac66dd1f2f2c6549c746608e6ccaaba0201c8cab54
+size 7224