Training in progress, step 400

.gitattributes

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

added_tokens.json

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

config.json

@@ -0,0 +1,29 @@
+{
+  "_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 sha256:9b2cad542652d7648a80829299f026ec0cfaf9eb39222e694e1cde0c4f62e57e
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83139fbadc93fc547a513ff7c709c9f111e0f5ecba0e8e485f0785fa72570a5b
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c71f501a4b8384a8ca18778b7de7fab474482c3c2e540ad2d1b17ee7fa032d3
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4008887da8516a9b963b08f011c24d99e34b8700f1dee9e54563bab15dc7eb0
+size 1089994880

model.safetensors.index.json

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

special_tokens_map.json

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

tokenizer.json

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

tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "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": "<|im_end|>",
+  "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 +1,439 @@
+{"current_steps": 1, "total_steps": 876, "loss": 1.5414, "lr": 1.1363636363636364e-07, "epoch": 0.003419777714448561, "percentage": 0.11, "elapsed_time": "0:00:43", "remaining_time": "10:31:19"}
+{"current_steps": 2, "total_steps": 876, "loss": 1.6625, "lr": 2.2727272727272729e-07, "epoch": 0.006839555428897122, "percentage": 0.23, "elapsed_time": "0:01:26", "remaining_time": "10:33:24"}
+{"current_steps": 3, "total_steps": 876, "loss": 1.5932, "lr": 3.409090909090909e-07, "epoch": 0.010259333143345683, "percentage": 0.34, "elapsed_time": "0:02:07", "remaining_time": "10:18:53"}
+{"current_steps": 4, "total_steps": 876, "loss": 1.5567, "lr": 4.5454545454545457e-07, "epoch": 0.013679110857794243, "percentage": 0.46, "elapsed_time": "0:02:49", "remaining_time": "10:17:34"}
+{"current_steps": 5, "total_steps": 876, "loss": 1.6347, "lr": 5.681818181818182e-07, "epoch": 0.017098888572242805, "percentage": 0.57, "elapsed_time": "0:03:31", "remaining_time": "10:13:29"}
+{"current_steps": 6, "total_steps": 876, "loss": 1.578, "lr": 6.818181818181818e-07, "epoch": 0.020518666286691366, "percentage": 0.68, "elapsed_time": "0:04:12", "remaining_time": "10:11:19"}
+{"current_steps": 7, "total_steps": 876, "loss": 1.6158, "lr": 7.954545454545455e-07, "epoch": 0.023938444001139925, "percentage": 0.8, "elapsed_time": "0:04:54", "remaining_time": "10:10:18"}
+{"current_steps": 8, "total_steps": 876, "loss": 1.5234, "lr": 9.090909090909091e-07, "epoch": 0.027358221715588486, "percentage": 0.91, "elapsed_time": "0:05:36", "remaining_time": "10:08:31"}
+{"current_steps": 9, "total_steps": 876, "loss": 1.4546, "lr": 1.0227272727272729e-06, "epoch": 0.030777999430037048, "percentage": 1.03, "elapsed_time": "0:06:18", "remaining_time": "10:07:25"}
+{"current_steps": 10, "total_steps": 876, "loss": 1.4514, "lr": 1.1363636363636364e-06, "epoch": 0.03419777714448561, "percentage": 1.14, "elapsed_time": "0:07:00", "remaining_time": "10:06:35"}
+{"current_steps": 11, "total_steps": 876, "loss": 1.4477, "lr": 1.25e-06, "epoch": 0.03761755485893417, "percentage": 1.26, "elapsed_time": "0:07:41", "remaining_time": "10:05:14"}
+{"current_steps": 12, "total_steps": 876, "loss": 1.44, "lr": 1.3636363636363636e-06, "epoch": 0.04103733257338273, "percentage": 1.37, "elapsed_time": "0:08:25", "remaining_time": "10:06:20"}
+{"current_steps": 13, "total_steps": 876, "loss": 1.3792, "lr": 1.4772727272727275e-06, "epoch": 0.04445711028783129, "percentage": 1.48, "elapsed_time": "0:09:05", "remaining_time": "10:03:35"}
+{"current_steps": 14, "total_steps": 876, "loss": 1.1825, "lr": 1.590909090909091e-06, "epoch": 0.04787688800227985, "percentage": 1.6, "elapsed_time": "0:09:51", "remaining_time": "10:06:38"}
+{"current_steps": 15, "total_steps": 876, "loss": 1.3114, "lr": 1.7045454545454546e-06, "epoch": 0.05129666571672841, "percentage": 1.71, "elapsed_time": "0:10:34", "remaining_time": "10:06:32"}
+{"current_steps": 16, "total_steps": 876, "loss": 1.2936, "lr": 1.8181818181818183e-06, "epoch": 0.05471644343117697, "percentage": 1.83, "elapsed_time": "0:11:15", "remaining_time": "10:05:23"}
+{"current_steps": 17, "total_steps": 876, "loss": 1.2741, "lr": 1.931818181818182e-06, "epoch": 0.058136221145625534, "percentage": 1.94, "elapsed_time": "0:11:57", "remaining_time": "10:04:19"}
+{"current_steps": 18, "total_steps": 876, "loss": 1.1998, "lr": 2.0454545454545457e-06, "epoch": 0.061555998860074096, "percentage": 2.05, "elapsed_time": "0:12:38", "remaining_time": "10:02:20"}
+{"current_steps": 19, "total_steps": 876, "loss": 1.2239, "lr": 2.1590909090909092e-06, "epoch": 0.06497577657452265, "percentage": 2.17, "elapsed_time": "0:13:22", "remaining_time": "10:03:06"}
+{"current_steps": 20, "total_steps": 876, "loss": 1.1305, "lr": 2.2727272727272728e-06, "epoch": 0.06839555428897122, "percentage": 2.28, "elapsed_time": "0:14:03", "remaining_time": "10:01:30"}
+{"current_steps": 21, "total_steps": 876, "loss": 1.1247, "lr": 2.3863636363636367e-06, "epoch": 0.07181533200341977, "percentage": 2.4, "elapsed_time": "0:14:46", "remaining_time": "10:01:14"}
+{"current_steps": 22, "total_steps": 876, "loss": 1.1385, "lr": 2.5e-06, "epoch": 0.07523510971786834, "percentage": 2.51, "elapsed_time": "0:15:29", "remaining_time": "10:01:14"}
+{"current_steps": 23, "total_steps": 876, "loss": 1.1009, "lr": 2.6136363636363637e-06, "epoch": 0.0786548874323169, "percentage": 2.63, "elapsed_time": "0:16:10", "remaining_time": "9:59:48"}
+{"current_steps": 24, "total_steps": 876, "loss": 1.0951, "lr": 2.7272727272727272e-06, "epoch": 0.08207466514676547, "percentage": 2.74, "elapsed_time": "0:16:52", "remaining_time": "9:58:58"}
+{"current_steps": 25, "total_steps": 876, "loss": 1.0731, "lr": 2.8409090909090916e-06, "epoch": 0.08549444286121402, "percentage": 2.85, "elapsed_time": "0:17:33", "remaining_time": "9:57:26"}
+{"current_steps": 26, "total_steps": 876, "loss": 1.0945, "lr": 2.954545454545455e-06, "epoch": 0.08891422057566258, "percentage": 2.97, "elapsed_time": "0:18:15", "remaining_time": "9:56:42"}
+{"current_steps": 27, "total_steps": 876, "loss": 0.9886, "lr": 3.0681818181818186e-06, "epoch": 0.09233399829011114, "percentage": 3.08, "elapsed_time": "0:18:56", "remaining_time": "9:55:22"}
+{"current_steps": 28, "total_steps": 876, "loss": 0.9824, "lr": 3.181818181818182e-06, "epoch": 0.0957537760045597, "percentage": 3.2, "elapsed_time": "0:19:39", "remaining_time": "9:55:16"}
+{"current_steps": 29, "total_steps": 876, "loss": 1.0123, "lr": 3.2954545454545456e-06, "epoch": 0.09917355371900827, "percentage": 3.31, "elapsed_time": "0:20:21", "remaining_time": "9:54:43"}
+{"current_steps": 30, "total_steps": 876, "loss": 1.0924, "lr": 3.409090909090909e-06, "epoch": 0.10259333143345682, "percentage": 3.42, "elapsed_time": "0:21:04", "remaining_time": "9:54:28"}
+{"current_steps": 31, "total_steps": 876, "loss": 1.0842, "lr": 3.522727272727273e-06, "epoch": 0.10601310914790539, "percentage": 3.54, "elapsed_time": "0:21:45", "remaining_time": "9:53:03"}
+{"current_steps": 32, "total_steps": 876, "loss": 0.9534, "lr": 3.6363636363636366e-06, "epoch": 0.10943288686235395, "percentage": 3.65, "elapsed_time": "0:22:26", "remaining_time": "9:51:44"}
+{"current_steps": 33, "total_steps": 876, "loss": 1.0274, "lr": 3.7500000000000005e-06, "epoch": 0.11285266457680251, "percentage": 3.77, "elapsed_time": "0:23:07", "remaining_time": "9:50:47"}
+{"current_steps": 34, "total_steps": 876, "loss": 0.9753, "lr": 3.863636363636364e-06, "epoch": 0.11627244229125107, "percentage": 3.88, "elapsed_time": "0:23:48", "remaining_time": "9:49:47"}
+{"current_steps": 35, "total_steps": 876, "loss": 0.9402, "lr": 3.9772727272727275e-06, "epoch": 0.11969222000569962, "percentage": 4.0, "elapsed_time": "0:24:31", "remaining_time": "9:49:06"}
+{"current_steps": 36, "total_steps": 876, "loss": 0.9462, "lr": 4.0909090909090915e-06, "epoch": 0.12311199772014819, "percentage": 4.11, "elapsed_time": "0:25:15", "remaining_time": "9:49:15"}
+{"current_steps": 1, "total_steps": 876, "loss": 1.5414, "lr": 1.1363636363636364e-07, "epoch": 0.003419777714448561, "percentage": 0.11, "elapsed_time": "0:00:43", "remaining_time": "10:30:35"}
+{"current_steps": 2, "total_steps": 876, "loss": 1.6625, "lr": 2.2727272727272729e-07, "epoch": 0.006839555428897122, "percentage": 0.23, "elapsed_time": "0:01:26", "remaining_time": "10:32:25"}
+{"current_steps": 3, "total_steps": 876, "loss": 1.5932, "lr": 3.409090909090909e-07, "epoch": 0.010259333143345683, "percentage": 0.34, "elapsed_time": "0:02:07", "remaining_time": "10:18:18"}
+{"current_steps": 4, "total_steps": 876, "loss": 1.5571, "lr": 4.5454545454545457e-07, "epoch": 0.013679110857794243, "percentage": 0.46, "elapsed_time": "0:02:49", "remaining_time": "10:16:59"}
+{"current_steps": 5, "total_steps": 876, "loss": 1.6355, "lr": 5.681818181818182e-07, "epoch": 0.017098888572242805, "percentage": 0.57, "elapsed_time": "0:03:31", "remaining_time": "10:12:54"}
+{"current_steps": 6, "total_steps": 876, "loss": 1.5772, "lr": 6.818181818181818e-07, "epoch": 0.020518666286691366, "percentage": 0.68, "elapsed_time": "0:04:12", "remaining_time": "10:10:44"}
+{"current_steps": 7, "total_steps": 876, "loss": 1.6161, "lr": 7.954545454545455e-07, "epoch": 0.023938444001139925, "percentage": 0.8, "elapsed_time": "0:04:54", "remaining_time": "10:09:42"}
+{"current_steps": 8, "total_steps": 876, "loss": 1.5239, "lr": 9.090909090909091e-07, "epoch": 0.027358221715588486, "percentage": 0.91, "elapsed_time": "0:05:36", "remaining_time": "10:08:05"}
+{"current_steps": 9, "total_steps": 876, "loss": 1.455, "lr": 1.0227272727272729e-06, "epoch": 0.030777999430037048, "percentage": 1.03, "elapsed_time": "0:06:18", "remaining_time": "10:07:01"}
+{"current_steps": 10, "total_steps": 876, "loss": 1.4506, "lr": 1.1363636363636364e-06, "epoch": 0.03419777714448561, "percentage": 1.14, "elapsed_time": "0:06:59", "remaining_time": "10:06:09"}
+{"current_steps": 11, "total_steps": 876, "loss": 1.447, "lr": 1.25e-06, "epoch": 0.03761755485893417, "percentage": 1.26, "elapsed_time": "0:07:41", "remaining_time": "10:04:48"}
+{"current_steps": 12, "total_steps": 876, "loss": 1.4402, "lr": 1.3636363636363636e-06, "epoch": 0.04103733257338273, "percentage": 1.37, "elapsed_time": "0:08:24", "remaining_time": "10:05:52"}
+{"current_steps": 13, "total_steps": 876, "loss": 1.3787, "lr": 1.4772727272727275e-06, "epoch": 0.04445711028783129, "percentage": 1.48, "elapsed_time": "0:09:05", "remaining_time": "10:03:04"}
+{"current_steps": 14, "total_steps": 876, "loss": 1.1824, "lr": 1.590909090909091e-06, "epoch": 0.04787688800227985, "percentage": 1.6, "elapsed_time": "0:09:50", "remaining_time": "10:06:04"}
+{"current_steps": 15, "total_steps": 876, "loss": 1.3111, "lr": 1.7045454545454546e-06, "epoch": 0.05129666571672841, "percentage": 1.71, "elapsed_time": "0:10:33", "remaining_time": "10:05:58"}
+{"current_steps": 16, "total_steps": 876, "loss": 1.2942, "lr": 1.8181818181818183e-06, "epoch": 0.05471644343117697, "percentage": 1.83, "elapsed_time": "0:11:15", "remaining_time": "10:04:54"}
+{"current_steps": 17, "total_steps": 876, "loss": 1.2742, "lr": 1.931818181818182e-06, "epoch": 0.058136221145625534, "percentage": 1.94, "elapsed_time": "0:11:57", "remaining_time": "10:03:52"}
+{"current_steps": 18, "total_steps": 876, "loss": 1.1996, "lr": 2.0454545454545457e-06, "epoch": 0.061555998860074096, "percentage": 2.05, "elapsed_time": "0:12:37", "remaining_time": "10:01:56"}
+{"current_steps": 19, "total_steps": 876, "loss": 1.224, "lr": 2.1590909090909092e-06, "epoch": 0.06497577657452265, "percentage": 2.17, "elapsed_time": "0:13:21", "remaining_time": "10:02:41"}
+{"current_steps": 20, "total_steps": 876, "loss": 1.1313, "lr": 2.2727272727272728e-06, "epoch": 0.06839555428897122, "percentage": 2.28, "elapsed_time": "0:14:02", "remaining_time": "10:01:06"}
+{"current_steps": 21, "total_steps": 876, "loss": 1.125, "lr": 2.3863636363636367e-06, "epoch": 0.07181533200341977, "percentage": 2.4, "elapsed_time": "0:14:45", "remaining_time": "10:00:48"}
+{"current_steps": 22, "total_steps": 876, "loss": 1.139, "lr": 2.5e-06, "epoch": 0.07523510971786834, "percentage": 2.51, "elapsed_time": "0:15:28", "remaining_time": "10:00:47"}
+{"current_steps": 23, "total_steps": 876, "loss": 1.1007, "lr": 2.6136363636363637e-06, "epoch": 0.0786548874323169, "percentage": 2.63, "elapsed_time": "0:16:09", "remaining_time": "9:59:23"}
+{"current_steps": 24, "total_steps": 876, "loss": 1.0949, "lr": 2.7272727272727272e-06, "epoch": 0.08207466514676547, "percentage": 2.74, "elapsed_time": "0:16:51", "remaining_time": "9:58:36"}
+{"current_steps": 25, "total_steps": 876, "loss": 1.0732, "lr": 2.8409090909090916e-06, "epoch": 0.08549444286121402, "percentage": 2.85, "elapsed_time": "0:17:32", "remaining_time": "9:57:04"}
+{"current_steps": 26, "total_steps": 876, "loss": 1.0948, "lr": 2.954545454545455e-06, "epoch": 0.08891422057566258, "percentage": 2.97, "elapsed_time": "0:18:14", "remaining_time": "9:56:19"}
+{"current_steps": 27, "total_steps": 876, "loss": 0.9884, "lr": 3.0681818181818186e-06, "epoch": 0.09233399829011114, "percentage": 3.08, "elapsed_time": "0:18:54", "remaining_time": "9:54:47"}
+{"current_steps": 28, "total_steps": 876, "loss": 0.9821, "lr": 3.181818181818182e-06, "epoch": 0.0957537760045597, "percentage": 3.2, "elapsed_time": "0:19:38", "remaining_time": "9:54:41"}
+{"current_steps": 29, "total_steps": 876, "loss": 1.0122, "lr": 3.2954545454545456e-06, "epoch": 0.09917355371900827, "percentage": 3.31, "elapsed_time": "0:20:20", "remaining_time": "9:54:07"}
+{"current_steps": 30, "total_steps": 876, "loss": 1.0924, "lr": 3.409090909090909e-06, "epoch": 0.10259333143345682, "percentage": 3.42, "elapsed_time": "0:21:03", "remaining_time": "9:53:50"}
+{"current_steps": 31, "total_steps": 876, "loss": 1.0842, "lr": 3.522727272727273e-06, "epoch": 0.10601310914790539, "percentage": 3.54, "elapsed_time": "0:21:44", "remaining_time": "9:52:26"}
+{"current_steps": 32, "total_steps": 876, "loss": 0.9536, "lr": 3.6363636363636366e-06, "epoch": 0.10943288686235395, "percentage": 3.65, "elapsed_time": "0:22:24", "remaining_time": "9:51:06"}
+{"current_steps": 33, "total_steps": 876, "loss": 1.0273, "lr": 3.7500000000000005e-06, "epoch": 0.11285266457680251, "percentage": 3.77, "elapsed_time": "0:23:06", "remaining_time": "9:50:10"}
+{"current_steps": 34, "total_steps": 876, "loss": 0.9749, "lr": 3.863636363636364e-06, "epoch": 0.11627244229125107, "percentage": 3.88, "elapsed_time": "0:23:47", "remaining_time": "9:49:11"}
+{"current_steps": 35, "total_steps": 876, "loss": 0.9403, "lr": 3.9772727272727275e-06, "epoch": 0.11969222000569962, "percentage": 4.0, "elapsed_time": "0:24:29", "remaining_time": "9:48:23"}
+{"current_steps": 36, "total_steps": 876, "loss": 0.9462, "lr": 4.0909090909090915e-06, "epoch": 0.12311199772014819, "percentage": 4.11, "elapsed_time": "0:25:12", "remaining_time": "9:48:18"}
+{"current_steps": 37, "total_steps": 876, "loss": 0.9279, "lr": 4.204545454545455e-06, "epoch": 0.12653177543459676, "percentage": 4.22, "elapsed_time": "0:25:55", "remaining_time": "9:47:54"}
+{"current_steps": 38, "total_steps": 876, "loss": 0.9447, "lr": 4.3181818181818185e-06, "epoch": 0.1299515531490453, "percentage": 4.34, "elapsed_time": "0:26:37", "remaining_time": "9:46:59"}
+{"current_steps": 39, "total_steps": 876, "loss": 0.923, "lr": 4.4318181818181824e-06, "epoch": 0.13337133086349387, "percentage": 4.45, "elapsed_time": "0:27:19", "remaining_time": "9:46:27"}
+{"current_steps": 40, "total_steps": 876, "loss": 0.9413, "lr": 4.5454545454545455e-06, "epoch": 0.13679110857794244, "percentage": 4.57, "elapsed_time": "0:28:02", "remaining_time": "9:46:03"}
+{"current_steps": 41, "total_steps": 876, "loss": 0.8966, "lr": 4.6590909090909095e-06, "epoch": 0.140210886292391, "percentage": 4.68, "elapsed_time": "0:28:44", "remaining_time": "9:45:15"}
+{"current_steps": 42, "total_steps": 876, "loss": 0.9183, "lr": 4.772727272727273e-06, "epoch": 0.14363066400683955, "percentage": 4.79, "elapsed_time": "0:29:26", "remaining_time": "9:44:38"}
+{"current_steps": 43, "total_steps": 876, "loss": 0.9541, "lr": 4.8863636363636365e-06, "epoch": 0.14705044172128812, "percentage": 4.91, "elapsed_time": "0:30:08", "remaining_time": "9:43:49"}
+{"current_steps": 44, "total_steps": 876, "loss": 0.9049, "lr": 5e-06, "epoch": 0.15047021943573669, "percentage": 5.02, "elapsed_time": "0:30:52", "remaining_time": "9:43:55"}
+{"current_steps": 45, "total_steps": 876, "loss": 0.917, "lr": 5.113636363636364e-06, "epoch": 0.15388999715018523, "percentage": 5.14, "elapsed_time": "0:31:34", "remaining_time": "9:43:09"}
+{"current_steps": 46, "total_steps": 876, "loss": 0.9281, "lr": 5.2272727272727274e-06, "epoch": 0.1573097748646338, "percentage": 5.25, "elapsed_time": "0:32:17", "remaining_time": "9:42:37"}
+{"current_steps": 47, "total_steps": 876, "loss": 0.9118, "lr": 5.340909090909091e-06, "epoch": 0.16072955257908236, "percentage": 5.37, "elapsed_time": "0:32:58", "remaining_time": "9:41:30"}
+{"current_steps": 48, "total_steps": 876, "loss": 0.9378, "lr": 5.4545454545454545e-06, "epoch": 0.16414933029353093, "percentage": 5.48, "elapsed_time": "0:33:39", "remaining_time": "9:40:44"}
+{"current_steps": 49, "total_steps": 876, "loss": 0.9053, "lr": 5.568181818181818e-06, "epoch": 0.16756910800797947, "percentage": 5.59, "elapsed_time": "0:34:24", "remaining_time": "9:40:47"}
+{"current_steps": 50, "total_steps": 876, "loss": 0.8853, "lr": 5.681818181818183e-06, "epoch": 0.17098888572242804, "percentage": 5.71, "elapsed_time": "0:35:05", "remaining_time": "9:39:37"}
+{"current_steps": 51, "total_steps": 876, "loss": 0.9516, "lr": 5.795454545454546e-06, "epoch": 0.1744086634368766, "percentage": 5.82, "elapsed_time": "0:35:46", "remaining_time": "9:38:44"}
+{"current_steps": 52, "total_steps": 876, "loss": 0.8306, "lr": 5.90909090909091e-06, "epoch": 0.17782844115132515, "percentage": 5.94, "elapsed_time": "0:36:28", "remaining_time": "9:37:55"}
+{"current_steps": 53, "total_steps": 876, "loss": 0.8584, "lr": 6.022727272727273e-06, "epoch": 0.18124821886577372, "percentage": 6.05, "elapsed_time": "0:37:09", "remaining_time": "9:36:56"}
+{"current_steps": 54, "total_steps": 876, "loss": 0.9275, "lr": 6.136363636363637e-06, "epoch": 0.1846679965802223, "percentage": 6.16, "elapsed_time": "0:37:51", "remaining_time": "9:36:12"}
+{"current_steps": 55, "total_steps": 876, "loss": 0.9185, "lr": 6.25e-06, "epoch": 0.18808777429467086, "percentage": 6.28, "elapsed_time": "0:38:32", "remaining_time": "9:35:12"}
+{"current_steps": 56, "total_steps": 876, "loss": 0.9026, "lr": 6.363636363636364e-06, "epoch": 0.1915075520091194, "percentage": 6.39, "elapsed_time": "0:39:13", "remaining_time": "9:34:15"}
+{"current_steps": 57, "total_steps": 876, "loss": 0.891, "lr": 6.477272727272727e-06, "epoch": 0.19492732972356797, "percentage": 6.51, "elapsed_time": "0:39:53", "remaining_time": "9:33:13"}
+{"current_steps": 58, "total_steps": 876, "loss": 0.8441, "lr": 6.590909090909091e-06, "epoch": 0.19834710743801653, "percentage": 6.62, "elapsed_time": "0:40:34", "remaining_time": "9:32:17"}
+{"current_steps": 59, "total_steps": 876, "loss": 0.8349, "lr": 6.704545454545454e-06, "epoch": 0.2017668851524651, "percentage": 6.74, "elapsed_time": "0:41:16", "remaining_time": "9:31:37"}
+{"current_steps": 60, "total_steps": 876, "loss": 0.8339, "lr": 6.818181818181818e-06, "epoch": 0.20518666286691364, "percentage": 6.85, "elapsed_time": "0:41:57", "remaining_time": "9:30:44"}
+{"current_steps": 61, "total_steps": 876, "loss": 0.8374, "lr": 6.931818181818183e-06, "epoch": 0.2086064405813622, "percentage": 6.96, "elapsed_time": "0:42:40", "remaining_time": "9:30:14"}
+{"current_steps": 62, "total_steps": 876, "loss": 0.8501, "lr": 7.045454545454546e-06, "epoch": 0.21202621829581078, "percentage": 7.08, "elapsed_time": "0:43:23", "remaining_time": "9:29:41"}
+{"current_steps": 63, "total_steps": 876, "loss": 0.8596, "lr": 7.15909090909091e-06, "epoch": 0.21544599601025932, "percentage": 7.19, "elapsed_time": "0:44:05", "remaining_time": "9:29:04"}
+{"current_steps": 64, "total_steps": 876, "loss": 0.8495, "lr": 7.272727272727273e-06, "epoch": 0.2188657737247079, "percentage": 7.31, "elapsed_time": "0:44:46", "remaining_time": "9:28:06"}
+{"current_steps": 65, "total_steps": 876, "loss": 0.8598, "lr": 7.386363636363637e-06, "epoch": 0.22228555143915646, "percentage": 7.42, "elapsed_time": "0:45:27", "remaining_time": "9:27:10"}
+{"current_steps": 66, "total_steps": 876, "loss": 0.8404, "lr": 7.500000000000001e-06, "epoch": 0.22570532915360503, "percentage": 7.53, "elapsed_time": "0:46:09", "remaining_time": "9:26:29"}
+{"current_steps": 67, "total_steps": 876, "loss": 0.8957, "lr": 7.613636363636364e-06, "epoch": 0.22912510686805357, "percentage": 7.65, "elapsed_time": "0:46:50", "remaining_time": "9:25:34"}
+{"current_steps": 68, "total_steps": 876, "loss": 0.8282, "lr": 7.727272727272727e-06, "epoch": 0.23254488458250214, "percentage": 7.76, "elapsed_time": "0:47:31", "remaining_time": "9:24:47"}
+{"current_steps": 69, "total_steps": 876, "loss": 0.8148, "lr": 7.840909090909091e-06, "epoch": 0.2359646622969507, "percentage": 7.88, "elapsed_time": "0:48:13", "remaining_time": "9:24:06"}
+{"current_steps": 70, "total_steps": 876, "loss": 0.8177, "lr": 7.954545454545455e-06, "epoch": 0.23938444001139925, "percentage": 7.99, "elapsed_time": "0:48:56", "remaining_time": "9:23:28"}
+{"current_steps": 71, "total_steps": 876, "loss": 0.836, "lr": 8.068181818181819e-06, "epoch": 0.24280421772584782, "percentage": 8.11, "elapsed_time": "0:49:37", "remaining_time": "9:22:34"}
+{"current_steps": 72, "total_steps": 876, "loss": 0.8442, "lr": 8.181818181818183e-06, "epoch": 0.24622399544029638, "percentage": 8.22, "elapsed_time": "0:50:17", "remaining_time": "9:21:39"}
+{"current_steps": 73, "total_steps": 876, "loss": 0.9051, "lr": 8.295454545454547e-06, "epoch": 0.24964377315474495, "percentage": 8.33, "elapsed_time": "0:50:59", "remaining_time": "9:20:50"}
+{"current_steps": 74, "total_steps": 876, "loss": 0.8147, "lr": 8.40909090909091e-06, "epoch": 0.2530635508691935, "percentage": 8.45, "elapsed_time": "0:51:41", "remaining_time": "9:20:17"}
+{"current_steps": 75, "total_steps": 876, "loss": 0.8252, "lr": 8.522727272727273e-06, "epoch": 0.2564833285836421, "percentage": 8.56, "elapsed_time": "0:52:24", "remaining_time": "9:19:42"}
+{"current_steps": 76, "total_steps": 876, "loss": 0.7973, "lr": 8.636363636363637e-06, "epoch": 0.2599031062980906, "percentage": 8.68, "elapsed_time": "0:53:07", "remaining_time": "9:19:13"}
+{"current_steps": 77, "total_steps": 876, "loss": 0.8301, "lr": 8.750000000000001e-06, "epoch": 0.26332288401253917, "percentage": 8.79, "elapsed_time": "0:53:49", "remaining_time": "9:18:35"}
+{"current_steps": 78, "total_steps": 876, "loss": 0.8598, "lr": 8.863636363636365e-06, "epoch": 0.26674266172698774, "percentage": 8.9, "elapsed_time": "0:54:31", "remaining_time": "9:17:50"}
+{"current_steps": 79, "total_steps": 876, "loss": 0.8677, "lr": 8.977272727272727e-06, "epoch": 0.2701624394414363, "percentage": 9.02, "elapsed_time": "0:55:13", "remaining_time": "9:17:04"}
+{"current_steps": 80, "total_steps": 876, "loss": 0.8536, "lr": 9.090909090909091e-06, "epoch": 0.2735822171558849, "percentage": 9.13, "elapsed_time": "0:55:54", "remaining_time": "9:16:20"}
+{"current_steps": 81, "total_steps": 876, "loss": 0.8548, "lr": 9.204545454545455e-06, "epoch": 0.27700199487033345, "percentage": 9.25, "elapsed_time": "0:56:39", "remaining_time": "9:16:01"}
+{"current_steps": 82, "total_steps": 876, "loss": 0.8725, "lr": 9.318181818181819e-06, "epoch": 0.280421772584782, "percentage": 9.36, "elapsed_time": "0:57:21", "remaining_time": "9:15:20"}
+{"current_steps": 83, "total_steps": 876, "loss": 0.7992, "lr": 9.431818181818183e-06, "epoch": 0.2838415502992305, "percentage": 9.47, "elapsed_time": "0:58:04", "remaining_time": "9:14:52"}
+{"current_steps": 84, "total_steps": 876, "loss": 0.802, "lr": 9.545454545454547e-06, "epoch": 0.2872613280136791, "percentage": 9.59, "elapsed_time": "0:58:46", "remaining_time": "9:14:09"}
+{"current_steps": 85, "total_steps": 876, "loss": 0.7731, "lr": 9.65909090909091e-06, "epoch": 0.29068110572812766, "percentage": 9.7, "elapsed_time": "0:59:27", "remaining_time": "9:13:16"}
+{"current_steps": 86, "total_steps": 876, "loss": 0.8846, "lr": 9.772727272727273e-06, "epoch": 0.29410088344257623, "percentage": 9.82, "elapsed_time": "1:00:07", "remaining_time": "9:12:22"}
+{"current_steps": 87, "total_steps": 876, "loss": 0.8487, "lr": 9.886363636363637e-06, "epoch": 0.2975206611570248, "percentage": 9.93, "elapsed_time": "1:00:49", "remaining_time": "9:11:37"}
+{"current_steps": 88, "total_steps": 876, "loss": 0.7879, "lr": 1e-05, "epoch": 0.30094043887147337, "percentage": 10.05, "elapsed_time": "1:01:32", "remaining_time": "9:11:01"}
+{"current_steps": 89, "total_steps": 876, "loss": 0.8333, "lr": 9.9999602637624e-06, "epoch": 0.30436021658592194, "percentage": 10.16, "elapsed_time": "1:02:13", "remaining_time": "9:10:16"}
+{"current_steps": 90, "total_steps": 876, "loss": 0.8535, "lr": 9.999841055681184e-06, "epoch": 0.30777999430037045, "percentage": 10.27, "elapsed_time": "1:02:57", "remaining_time": "9:09:50"}
+{"current_steps": 91, "total_steps": 876, "loss": 0.8524, "lr": 9.999642377651106e-06, "epoch": 0.311199772014819, "percentage": 10.39, "elapsed_time": "1:03:38", "remaining_time": "9:09:00"}
+{"current_steps": 92, "total_steps": 876, "loss": 0.7886, "lr": 9.999364232830053e-06, "epoch": 0.3146195497292676, "percentage": 10.5, "elapsed_time": "1:04:19", "remaining_time": "9:08:05"}
+{"current_steps": 93, "total_steps": 876, "loss": 0.7732, "lr": 9.999006625638994e-06, "epoch": 0.31803932744371616, "percentage": 10.62, "elapsed_time": "1:05:00", "remaining_time": "9:07:20"}
+{"current_steps": 94, "total_steps": 876, "loss": 0.7559, "lr": 9.99856956176192e-06, "epoch": 0.3214591051581647, "percentage": 10.73, "elapsed_time": "1:05:42", "remaining_time": "9:06:42"}
+{"current_steps": 95, "total_steps": 876, "loss": 0.7851, "lr": 9.998053048145735e-06, "epoch": 0.3248788828726133, "percentage": 10.84, "elapsed_time": "1:06:24", "remaining_time": "9:06:00"}
+{"current_steps": 96, "total_steps": 876, "loss": 0.8172, "lr": 9.997457093000165e-06, "epoch": 0.32829866058706186, "percentage": 10.96, "elapsed_time": "1:07:06", "remaining_time": "9:05:15"}
+{"current_steps": 97, "total_steps": 876, "loss": 0.8562, "lr": 9.996781705797615e-06, "epoch": 0.3317184383015104, "percentage": 11.07, "elapsed_time": "1:07:48", "remaining_time": "9:04:36"}
+{"current_steps": 98, "total_steps": 876, "loss": 0.8589, "lr": 9.996026897273024e-06, "epoch": 0.33513821601595895, "percentage": 11.19, "elapsed_time": "1:08:33", "remaining_time": "9:04:16"}
+{"current_steps": 99, "total_steps": 876, "loss": 0.8289, "lr": 9.995192679423692e-06, "epoch": 0.3385579937304075, "percentage": 11.3, "elapsed_time": "1:09:17", "remaining_time": "9:03:47"}
+{"current_steps": 100, "total_steps": 876, "loss": 0.7737, "lr": 9.994279065509094e-06, "epoch": 0.3419777714448561, "percentage": 11.42, "elapsed_time": "1:10:00", "remaining_time": "9:03:13"}
+{"current_steps": 101, "total_steps": 876, "loss": 0.815, "lr": 9.993286070050656e-06, "epoch": 0.34539754915930465, "percentage": 11.53, "elapsed_time": "1:10:43", "remaining_time": "9:02:42"}
+{"current_steps": 102, "total_steps": 876, "loss": 0.7821, "lr": 9.992213708831542e-06, "epoch": 0.3488173268737532, "percentage": 11.64, "elapsed_time": "1:11:25", "remaining_time": "9:01:55"}
+{"current_steps": 103, "total_steps": 876, "loss": 0.8188, "lr": 9.991061998896395e-06, "epoch": 0.3522371045882018, "percentage": 11.76, "elapsed_time": "1:12:06", "remaining_time": "9:01:11"}
+{"current_steps": 104, "total_steps": 876, "loss": 0.8286, "lr": 9.989830958551058e-06, "epoch": 0.3556568823026503, "percentage": 11.87, "elapsed_time": "1:12:50", "remaining_time": "9:00:41"}
+{"current_steps": 105, "total_steps": 876, "loss": 0.8388, "lr": 9.988520607362297e-06, "epoch": 0.35907666001709887, "percentage": 11.99, "elapsed_time": "1:13:32", "remaining_time": "8:59:59"}
+{"current_steps": 106, "total_steps": 876, "loss": 0.831, "lr": 9.987130966157486e-06, "epoch": 0.36249643773154744, "percentage": 12.1, "elapsed_time": "1:14:15", "remaining_time": "8:59:22"}
+{"current_steps": 107, "total_steps": 876, "loss": 0.7902, "lr": 9.985662057024267e-06, "epoch": 0.365916215445996, "percentage": 12.21, "elapsed_time": "1:14:59", "remaining_time": "8:58:59"}
+{"current_steps": 108, "total_steps": 876, "loss": 0.8463, "lr": 9.984113903310206e-06, "epoch": 0.3693359931604446, "percentage": 12.33, "elapsed_time": "1:15:41", "remaining_time": "8:58:15"}
+{"current_steps": 109, "total_steps": 876, "loss": 0.7466, "lr": 9.98248652962243e-06, "epoch": 0.37275577087489314, "percentage": 12.44, "elapsed_time": "1:16:27", "remaining_time": "8:58:02"}
+{"current_steps": 110, "total_steps": 876, "loss": 0.8053, "lr": 9.98077996182722e-06, "epoch": 0.3761755485893417, "percentage": 12.56, "elapsed_time": "1:17:09", "remaining_time": "8:57:20"}
+{"current_steps": 111, "total_steps": 876, "loss": 0.7833, "lr": 9.97899422704961e-06, "epoch": 0.3795953263037902, "percentage": 12.67, "elapsed_time": "1:17:51", "remaining_time": "8:56:33"}
+{"current_steps": 112, "total_steps": 876, "loss": 0.797, "lr": 9.977129353672951e-06, "epoch": 0.3830151040182388, "percentage": 12.79, "elapsed_time": "1:18:32", "remaining_time": "8:55:42"}
+{"current_steps": 113, "total_steps": 876, "loss": 0.7529, "lr": 9.975185371338464e-06, "epoch": 0.38643488173268736, "percentage": 12.9, "elapsed_time": "1:19:13", "remaining_time": "8:54:55"}
+{"current_steps": 114, "total_steps": 876, "loss": 0.8124, "lr": 9.973162310944769e-06, "epoch": 0.38985465944713593, "percentage": 13.01, "elapsed_time": "1:19:55", "remaining_time": "8:54:12"}
+{"current_steps": 115, "total_steps": 876, "loss": 0.823, "lr": 9.971060204647384e-06, "epoch": 0.3932744371615845, "percentage": 13.13, "elapsed_time": "1:20:37", "remaining_time": "8:53:34"}
+{"current_steps": 116, "total_steps": 876, "loss": 0.8199, "lr": 9.968879085858234e-06, "epoch": 0.39669421487603307, "percentage": 13.24, "elapsed_time": "1:21:22", "remaining_time": "8:53:08"}
+{"current_steps": 117, "total_steps": 876, "loss": 0.7838, "lr": 9.966618989245096e-06, "epoch": 0.40011399259048164, "percentage": 13.36, "elapsed_time": "1:22:05", "remaining_time": "8:52:34"}
+{"current_steps": 118, "total_steps": 876, "loss": 0.8039, "lr": 9.964279950731066e-06, "epoch": 0.4035337703049302, "percentage": 13.47, "elapsed_time": "1:22:46", "remaining_time": "8:51:46"}
+{"current_steps": 119, "total_steps": 876, "loss": 0.8238, "lr": 9.961862007493981e-06, "epoch": 0.4069535480193787, "percentage": 13.58, "elapsed_time": "1:23:30", "remaining_time": "8:51:11"}
+{"current_steps": 120, "total_steps": 876, "loss": 0.7899, "lr": 9.959365197965824e-06, "epoch": 0.4103733257338273, "percentage": 13.7, "elapsed_time": "1:24:12", "remaining_time": "8:50:31"}
+{"current_steps": 121, "total_steps": 876, "loss": 0.7501, "lr": 9.956789561832127e-06, "epoch": 0.41379310344827586, "percentage": 13.81, "elapsed_time": "1:24:54", "remaining_time": "8:49:50"}
+{"current_steps": 122, "total_steps": 876, "loss": 0.7655, "lr": 9.954135140031322e-06, "epoch": 0.4172128811627244, "percentage": 13.93, "elapsed_time": "1:25:37", "remaining_time": "8:49:08"}
+{"current_steps": 123, "total_steps": 876, "loss": 0.7799, "lr": 9.951401974754103e-06, "epoch": 0.420632658877173, "percentage": 14.04, "elapsed_time": "1:26:18", "remaining_time": "8:48:23"}
+{"current_steps": 124, "total_steps": 876, "loss": 0.7772, "lr": 9.948590109442755e-06, "epoch": 0.42405243659162156, "percentage": 14.16, "elapsed_time": "1:26:59", "remaining_time": "8:47:34"}
+{"current_steps": 125, "total_steps": 876, "loss": 0.7817, "lr": 9.945699588790455e-06, "epoch": 0.42747221430607013, "percentage": 14.27, "elapsed_time": "1:27:41", "remaining_time": "8:46:50"}
+{"current_steps": 126, "total_steps": 876, "loss": 0.7938, "lr": 9.942730458740568e-06, "epoch": 0.43089199202051864, "percentage": 14.38, "elapsed_time": "1:28:23", "remaining_time": "8:46:07"}
+{"current_steps": 127, "total_steps": 876, "loss": 0.852, "lr": 9.93968276648592e-06, "epoch": 0.4343117697349672, "percentage": 14.5, "elapsed_time": "1:29:06", "remaining_time": "8:45:30"}
+{"current_steps": 128, "total_steps": 876, "loss": 0.7839, "lr": 9.936556560468037e-06, "epoch": 0.4377315474494158, "percentage": 14.61, "elapsed_time": "1:29:49", "remaining_time": "8:44:54"}
+{"current_steps": 129, "total_steps": 876, "loss": 0.8082, "lr": 9.933351890376389e-06, "epoch": 0.44115132516386435, "percentage": 14.73, "elapsed_time": "1:30:30", "remaining_time": "8:44:06"}
+{"current_steps": 130, "total_steps": 876, "loss": 0.8263, "lr": 9.930068807147585e-06, "epoch": 0.4445711028783129, "percentage": 14.84, "elapsed_time": "1:31:12", "remaining_time": "8:43:23"}
+{"current_steps": 131, "total_steps": 876, "loss": 0.7795, "lr": 9.926707362964579e-06, "epoch": 0.4479908805927615, "percentage": 14.95, "elapsed_time": "1:31:53", "remaining_time": "8:42:36"}
+{"current_steps": 132, "total_steps": 876, "loss": 0.8044, "lr": 9.923267611255824e-06, "epoch": 0.45141065830721006, "percentage": 15.07, "elapsed_time": "1:32:37", "remaining_time": "8:42:03"}
+{"current_steps": 133, "total_steps": 876, "loss": 0.7523, "lr": 9.91974960669444e-06, "epoch": 0.45483043602165857, "percentage": 15.18, "elapsed_time": "1:33:18", "remaining_time": "8:41:13"}
+{"current_steps": 134, "total_steps": 876, "loss": 0.8252, "lr": 9.916153405197333e-06, "epoch": 0.45825021373610714, "percentage": 15.3, "elapsed_time": "1:34:01", "remaining_time": "8:40:37"}
+{"current_steps": 135, "total_steps": 876, "loss": 0.7697, "lr": 9.912479063924309e-06, "epoch": 0.4616699914505557, "percentage": 15.41, "elapsed_time": "1:34:43", "remaining_time": "8:39:58"}
+{"current_steps": 136, "total_steps": 876, "loss": 0.7767, "lr": 9.908726641277167e-06, "epoch": 0.4650897691650043, "percentage": 15.53, "elapsed_time": "1:35:26", "remaining_time": "8:39:17"}
+{"current_steps": 137, "total_steps": 876, "loss": 0.7751, "lr": 9.904896196898771e-06, "epoch": 0.46850954687945284, "percentage": 15.64, "elapsed_time": "1:36:08", "remaining_time": "8:38:36"}
+{"current_steps": 138, "total_steps": 876, "loss": 0.7986, "lr": 9.9009877916721e-06, "epoch": 0.4719293245939014, "percentage": 15.75, "elapsed_time": "1:36:49", "remaining_time": "8:37:50"}
+{"current_steps": 139, "total_steps": 876, "loss": 0.7678, "lr": 9.89700148771928e-06, "epoch": 0.47534910230835, "percentage": 15.87, "elapsed_time": "1:37:32", "remaining_time": "8:37:10"}
+{"current_steps": 140, "total_steps": 876, "loss": 0.765, "lr": 9.8929373484006e-06, "epoch": 0.4787688800227985, "percentage": 15.98, "elapsed_time": "1:38:17", "remaining_time": "8:36:44"}
+{"current_steps": 141, "total_steps": 876, "loss": 0.772, "lr": 9.888795438313506e-06, "epoch": 0.48218865773724706, "percentage": 16.1, "elapsed_time": "1:39:00", "remaining_time": "8:36:05"}
+{"current_steps": 142, "total_steps": 876, "loss": 0.7549, "lr": 9.884575823291561e-06, "epoch": 0.48560843545169563, "percentage": 16.21, "elapsed_time": "1:39:42", "remaining_time": "8:35:23"}
+{"current_steps": 143, "total_steps": 876, "loss": 0.7787, "lr": 9.880278570403419e-06, "epoch": 0.4890282131661442, "percentage": 16.32, "elapsed_time": "1:40:26", "remaining_time": "8:34:50"}
+{"current_steps": 144, "total_steps": 876, "loss": 0.7838, "lr": 9.875903747951742e-06, "epoch": 0.49244799088059277, "percentage": 16.44, "elapsed_time": "1:41:08", "remaining_time": "8:34:06"}
+{"current_steps": 145, "total_steps": 876, "loss": 0.7346, "lr": 9.871451425472128e-06, "epoch": 0.49586776859504134, "percentage": 16.55, "elapsed_time": "1:41:51", "remaining_time": "8:33:29"}
+{"current_steps": 146, "total_steps": 876, "loss": 0.7602, "lr": 9.866921673731991e-06, "epoch": 0.4992875463094899, "percentage": 16.67, "elapsed_time": "1:42:32", "remaining_time": "8:32:44"}
+{"current_steps": 147, "total_steps": 876, "loss": 0.7167, "lr": 9.86231456472945e-06, "epoch": 0.5027073240239385, "percentage": 16.78, "elapsed_time": "1:43:14", "remaining_time": "8:31:58"}
+{"current_steps": 148, "total_steps": 876, "loss": 0.7581, "lr": 9.857630171692175e-06, "epoch": 0.506127101738387, "percentage": 16.89, "elapsed_time": "1:43:55", "remaining_time": "8:31:13"}
+{"current_steps": 149, "total_steps": 876, "loss": 0.7696, "lr": 9.852868569076225e-06, "epoch": 0.5095468794528356, "percentage": 17.01, "elapsed_time": "1:44:38", "remaining_time": "8:30:34"}
+{"current_steps": 150, "total_steps": 876, "loss": 0.8441, "lr": 9.848029832564875e-06, "epoch": 0.5129666571672842, "percentage": 17.12, "elapsed_time": "1:45:21", "remaining_time": "8:29:55"}
+{"current_steps": 151, "total_steps": 876, "loss": 0.7731, "lr": 9.843114039067395e-06, "epoch": 0.5163864348817326, "percentage": 17.24, "elapsed_time": "1:46:04", "remaining_time": "8:29:16"}
+{"current_steps": 152, "total_steps": 876, "loss": 0.7471, "lr": 9.83812126671784e-06, "epoch": 0.5198062125961812, "percentage": 17.35, "elapsed_time": "1:46:44", "remaining_time": "8:28:27"}
+{"current_steps": 153, "total_steps": 876, "loss": 0.7752, "lr": 9.833051594873805e-06, "epoch": 0.5232259903106298, "percentage": 17.47, "elapsed_time": "1:47:27", "remaining_time": "8:27:47"}
+{"current_steps": 154, "total_steps": 876, "loss": 0.7733, "lr": 9.827905104115167e-06, "epoch": 0.5266457680250783, "percentage": 17.58, "elapsed_time": "1:48:07", "remaining_time": "8:26:55"}
+{"current_steps": 155, "total_steps": 876, "loss": 0.7583, "lr": 9.822681876242797e-06, "epoch": 0.5300655457395269, "percentage": 17.69, "elapsed_time": "1:48:49", "remaining_time": "8:26:13"}
+{"current_steps": 156, "total_steps": 876, "loss": 0.741, "lr": 9.81738199427726e-06, "epoch": 0.5334853234539755, "percentage": 17.81, "elapsed_time": "1:49:30", "remaining_time": "8:25:26"}
+{"current_steps": 157, "total_steps": 876, "loss": 0.7838, "lr": 9.81200554245751e-06, "epoch": 0.536905101168424, "percentage": 17.92, "elapsed_time": "1:50:13", "remaining_time": "8:24:48"}
+{"current_steps": 158, "total_steps": 876, "loss": 0.7665, "lr": 9.80655260623953e-06, "epoch": 0.5403248788828726, "percentage": 18.04, "elapsed_time": "1:50:55", "remaining_time": "8:24:03"}
+{"current_steps": 159, "total_steps": 876, "loss": 0.7813, "lr": 9.80102327229499e-06, "epoch": 0.5437446565973212, "percentage": 18.15, "elapsed_time": "1:51:36", "remaining_time": "8:23:16"}
+{"current_steps": 160, "total_steps": 876, "loss": 0.7756, "lr": 9.795417628509857e-06, "epoch": 0.5471644343117698, "percentage": 18.26, "elapsed_time": "1:52:18", "remaining_time": "8:22:33"}
+{"current_steps": 161, "total_steps": 876, "loss": 0.7811, "lr": 9.789735763983013e-06, "epoch": 0.5505842120262183, "percentage": 18.38, "elapsed_time": "1:52:59", "remaining_time": "8:21:49"}
+{"current_steps": 162, "total_steps": 876, "loss": 0.7944, "lr": 9.783977769024821e-06, "epoch": 0.5540039897406669, "percentage": 18.49, "elapsed_time": "1:53:41", "remaining_time": "8:21:03"}
+{"current_steps": 163, "total_steps": 876, "loss": 0.7909, "lr": 9.778143735155709e-06, "epoch": 0.5574237674551155, "percentage": 18.61, "elapsed_time": "1:54:22", "remaining_time": "8:20:16"}
+{"current_steps": 164, "total_steps": 876, "loss": 0.8043, "lr": 9.772233755104695e-06, "epoch": 0.560843545169564, "percentage": 18.72, "elapsed_time": "1:55:04", "remaining_time": "8:19:34"}
+{"current_steps": 165, "total_steps": 876, "loss": 0.7798, "lr": 9.766247922807927e-06, "epoch": 0.5642633228840125, "percentage": 18.84, "elapsed_time": "1:55:47", "remaining_time": "8:18:58"}
+{"current_steps": 166, "total_steps": 876, "loss": 0.7296, "lr": 9.76018633340719e-06, "epoch": 0.567683100598461, "percentage": 18.95, "elapsed_time": "1:56:29", "remaining_time": "8:18:13"}
+{"current_steps": 167, "total_steps": 876, "loss": 0.7616, "lr": 9.754049083248384e-06, "epoch": 0.5711028783129096, "percentage": 19.06, "elapsed_time": "1:57:10", "remaining_time": "8:17:26"}
+{"current_steps": 168, "total_steps": 876, "loss": 0.7576, "lr": 9.747836269880005e-06, "epoch": 0.5745226560273582, "percentage": 19.18, "elapsed_time": "1:57:52", "remaining_time": "8:16:45"}
+{"current_steps": 169, "total_steps": 876, "loss": 0.8145, "lr": 9.74154799205158e-06, "epoch": 0.5779424337418068, "percentage": 19.29, "elapsed_time": "1:58:34", "remaining_time": "8:16:05"}
+{"current_steps": 170, "total_steps": 876, "loss": 0.7338, "lr": 9.73518434971211e-06, "epoch": 0.5813622114562553, "percentage": 19.41, "elapsed_time": "1:59:16", "remaining_time": "8:15:21"}
+{"current_steps": 171, "total_steps": 876, "loss": 0.7607, "lr": 9.728745444008481e-06, "epoch": 0.5847819891707039, "percentage": 19.52, "elapsed_time": "1:59:59", "remaining_time": "8:14:40"}
+{"current_steps": 172, "total_steps": 876, "loss": 0.7551, "lr": 9.722231377283841e-06, "epoch": 0.5882017668851525, "percentage": 19.63, "elapsed_time": "2:00:40", "remaining_time": "8:13:56"}
+{"current_steps": 173, "total_steps": 876, "loss": 0.7888, "lr": 9.715642253075998e-06, "epoch": 0.591621544599601, "percentage": 19.75, "elapsed_time": "2:01:24", "remaining_time": "8:13:20"}
+{"current_steps": 174, "total_steps": 876, "loss": 0.7613, "lr": 9.70897817611575e-06, "epoch": 0.5950413223140496, "percentage": 19.86, "elapsed_time": "2:02:07", "remaining_time": "8:12:41"}
+{"current_steps": 175, "total_steps": 876, "loss": 0.7366, "lr": 9.702239252325237e-06, "epoch": 0.5984611000284982, "percentage": 19.98, "elapsed_time": "2:02:50", "remaining_time": "8:12:04"}
+{"current_steps": 176, "total_steps": 876, "loss": 0.7535, "lr": 9.695425588816248e-06, "epoch": 0.6018808777429467, "percentage": 20.09, "elapsed_time": "2:03:32", "remaining_time": "8:11:21"}
+{"current_steps": 177, "total_steps": 876, "loss": 0.7959, "lr": 9.688537293888525e-06, "epoch": 0.6053006554573953, "percentage": 20.21, "elapsed_time": "2:04:12", "remaining_time": "8:10:32"}
+{"current_steps": 178, "total_steps": 876, "loss": 0.7965, "lr": 9.681574477028039e-06, "epoch": 0.6087204331718439, "percentage": 20.32, "elapsed_time": "2:04:55", "remaining_time": "8:09:54"}
+{"current_steps": 179, "total_steps": 876, "loss": 0.697, "lr": 9.674537248905244e-06, "epoch": 0.6121402108862924, "percentage": 20.43, "elapsed_time": "2:05:37", "remaining_time": "8:09:09"}
+{"current_steps": 180, "total_steps": 876, "loss": 0.7651, "lr": 9.667425721373333e-06, "epoch": 0.6155599886007409, "percentage": 20.55, "elapsed_time": "2:06:20", "remaining_time": "8:08:29"}
+{"current_steps": 181, "total_steps": 876, "loss": 0.7286, "lr": 9.66024000746644e-06, "epoch": 0.6189797663151895, "percentage": 20.66, "elapsed_time": "2:07:01", "remaining_time": "8:07:45"}
+{"current_steps": 182, "total_steps": 876, "loss": 0.798, "lr": 9.65298022139786e-06, "epoch": 0.622399544029638, "percentage": 20.78, "elapsed_time": "2:07:45", "remaining_time": "8:07:09"}
+{"current_steps": 183, "total_steps": 876, "loss": 0.8022, "lr": 9.64564647855823e-06, "epoch": 0.6258193217440866, "percentage": 20.89, "elapsed_time": "2:08:28", "remaining_time": "8:06:32"}
+{"current_steps": 184, "total_steps": 876, "loss": 0.7432, "lr": 9.638238895513687e-06, "epoch": 0.6292390994585352, "percentage": 21.0, "elapsed_time": "2:09:10", "remaining_time": "8:05:49"}
+{"current_steps": 185, "total_steps": 876, "loss": 0.7303, "lr": 9.630757590004023e-06, "epoch": 0.6326588771729837, "percentage": 21.12, "elapsed_time": "2:09:52", "remaining_time": "8:05:06"}
+{"current_steps": 186, "total_steps": 876, "loss": 0.7647, "lr": 9.623202680940811e-06, "epoch": 0.6360786548874323, "percentage": 21.23, "elapsed_time": "2:10:34", "remaining_time": "8:04:23"}
+{"current_steps": 187, "total_steps": 876, "loss": 0.7413, "lr": 9.615574288405515e-06, "epoch": 0.6394984326018809, "percentage": 21.35, "elapsed_time": "2:11:16", "remaining_time": "8:03:39"}
+{"current_steps": 188, "total_steps": 876, "loss": 0.7602, "lr": 9.607872533647584e-06, "epoch": 0.6429182103163295, "percentage": 21.46, "elapsed_time": "2:11:59", "remaining_time": "8:03:00"}
+{"current_steps": 189, "total_steps": 876, "loss": 0.7327, "lr": 9.60009753908252e-06, "epoch": 0.646337988030778, "percentage": 21.58, "elapsed_time": "2:12:41", "remaining_time": "8:02:18"}
+{"current_steps": 190, "total_steps": 876, "loss": 0.774, "lr": 9.592249428289935e-06, "epoch": 0.6497577657452266, "percentage": 21.69, "elapsed_time": "2:13:23", "remaining_time": "8:01:37"}
+{"current_steps": 191, "total_steps": 876, "loss": 0.7007, "lr": 9.584328326011588e-06, "epoch": 0.6531775434596752, "percentage": 21.8, "elapsed_time": "2:14:04", "remaining_time": "8:00:50"}
+{"current_steps": 192, "total_steps": 876, "loss": 0.7067, "lr": 9.5763343581494e-06, "epoch": 0.6565973211741237, "percentage": 21.92, "elapsed_time": "2:14:45", "remaining_time": "8:00:05"}
+{"current_steps": 193, "total_steps": 876, "loss": 0.768, "lr": 9.568267651763453e-06, "epoch": 0.6600170988885723, "percentage": 22.03, "elapsed_time": "2:15:27", "remaining_time": "7:59:23"}
+{"current_steps": 194, "total_steps": 876, "loss": 0.7779, "lr": 9.560128335069971e-06, "epoch": 0.6634368766030208, "percentage": 22.15, "elapsed_time": "2:16:10", "remaining_time": "7:58:44"}
+{"current_steps": 195, "total_steps": 876, "loss": 0.7922, "lr": 9.551916537439282e-06, "epoch": 0.6668566543174693, "percentage": 22.26, "elapsed_time": "2:16:52", "remaining_time": "7:58:00"}
+{"current_steps": 196, "total_steps": 876, "loss": 0.7655, "lr": 9.543632389393767e-06, "epoch": 0.6702764320319179, "percentage": 22.37, "elapsed_time": "2:17:33", "remaining_time": "7:57:14"}
+{"current_steps": 197, "total_steps": 876, "loss": 0.791, "lr": 9.535276022605772e-06, "epoch": 0.6736962097463665, "percentage": 22.49, "elapsed_time": "2:18:15", "remaining_time": "7:56:32"}
+{"current_steps": 198, "total_steps": 876, "loss": 0.7394, "lr": 9.526847569895529e-06, "epoch": 0.677115987460815, "percentage": 22.6, "elapsed_time": "2:18:57", "remaining_time": "7:55:48"}
+{"current_steps": 199, "total_steps": 876, "loss": 0.7739, "lr": 9.518347165229038e-06, "epoch": 0.6805357651752636, "percentage": 22.72, "elapsed_time": "2:19:39", "remaining_time": "7:55:06"}
+{"current_steps": 200, "total_steps": 876, "loss": 0.7519, "lr": 9.50977494371594e-06, "epoch": 0.6839555428897122, "percentage": 22.83, "elapsed_time": "2:20:20", "remaining_time": "7:54:21"}
+{"current_steps": 201, "total_steps": 876, "loss": 0.7499, "lr": 9.501131041607364e-06, "epoch": 0.6873753206041607, "percentage": 22.95, "elapsed_time": "2:21:01", "remaining_time": "7:53:36"}
+{"current_steps": 202, "total_steps": 876, "loss": 0.7249, "lr": 9.49241559629377e-06, "epoch": 0.6907950983186093, "percentage": 23.06, "elapsed_time": "2:21:43", "remaining_time": "7:52:53"}
+{"current_steps": 203, "total_steps": 876, "loss": 0.7775, "lr": 9.483628746302761e-06, "epoch": 0.6942148760330579, "percentage": 23.17, "elapsed_time": "2:22:25", "remaining_time": "7:52:11"}
+{"current_steps": 204, "total_steps": 876, "loss": 0.6973, "lr": 9.474770631296882e-06, "epoch": 0.6976346537475064, "percentage": 23.29, "elapsed_time": "2:23:06", "remaining_time": "7:51:26"}
+{"current_steps": 205, "total_steps": 876, "loss": 0.7024, "lr": 9.465841392071396e-06, "epoch": 0.701054431461955, "percentage": 23.4, "elapsed_time": "2:23:47", "remaining_time": "7:50:39"}
+{"current_steps": 206, "total_steps": 876, "loss": 0.7308, "lr": 9.456841170552054e-06, "epoch": 0.7044742091764036, "percentage": 23.52, "elapsed_time": "2:24:29", "remaining_time": "7:49:56"}
+{"current_steps": 207, "total_steps": 876, "loss": 0.6597, "lr": 9.447770109792828e-06, "epoch": 0.7078939868908521, "percentage": 23.63, "elapsed_time": "2:25:09", "remaining_time": "7:49:09"}
+{"current_steps": 208, "total_steps": 876, "loss": 0.7234, "lr": 9.438628353973654e-06, "epoch": 0.7113137646053006, "percentage": 23.74, "elapsed_time": "2:25:53", "remaining_time": "7:48:32"}
+{"current_steps": 209, "total_steps": 876, "loss": 0.7581, "lr": 9.42941604839812e-06, "epoch": 0.7147335423197492, "percentage": 23.86, "elapsed_time": "2:26:36", "remaining_time": "7:47:53"}
+{"current_steps": 210, "total_steps": 876, "loss": 0.7862, "lr": 9.420133339491171e-06, "epoch": 0.7181533200341977, "percentage": 23.97, "elapsed_time": "2:27:19", "remaining_time": "7:47:12"}
+{"current_steps": 211, "total_steps": 876, "loss": 0.7187, "lr": 9.410780374796782e-06, "epoch": 0.7215730977486463, "percentage": 24.09, "elapsed_time": "2:28:01", "remaining_time": "7:46:30"}
+{"current_steps": 212, "total_steps": 876, "loss": 0.7877, "lr": 9.4013573029756e-06, "epoch": 0.7249928754630949, "percentage": 24.2, "elapsed_time": "2:28:44", "remaining_time": "7:45:52"}
+{"current_steps": 213, "total_steps": 876, "loss": 0.7158, "lr": 9.391864273802596e-06, "epoch": 0.7284126531775434, "percentage": 24.32, "elapsed_time": "2:29:26", "remaining_time": "7:45:09"}
+{"current_steps": 214, "total_steps": 876, "loss": 0.7118, "lr": 9.382301438164673e-06, "epoch": 0.731832430891992, "percentage": 24.43, "elapsed_time": "2:30:09", "remaining_time": "7:44:30"}
+{"current_steps": 215, "total_steps": 876, "loss": 0.7764, "lr": 9.372668948058276e-06, "epoch": 0.7352522086064406, "percentage": 24.54, "elapsed_time": "2:30:50", "remaining_time": "7:43:46"}
+{"current_steps": 216, "total_steps": 876, "loss": 0.7469, "lr": 9.36296695658697e-06, "epoch": 0.7386719863208892, "percentage": 24.66, "elapsed_time": "2:31:34", "remaining_time": "7:43:09"}
+{"current_steps": 217, "total_steps": 876, "loss": 0.7397, "lr": 9.353195617959011e-06, "epoch": 0.7420917640353377, "percentage": 24.77, "elapsed_time": "2:32:15", "remaining_time": "7:42:24"}
+{"current_steps": 218, "total_steps": 876, "loss": 0.7018, "lr": 9.343355087484893e-06, "epoch": 0.7455115417497863, "percentage": 24.89, "elapsed_time": "2:32:57", "remaining_time": "7:41:41"}
+{"current_steps": 219, "total_steps": 876, "loss": 0.6836, "lr": 9.333445521574878e-06, "epoch": 0.7489313194642349, "percentage": 25.0, "elapsed_time": "2:33:38", "remaining_time": "7:40:55"}
+{"current_steps": 220, "total_steps": 876, "loss": 0.7351, "lr": 9.323467077736513e-06, "epoch": 0.7523510971786834, "percentage": 25.11, "elapsed_time": "2:34:21", "remaining_time": "7:40:15"}
+{"current_steps": 221, "total_steps": 876, "loss": 0.7569, "lr": 9.313419914572122e-06, "epoch": 0.755770874893132, "percentage": 25.23, "elapsed_time": "2:35:04", "remaining_time": "7:39:35"}
+{"current_steps": 222, "total_steps": 876, "loss": 0.7435, "lr": 9.303304191776291e-06, "epoch": 0.7591906526075805, "percentage": 25.34, "elapsed_time": "2:35:45", "remaining_time": "7:38:52"}
+{"current_steps": 223, "total_steps": 876, "loss": 0.7647, "lr": 9.293120070133327e-06, "epoch": 0.762610430322029, "percentage": 25.46, "elapsed_time": "2:36:28", "remaining_time": "7:38:11"}
+{"current_steps": 224, "total_steps": 876, "loss": 0.7208, "lr": 9.282867711514703e-06, "epoch": 0.7660302080364776, "percentage": 25.57, "elapsed_time": "2:37:09", "remaining_time": "7:37:26"}
+{"current_steps": 225, "total_steps": 876, "loss": 0.7395, "lr": 9.272547278876475e-06, "epoch": 0.7694499857509262, "percentage": 25.68, "elapsed_time": "2:37:53", "remaining_time": "7:36:50"}
+{"current_steps": 226, "total_steps": 876, "loss": 0.7955, "lr": 9.262158936256717e-06, "epoch": 0.7728697634653747, "percentage": 25.8, "elapsed_time": "2:38:37", "remaining_time": "7:36:12"}
+{"current_steps": 227, "total_steps": 876, "loss": 0.7159, "lr": 9.251702848772884e-06, "epoch": 0.7762895411798233, "percentage": 25.91, "elapsed_time": "2:39:20", "remaining_time": "7:35:33"}
+{"current_steps": 228, "total_steps": 876, "loss": 0.7004, "lr": 9.241179182619207e-06, "epoch": 0.7797093188942719, "percentage": 26.03, "elapsed_time": "2:40:00", "remaining_time": "7:34:46"}
+{"current_steps": 229, "total_steps": 876, "loss": 0.7216, "lr": 9.230588105064047e-06, "epoch": 0.7831290966087204, "percentage": 26.14, "elapsed_time": "2:40:42", "remaining_time": "7:34:02"}
+{"current_steps": 230, "total_steps": 876, "loss": 0.7555, "lr": 9.219929784447232e-06, "epoch": 0.786548874323169, "percentage": 26.26, "elapsed_time": "2:41:23", "remaining_time": "7:33:19"}
+{"current_steps": 231, "total_steps": 876, "loss": 0.7286, "lr": 9.209204390177388e-06, "epoch": 0.7899686520376176, "percentage": 26.37, "elapsed_time": "2:42:04", "remaining_time": "7:32:34"}
+{"current_steps": 232, "total_steps": 876, "loss": 0.7259, "lr": 9.19841209272924e-06, "epoch": 0.7933884297520661, "percentage": 26.48, "elapsed_time": "2:42:48", "remaining_time": "7:31:56"}
+{"current_steps": 233, "total_steps": 876, "loss": 0.7682, "lr": 9.187553063640905e-06, "epoch": 0.7968082074665147, "percentage": 26.6, "elapsed_time": "2:43:30", "remaining_time": "7:31:12"}
+{"current_steps": 234, "total_steps": 876, "loss": 0.7338, "lr": 9.176627475511171e-06, "epoch": 0.8002279851809633, "percentage": 26.71, "elapsed_time": "2:44:12", "remaining_time": "7:30:30"}
+{"current_steps": 235, "total_steps": 876, "loss": 0.6741, "lr": 9.16563550199674e-06, "epoch": 0.8036477628954118, "percentage": 26.83, "elapsed_time": "2:44:54", "remaining_time": "7:29:47"}
+{"current_steps": 236, "total_steps": 876, "loss": 0.7409, "lr": 9.154577317809483e-06, "epoch": 0.8070675406098604, "percentage": 26.94, "elapsed_time": "2:45:36", "remaining_time": "7:29:05"}
+{"current_steps": 237, "total_steps": 876, "loss": 0.728, "lr": 9.143453098713655e-06, "epoch": 0.8104873183243089, "percentage": 27.05, "elapsed_time": "2:46:17", "remaining_time": "7:28:20"}
+{"current_steps": 238, "total_steps": 876, "loss": 0.733, "lr": 9.132263021523096e-06, "epoch": 0.8139070960387574, "percentage": 27.17, "elapsed_time": "2:46:57", "remaining_time": "7:27:34"}
+{"current_steps": 239, "total_steps": 876, "loss": 0.7647, "lr": 9.121007264098438e-06, "epoch": 0.817326873753206, "percentage": 27.28, "elapsed_time": "2:47:39", "remaining_time": "7:26:51"}
+{"current_steps": 240, "total_steps": 876, "loss": 0.7708, "lr": 9.109686005344258e-06, "epoch": 0.8207466514676546, "percentage": 27.4, "elapsed_time": "2:48:22", "remaining_time": "7:26:11"}
+{"current_steps": 241, "total_steps": 876, "loss": 0.7486, "lr": 9.098299425206248e-06, "epoch": 0.8241664291821031, "percentage": 27.51, "elapsed_time": "2:49:06", "remaining_time": "7:25:34"}
+{"current_steps": 242, "total_steps": 876, "loss": 0.7467, "lr": 9.086847704668352e-06, "epoch": 0.8275862068965517, "percentage": 27.63, "elapsed_time": "2:49:49", "remaining_time": "7:24:53"}
+{"current_steps": 243, "total_steps": 876, "loss": 0.6948, "lr": 9.07533102574988e-06, "epoch": 0.8310059846110003, "percentage": 27.74, "elapsed_time": "2:50:30", "remaining_time": "7:24:10"}
+{"current_steps": 244, "total_steps": 876, "loss": 0.7371, "lr": 9.063749571502633e-06, "epoch": 0.8344257623254488, "percentage": 27.85, "elapsed_time": "2:51:13", "remaining_time": "7:23:29"}
+{"current_steps": 245, "total_steps": 876, "loss": 0.7565, "lr": 9.052103526007976e-06, "epoch": 0.8378455400398974, "percentage": 27.97, "elapsed_time": "2:51:56", "remaining_time": "7:22:49"}
+{"current_steps": 246, "total_steps": 876, "loss": 0.7766, "lr": 9.040393074373921e-06, "epoch": 0.841265317754346, "percentage": 28.08, "elapsed_time": "2:52:39", "remaining_time": "7:22:09"}
+{"current_steps": 247, "total_steps": 876, "loss": 0.7276, "lr": 9.028618402732185e-06, "epoch": 0.8446850954687946, "percentage": 28.2, "elapsed_time": "2:53:19", "remaining_time": "7:21:23"}
+{"current_steps": 248, "total_steps": 876, "loss": 0.7241, "lr": 9.016779698235227e-06, "epoch": 0.8481048731832431, "percentage": 28.31, "elapsed_time": "2:54:01", "remaining_time": "7:20:40"}
+{"current_steps": 249, "total_steps": 876, "loss": 0.7894, "lr": 9.004877149053277e-06, "epoch": 0.8515246508976917, "percentage": 28.42, "elapsed_time": "2:54:43", "remaining_time": "7:19:57"}
+{"current_steps": 250, "total_steps": 876, "loss": 0.7455, "lr": 8.992910944371343e-06, "epoch": 0.8549444286121403, "percentage": 28.54, "elapsed_time": "2:55:25", "remaining_time": "7:19:15"}
+{"current_steps": 251, "total_steps": 876, "loss": 0.6952, "lr": 8.980881274386207e-06, "epoch": 0.8583642063265887, "percentage": 28.65, "elapsed_time": "2:56:07", "remaining_time": "7:18:32"}
+{"current_steps": 252, "total_steps": 876, "loss": 0.7474, "lr": 8.9687883303034e-06, "epoch": 0.8617839840410373, "percentage": 28.77, "elapsed_time": "2:56:48", "remaining_time": "7:17:49"}
+{"current_steps": 253, "total_steps": 876, "loss": 0.7323, "lr": 8.956632304334158e-06, "epoch": 0.8652037617554859, "percentage": 28.88, "elapsed_time": "2:57:29", "remaining_time": "7:17:04"}
+{"current_steps": 254, "total_steps": 876, "loss": 0.7479, "lr": 8.94441338969238e-06, "epoch": 0.8686235394699344, "percentage": 29.0, "elapsed_time": "2:58:11", "remaining_time": "7:16:22"}
+{"current_steps": 255, "total_steps": 876, "loss": 0.7248, "lr": 8.932131780591542e-06, "epoch": 0.872043317184383, "percentage": 29.11, "elapsed_time": "2:58:53", "remaining_time": "7:15:40"}
+{"current_steps": 256, "total_steps": 876, "loss": 0.7423, "lr": 8.919787672241619e-06, "epoch": 0.8754630948988316, "percentage": 29.22, "elapsed_time": "2:59:37", "remaining_time": "7:15:02"}
+{"current_steps": 257, "total_steps": 876, "loss": 0.7562, "lr": 8.90738126084598e-06, "epoch": 0.8788828726132801, "percentage": 29.34, "elapsed_time": "3:00:18", "remaining_time": "7:14:17"}
+{"current_steps": 258, "total_steps": 876, "loss": 0.7621, "lr": 8.894912743598269e-06, "epoch": 0.8823026503277287, "percentage": 29.45, "elapsed_time": "3:00:59", "remaining_time": "7:13:31"}
+{"current_steps": 259, "total_steps": 876, "loss": 0.69, "lr": 8.882382318679272e-06, "epoch": 0.8857224280421773, "percentage": 29.57, "elapsed_time": "3:01:42", "remaining_time": "7:12:52"}
+{"current_steps": 260, "total_steps": 876, "loss": 0.7642, "lr": 8.869790185253766e-06, "epoch": 0.8891422057566258, "percentage": 29.68, "elapsed_time": "3:02:24", "remaining_time": "7:12:10"}
+{"current_steps": 261, "total_steps": 876, "loss": 0.7308, "lr": 8.857136543467353e-06, "epoch": 0.8925619834710744, "percentage": 29.79, "elapsed_time": "3:03:07", "remaining_time": "7:11:30"}
+{"current_steps": 262, "total_steps": 876, "loss": 0.7139, "lr": 8.84442159444328e-06, "epoch": 0.895981761185523, "percentage": 29.91, "elapsed_time": "3:03:50", "remaining_time": "7:10:49"}
+{"current_steps": 263, "total_steps": 876, "loss": 0.7345, "lr": 8.831645540279239e-06, "epoch": 0.8994015388999715, "percentage": 30.02, "elapsed_time": "3:04:31", "remaining_time": "7:10:05"}
+{"current_steps": 264, "total_steps": 876, "loss": 0.7299, "lr": 8.818808584044163e-06, "epoch": 0.9028213166144201, "percentage": 30.14, "elapsed_time": "3:05:13", "remaining_time": "7:09:24"}
+{"current_steps": 265, "total_steps": 876, "loss": 0.753, "lr": 8.805910929774989e-06, "epoch": 0.9062410943288686, "percentage": 30.25, "elapsed_time": "3:05:57", "remaining_time": "7:08:45"}
+{"current_steps": 266, "total_steps": 876, "loss": 0.7449, "lr": 8.792952782473415e-06, "epoch": 0.9096608720433171, "percentage": 30.37, "elapsed_time": "3:06:38", "remaining_time": "7:08:01"}
+{"current_steps": 267, "total_steps": 876, "loss": 0.7272, "lr": 8.779934348102652e-06, "epoch": 0.9130806497577657, "percentage": 30.48, "elapsed_time": "3:07:20", "remaining_time": "7:07:18"}
+{"current_steps": 268, "total_steps": 876, "loss": 0.7399, "lr": 8.76685583358414e-06, "epoch": 0.9165004274722143, "percentage": 30.59, "elapsed_time": "3:08:02", "remaining_time": "7:06:36"}
+{"current_steps": 269, "total_steps": 876, "loss": 0.7404, "lr": 8.753717446794262e-06, "epoch": 0.9199202051866628, "percentage": 30.71, "elapsed_time": "3:08:43", "remaining_time": "7:05:51"}
+{"current_steps": 270, "total_steps": 876, "loss": 0.6951, "lr": 8.740519396561045e-06, "epoch": 0.9233399829011114, "percentage": 30.82, "elapsed_time": "3:09:24", "remaining_time": "7:05:07"}
+{"current_steps": 271, "total_steps": 876, "loss": 0.7018, "lr": 8.72726189266083e-06, "epoch": 0.92675976061556, "percentage": 30.94, "elapsed_time": "3:10:07", "remaining_time": "7:04:26"}
+{"current_steps": 272, "total_steps": 876, "loss": 0.7006, "lr": 8.713945145814948e-06, "epoch": 0.9301795383300085, "percentage": 31.05, "elapsed_time": "3:10:49", "remaining_time": "7:03:44"}
+{"current_steps": 273, "total_steps": 876, "loss": 0.6998, "lr": 8.700569367686365e-06, "epoch": 0.9335993160444571, "percentage": 31.16, "elapsed_time": "3:11:30", "remaining_time": "7:02:59"}
+{"current_steps": 274, "total_steps": 876, "loss": 0.7088, "lr": 8.68713477087632e-06, "epoch": 0.9370190937589057, "percentage": 31.28, "elapsed_time": "3:12:15", "remaining_time": "7:02:23"}
+{"current_steps": 275, "total_steps": 876, "loss": 0.6882, "lr": 8.673641568920944e-06, "epoch": 0.9404388714733543, "percentage": 31.39, "elapsed_time": "3:12:57", "remaining_time": "7:01:42"}
+{"current_steps": 276, "total_steps": 876, "loss": 0.7262, "lr": 8.660089976287875e-06, "epoch": 0.9438586491878028, "percentage": 31.51, "elapsed_time": "3:13:38", "remaining_time": "7:00:58"}
+{"current_steps": 277, "total_steps": 876, "loss": 0.7704, "lr": 8.646480208372829e-06, "epoch": 0.9472784269022514, "percentage": 31.62, "elapsed_time": "3:14:21", "remaining_time": "7:00:18"}
+{"current_steps": 278, "total_steps": 876, "loss": 0.7287, "lr": 8.632812481496195e-06, "epoch": 0.9506982046167, "percentage": 31.74, "elapsed_time": "3:15:04", "remaining_time": "6:59:36"}
+{"current_steps": 279, "total_steps": 876, "loss": 0.7045, "lr": 8.619087012899591e-06, "epoch": 0.9541179823311485, "percentage": 31.85, "elapsed_time": "3:15:45", "remaining_time": "6:58:53"}
+{"current_steps": 280, "total_steps": 876, "loss": 0.7022, "lr": 8.60530402074241e-06, "epoch": 0.957537760045597, "percentage": 31.96, "elapsed_time": "3:16:27", "remaining_time": "6:58:10"}
+{"current_steps": 281, "total_steps": 876, "loss": 0.752, "lr": 8.591463724098352e-06, "epoch": 0.9609575377600456, "percentage": 32.08, "elapsed_time": "3:17:10", "remaining_time": "6:57:29"}
+{"current_steps": 282, "total_steps": 876, "loss": 0.786, "lr": 8.577566342951944e-06, "epoch": 0.9643773154744941, "percentage": 32.19, "elapsed_time": "3:17:52", "remaining_time": "6:56:48"}
+{"current_steps": 283, "total_steps": 876, "loss": 0.7117, "lr": 8.56361209819504e-06, "epoch": 0.9677970931889427, "percentage": 32.31, "elapsed_time": "3:18:34", "remaining_time": "6:56:04"}
+{"current_steps": 284, "total_steps": 876, "loss": 0.6863, "lr": 8.549601211623316e-06, "epoch": 0.9712168709033913, "percentage": 32.42, "elapsed_time": "3:19:15", "remaining_time": "6:55:22"}
+{"current_steps": 285, "total_steps": 876, "loss": 0.7148, "lr": 8.535533905932739e-06, "epoch": 0.9746366486178398, "percentage": 32.53, "elapsed_time": "3:19:56", "remaining_time": "6:54:37"}
+{"current_steps": 286, "total_steps": 876, "loss": 0.696, "lr": 8.521410404716029e-06, "epoch": 0.9780564263322884, "percentage": 32.65, "elapsed_time": "3:20:39", "remaining_time": "6:53:57"}
+{"current_steps": 287, "total_steps": 876, "loss": 0.7036, "lr": 8.507230932459102e-06, "epoch": 0.981476204046737, "percentage": 32.76, "elapsed_time": "3:21:20", "remaining_time": "6:53:12"}
+{"current_steps": 288, "total_steps": 876, "loss": 0.7001, "lr": 8.492995714537519e-06, "epoch": 0.9848959817611855, "percentage": 32.88, "elapsed_time": "3:22:01", "remaining_time": "6:52:27"}
+{"current_steps": 289, "total_steps": 876, "loss": 0.7062, "lr": 8.478704977212873e-06, "epoch": 0.9883157594756341, "percentage": 32.99, "elapsed_time": "3:22:42", "remaining_time": "6:51:44"}
+{"current_steps": 290, "total_steps": 876, "loss": 0.7013, "lr": 8.46435894762922e-06, "epoch": 0.9917355371900827, "percentage": 33.11, "elapsed_time": "3:23:25", "remaining_time": "6:51:03"}
+{"current_steps": 291, "total_steps": 876, "loss": 0.7637, "lr": 8.449957853809453e-06, "epoch": 0.9951553149045312, "percentage": 33.22, "elapsed_time": "3:24:07", "remaining_time": "6:50:20"}
+{"current_steps": 292, "total_steps": 876, "loss": 0.7363, "lr": 8.43550192465169e-06, "epoch": 0.9985750926189798, "percentage": 33.33, "elapsed_time": "3:24:48", "remaining_time": "6:49:36"}
+{"current_steps": 293, "total_steps": 876, "loss": 1.1154, "lr": 8.420991389925627e-06, "epoch": 1.0019948703334283, "percentage": 33.45, "elapsed_time": "3:25:30", "remaining_time": "6:48:54"}
+{"current_steps": 294, "total_steps": 876, "loss": 0.5776, "lr": 8.406426480268881e-06, "epoch": 1.005414648047877, "percentage": 33.56, "elapsed_time": "3:26:11", "remaining_time": "6:48:10"}
+{"current_steps": 295, "total_steps": 876, "loss": 0.5697, "lr": 8.39180742718334e-06, "epoch": 1.0088344257623254, "percentage": 33.68, "elapsed_time": "3:26:53", "remaining_time": "6:47:28"}
+{"current_steps": 296, "total_steps": 876, "loss": 0.5745, "lr": 8.377134463031468e-06, "epoch": 1.012254203476774, "percentage": 33.79, "elapsed_time": "3:27:37", "remaining_time": "6:46:49"}
+{"current_steps": 297, "total_steps": 876, "loss": 0.5214, "lr": 8.362407821032623e-06, "epoch": 1.0156739811912225, "percentage": 33.9, "elapsed_time": "3:28:20", "remaining_time": "6:46:08"}
+{"current_steps": 298, "total_steps": 876, "loss": 0.5839, "lr": 8.347627735259344e-06, "epoch": 1.0190937589056712, "percentage": 34.02, "elapsed_time": "3:29:05", "remaining_time": "6:45:32"}
+{"current_steps": 299, "total_steps": 876, "loss": 0.5362, "lr": 8.33279444063363e-06, "epoch": 1.0225135366201197, "percentage": 34.13, "elapsed_time": "3:29:46", "remaining_time": "6:44:49"}
+{"current_steps": 300, "total_steps": 876, "loss": 0.5884, "lr": 8.317908172923207e-06, "epoch": 1.0259333143345684, "percentage": 34.25, "elapsed_time": "3:30:29", "remaining_time": "6:44:07"}
+{"current_steps": 301, "total_steps": 876, "loss": 0.5414, "lr": 8.302969168737784e-06, "epoch": 1.0293530920490168, "percentage": 34.36, "elapsed_time": "3:31:11", "remaining_time": "6:43:25"}
+{"current_steps": 302, "total_steps": 876, "loss": 0.5593, "lr": 8.287977665525292e-06, "epoch": 1.0327728697634653, "percentage": 34.47, "elapsed_time": "3:31:52", "remaining_time": "6:42:41"}
+{"current_steps": 303, "total_steps": 876, "loss": 0.5564, "lr": 8.272933901568102e-06, "epoch": 1.036192647477914, "percentage": 34.59, "elapsed_time": "3:32:33", "remaining_time": "6:41:57"}
+{"current_steps": 304, "total_steps": 876, "loss": 0.5832, "lr": 8.257838115979244e-06, "epoch": 1.0396124251923624, "percentage": 34.7, "elapsed_time": "3:33:16", "remaining_time": "6:41:18"}
+{"current_steps": 305, "total_steps": 876, "loss": 0.5374, "lr": 8.242690548698611e-06, "epoch": 1.043032202906811, "percentage": 34.82, "elapsed_time": "3:33:57", "remaining_time": "6:40:33"}
+{"current_steps": 306, "total_steps": 876, "loss": 0.5616, "lr": 8.227491440489134e-06, "epoch": 1.0464519806212595, "percentage": 34.93, "elapsed_time": "3:34:38", "remaining_time": "6:39:48"}
+{"current_steps": 307, "total_steps": 876, "loss": 0.5172, "lr": 8.212241032932962e-06, "epoch": 1.0498717583357082, "percentage": 35.05, "elapsed_time": "3:35:19", "remaining_time": "6:39:05"}
+{"current_steps": 308, "total_steps": 876, "loss": 0.5786, "lr": 8.196939568427624e-06, "epoch": 1.0532915360501567, "percentage": 35.16, "elapsed_time": "3:36:02", "remaining_time": "6:38:24"}
+{"current_steps": 309, "total_steps": 876, "loss": 0.5259, "lr": 8.18158729018217e-06, "epoch": 1.0567113137646054, "percentage": 35.27, "elapsed_time": "3:36:43", "remaining_time": "6:37:41"}
+{"current_steps": 310, "total_steps": 876, "loss": 0.5519, "lr": 8.166184442213314e-06, "epoch": 1.0601310914790538, "percentage": 35.39, "elapsed_time": "3:37:26", "remaining_time": "6:36:59"}
+{"current_steps": 311, "total_steps": 876, "loss": 0.5873, "lr": 8.150731269341544e-06, "epoch": 1.0635508691935025, "percentage": 35.5, "elapsed_time": "3:38:07", "remaining_time": "6:36:15"}
+{"current_steps": 312, "total_steps": 876, "loss": 0.521, "lr": 8.135228017187238e-06, "epoch": 1.066970646907951, "percentage": 35.62, "elapsed_time": "3:38:48", "remaining_time": "6:35:32"}
+{"current_steps": 313, "total_steps": 876, "loss": 0.5685, "lr": 8.119674932166763e-06, "epoch": 1.0703904246223996, "percentage": 35.73, "elapsed_time": "3:39:31", "remaining_time": "6:34:51"}
+{"current_steps": 314, "total_steps": 876, "loss": 0.5206, "lr": 8.10407226148855e-06, "epoch": 1.073810202336848, "percentage": 35.84, "elapsed_time": "3:40:12", "remaining_time": "6:34:08"}
+{"current_steps": 315, "total_steps": 876, "loss": 0.5661, "lr": 8.088420253149173e-06, "epoch": 1.0772299800512966, "percentage": 35.96, "elapsed_time": "3:40:53", "remaining_time": "6:33:24"}
+{"current_steps": 316, "total_steps": 876, "loss": 0.5497, "lr": 8.0727191559294e-06, "epoch": 1.0806497577657452, "percentage": 36.07, "elapsed_time": "3:41:34", "remaining_time": "6:32:39"}
+{"current_steps": 317, "total_steps": 876, "loss": 0.589, "lr": 8.05696921939024e-06, "epoch": 1.0840695354801937, "percentage": 36.19, "elapsed_time": "3:42:16", "remaining_time": "6:31:58"}
+{"current_steps": 318, "total_steps": 876, "loss": 0.5708, "lr": 8.041170693868985e-06, "epoch": 1.0874893131946424, "percentage": 36.3, "elapsed_time": "3:42:58", "remaining_time": "6:31:14"}
+{"current_steps": 319, "total_steps": 876, "loss": 0.6075, "lr": 8.02532383047522e-06, "epoch": 1.0909090909090908, "percentage": 36.42, "elapsed_time": "3:43:40", "remaining_time": "6:30:33"}
+{"current_steps": 320, "total_steps": 876, "loss": 0.5606, "lr": 8.009428881086836e-06, "epoch": 1.0943288686235395, "percentage": 36.53, "elapsed_time": "3:44:23", "remaining_time": "6:29:53"}
+{"current_steps": 321, "total_steps": 876, "loss": 0.5328, "lr": 7.993486098346026e-06, "epoch": 1.097748646337988, "percentage": 36.64, "elapsed_time": "3:45:06", "remaining_time": "6:29:12"}
+{"current_steps": 322, "total_steps": 876, "loss": 0.5653, "lr": 7.977495735655271e-06, "epoch": 1.1011684240524366, "percentage": 36.76, "elapsed_time": "3:45:48", "remaining_time": "6:28:30"}
+{"current_steps": 323, "total_steps": 876, "loss": 0.521, "lr": 7.961458047173316e-06, "epoch": 1.104588201766885, "percentage": 36.87, "elapsed_time": "3:46:32", "remaining_time": "6:27:50"}
+{"current_steps": 324, "total_steps": 876, "loss": 0.5616, "lr": 7.945373287811116e-06, "epoch": 1.1080079794813338, "percentage": 36.99, "elapsed_time": "3:47:15", "remaining_time": "6:27:10"}
+{"current_steps": 325, "total_steps": 876, "loss": 0.5657, "lr": 7.9292417132278e-06, "epoch": 1.1114277571957822, "percentage": 37.1, "elapsed_time": "3:47:57", "remaining_time": "6:26:27"}
+{"current_steps": 326, "total_steps": 876, "loss": 0.5345, "lr": 7.913063579826601e-06, "epoch": 1.114847534910231, "percentage": 37.21, "elapsed_time": "3:48:38", "remaining_time": "6:25:44"}
+{"current_steps": 327, "total_steps": 876, "loss": 0.5755, "lr": 7.896839144750782e-06, "epoch": 1.1182673126246794, "percentage": 37.33, "elapsed_time": "3:49:20", "remaining_time": "6:25:02"}
+{"current_steps": 328, "total_steps": 876, "loss": 0.5434, "lr": 7.880568665879542e-06, "epoch": 1.121687090339128, "percentage": 37.44, "elapsed_time": "3:50:02", "remaining_time": "6:24:20"}
+{"current_steps": 329, "total_steps": 876, "loss": 0.5616, "lr": 7.864252401823928e-06, "epoch": 1.1251068680535765, "percentage": 37.56, "elapsed_time": "3:50:44", "remaining_time": "6:23:37"}
+{"current_steps": 330, "total_steps": 876, "loss": 0.5442, "lr": 7.847890611922721e-06, "epoch": 1.1285266457680252, "percentage": 37.67, "elapsed_time": "3:51:25", "remaining_time": "6:22:54"}
+{"current_steps": 331, "total_steps": 876, "loss": 0.5212, "lr": 7.831483556238306e-06, "epoch": 1.1319464234824737, "percentage": 37.79, "elapsed_time": "3:52:07", "remaining_time": "6:22:11"}
+{"current_steps": 332, "total_steps": 876, "loss": 0.5549, "lr": 7.81503149555255e-06, "epoch": 1.135366201196922, "percentage": 37.9, "elapsed_time": "3:52:50", "remaining_time": "6:21:31"}
+{"current_steps": 333, "total_steps": 876, "loss": 0.5945, "lr": 7.798534691362647e-06, "epoch": 1.1387859789113708, "percentage": 38.01, "elapsed_time": "3:53:33", "remaining_time": "6:20:50"}
+{"current_steps": 334, "total_steps": 876, "loss": 0.5727, "lr": 7.781993405876974e-06, "epoch": 1.1422057566258192, "percentage": 38.13, "elapsed_time": "3:54:15", "remaining_time": "6:20:07"}
+{"current_steps": 335, "total_steps": 876, "loss": 0.5512, "lr": 7.76540790201091e-06, "epoch": 1.145625534340268, "percentage": 38.24, "elapsed_time": "3:54:57", "remaining_time": "6:19:27"}
+{"current_steps": 336, "total_steps": 876, "loss": 0.5197, "lr": 7.748778443382658e-06, "epoch": 1.1490453120547164, "percentage": 38.36, "elapsed_time": "3:55:40", "remaining_time": "6:18:45"}
+{"current_steps": 337, "total_steps": 876, "loss": 0.579, "lr": 7.732105294309074e-06, "epoch": 1.152465089769165, "percentage": 38.47, "elapsed_time": "3:56:22", "remaining_time": "6:18:03"}
+{"current_steps": 338, "total_steps": 876, "loss": 0.5646, "lr": 7.715388719801437e-06, "epoch": 1.1558848674836135, "percentage": 38.58, "elapsed_time": "3:57:03", "remaining_time": "6:17:20"}
+{"current_steps": 339, "total_steps": 876, "loss": 0.5827, "lr": 7.698628985561263e-06, "epoch": 1.1593046451980622, "percentage": 38.7, "elapsed_time": "3:57:47", "remaining_time": "6:16:40"}
+{"current_steps": 340, "total_steps": 876, "loss": 0.5742, "lr": 7.68182635797606e-06, "epoch": 1.1627244229125107, "percentage": 38.81, "elapsed_time": "3:58:29", "remaining_time": "6:15:58"}
+{"current_steps": 341, "total_steps": 876, "loss": 0.5616, "lr": 7.664981104115115e-06, "epoch": 1.1661442006269593, "percentage": 38.93, "elapsed_time": "3:59:10", "remaining_time": "6:15:14"}
+{"current_steps": 342, "total_steps": 876, "loss": 0.5792, "lr": 7.648093491725224e-06, "epoch": 1.1695639783414078, "percentage": 39.04, "elapsed_time": "3:59:53", "remaining_time": "6:14:34"}
+{"current_steps": 343, "total_steps": 876, "loss": 0.5777, "lr": 7.631163789226463e-06, "epoch": 1.1729837560558565, "percentage": 39.16, "elapsed_time": "4:00:37", "remaining_time": "6:13:54"}
+{"current_steps": 344, "total_steps": 876, "loss": 0.5552, "lr": 7.6141922657079045e-06, "epoch": 1.176403533770305, "percentage": 39.27, "elapsed_time": "4:01:18", "remaining_time": "6:13:11"}
+{"current_steps": 345, "total_steps": 876, "loss": 0.5496, "lr": 7.597179190923343e-06, "epoch": 1.1798233114847534, "percentage": 39.38, "elapsed_time": "4:02:00", "remaining_time": "6:12:29"}
+{"current_steps": 346, "total_steps": 876, "loss": 0.569, "lr": 7.580124835287013e-06, "epoch": 1.183243089199202, "percentage": 39.5, "elapsed_time": "4:02:43", "remaining_time": "6:11:47"}
+{"current_steps": 347, "total_steps": 876, "loss": 0.5264, "lr": 7.563029469869283e-06, "epoch": 1.1866628669136505, "percentage": 39.61, "elapsed_time": "4:03:25", "remaining_time": "6:11:06"}
+{"current_steps": 348, "total_steps": 876, "loss": 0.5862, "lr": 7.545893366392358e-06, "epoch": 1.1900826446280992, "percentage": 39.73, "elapsed_time": "4:04:07", "remaining_time": "6:10:24"}
+{"current_steps": 349, "total_steps": 876, "loss": 0.5818, "lr": 7.5287167972259455e-06, "epoch": 1.1935024223425477, "percentage": 39.84, "elapsed_time": "4:04:50", "remaining_time": "6:09:42"}
+{"current_steps": 350, "total_steps": 876, "loss": 0.5555, "lr": 7.511500035382943e-06, "epoch": 1.1969222000569963, "percentage": 39.95, "elapsed_time": "4:05:33", "remaining_time": "6:09:02"}
+{"current_steps": 351, "total_steps": 876, "loss": 0.609, "lr": 7.494243354515083e-06, "epoch": 1.2003419777714448, "percentage": 40.07, "elapsed_time": "4:06:18", "remaining_time": "6:08:24"}
+{"current_steps": 352, "total_steps": 876, "loss": 0.5554, "lr": 7.476947028908595e-06, "epoch": 1.2037617554858935, "percentage": 40.18, "elapsed_time": "4:07:01", "remaining_time": "6:07:43"}
+{"current_steps": 353, "total_steps": 876, "loss": 0.5898, "lr": 7.45961133347984e-06, "epoch": 1.207181533200342, "percentage": 40.3, "elapsed_time": "4:07:43", "remaining_time": "6:07:01"}
+{"current_steps": 354, "total_steps": 876, "loss": 0.5734, "lr": 7.442236543770945e-06, "epoch": 1.2106013109147906, "percentage": 40.41, "elapsed_time": "4:08:25", "remaining_time": "6:06:19"}
+{"current_steps": 355, "total_steps": 876, "loss": 0.583, "lr": 7.424822935945416e-06, "epoch": 1.214021088629239, "percentage": 40.53, "elapsed_time": "4:09:08", "remaining_time": "6:05:38"}
+{"current_steps": 356, "total_steps": 876, "loss": 0.5594, "lr": 7.407370786783757e-06, "epoch": 1.2174408663436878, "percentage": 40.64, "elapsed_time": "4:09:50", "remaining_time": "6:04:55"}
+{"current_steps": 357, "total_steps": 876, "loss": 0.5966, "lr": 7.3898803736790706e-06, "epoch": 1.2208606440581362, "percentage": 40.75, "elapsed_time": "4:10:32", "remaining_time": "6:04:13"}
+{"current_steps": 358, "total_steps": 876, "loss": 0.529, "lr": 7.372351974632634e-06, "epoch": 1.2242804217725847, "percentage": 40.87, "elapsed_time": "4:11:14", "remaining_time": "6:03:31"}
+{"current_steps": 359, "total_steps": 876, "loss": 0.5716, "lr": 7.354785868249504e-06, "epoch": 1.2277001994870334, "percentage": 40.98, "elapsed_time": "4:11:55", "remaining_time": "6:02:48"}
+{"current_steps": 360, "total_steps": 876, "loss": 0.6037, "lr": 7.33718233373407e-06, "epoch": 1.2311199772014818, "percentage": 41.1, "elapsed_time": "4:12:38", "remaining_time": "6:02:06"}
+{"current_steps": 361, "total_steps": 876, "loss": 0.5657, "lr": 7.319541650885623e-06, "epoch": 1.2345397549159305, "percentage": 41.21, "elapsed_time": "4:13:20", "remaining_time": "6:01:24"}
+{"current_steps": 362, "total_steps": 876, "loss": 0.5566, "lr": 7.3018641000939115e-06, "epoch": 1.237959532630379, "percentage": 41.32, "elapsed_time": "4:14:01", "remaining_time": "6:00:40"}
+{"current_steps": 363, "total_steps": 876, "loss": 0.5689, "lr": 7.2841499623346755e-06, "epoch": 1.2413793103448276, "percentage": 41.44, "elapsed_time": "4:14:42", "remaining_time": "5:59:57"}
+{"current_steps": 364, "total_steps": 876, "loss": 0.5437, "lr": 7.266399519165193e-06, "epoch": 1.244799088059276, "percentage": 41.55, "elapsed_time": "4:15:23", "remaining_time": "5:59:14"}
+{"current_steps": 365, "total_steps": 876, "loss": 0.5398, "lr": 7.248613052719793e-06, "epoch": 1.2482188657737248, "percentage": 41.67, "elapsed_time": "4:16:06", "remaining_time": "5:58:33"}
+{"current_steps": 366, "total_steps": 876, "loss": 0.5643, "lr": 7.2307908457053786e-06, "epoch": 1.2516386434881732, "percentage": 41.78, "elapsed_time": "4:16:49", "remaining_time": "5:57:52"}
+{"current_steps": 367, "total_steps": 876, "loss": 0.5905, "lr": 7.2129331813969315e-06, "epoch": 1.255058421202622, "percentage": 41.89, "elapsed_time": "4:17:31", "remaining_time": "5:57:09"}
+{"current_steps": 368, "total_steps": 876, "loss": 0.5902, "lr": 7.195040343633006e-06, "epoch": 1.2584781989170704, "percentage": 42.01, "elapsed_time": "4:18:12", "remaining_time": "5:56:26"}
+{"current_steps": 369, "total_steps": 876, "loss": 0.5705, "lr": 7.177112616811228e-06, "epoch": 1.261897976631519, "percentage": 42.12, "elapsed_time": "4:18:54", "remaining_time": "5:55:44"}
+{"current_steps": 370, "total_steps": 876, "loss": 0.5726, "lr": 7.159150285883757e-06, "epoch": 1.2653177543459675, "percentage": 42.24, "elapsed_time": "4:19:36", "remaining_time": "5:55:02"}
+{"current_steps": 371, "total_steps": 876, "loss": 0.6144, "lr": 7.141153636352778e-06, "epoch": 1.268737532060416, "percentage": 42.35, "elapsed_time": "4:20:17", "remaining_time": "5:54:18"}
+{"current_steps": 372, "total_steps": 876, "loss": 0.5557, "lr": 7.123122954265942e-06, "epoch": 1.2721573097748646, "percentage": 42.47, "elapsed_time": "4:20:59", "remaining_time": "5:53:36"}
+{"current_steps": 373, "total_steps": 876, "loss": 0.5938, "lr": 7.105058526211842e-06, "epoch": 1.2755770874893133, "percentage": 42.58, "elapsed_time": "4:21:41", "remaining_time": "5:52:54"}
+{"current_steps": 374, "total_steps": 876, "loss": 0.5723, "lr": 7.086960639315437e-06, "epoch": 1.2789968652037618, "percentage": 42.69, "elapsed_time": "4:22:22", "remaining_time": "5:52:10"}
+{"current_steps": 375, "total_steps": 876, "loss": 0.6185, "lr": 7.0688295812334995e-06, "epoch": 1.2824166429182102, "percentage": 42.81, "elapsed_time": "4:23:04", "remaining_time": "5:51:28"}
+{"current_steps": 376, "total_steps": 876, "loss": 0.5739, "lr": 7.050665640150045e-06, "epoch": 1.285836420632659, "percentage": 42.92, "elapsed_time": "4:23:45", "remaining_time": "5:50:45"}
+{"current_steps": 377, "total_steps": 876, "loss": 0.5928, "lr": 7.032469104771742e-06, "epoch": 1.2892561983471074, "percentage": 43.04, "elapsed_time": "4:24:27", "remaining_time": "5:50:02"}
+{"current_steps": 378, "total_steps": 876, "loss": 0.5261, "lr": 7.0142402643233346e-06, "epoch": 1.292675976061556, "percentage": 43.15, "elapsed_time": "4:25:10", "remaining_time": "5:49:21"}
+{"current_steps": 379, "total_steps": 876, "loss": 0.5592, "lr": 6.995979408543034e-06, "epoch": 1.2960957537760045, "percentage": 43.26, "elapsed_time": "4:25:53", "remaining_time": "5:48:40"}
+{"current_steps": 380, "total_steps": 876, "loss": 0.5425, "lr": 6.977686827677926e-06, "epoch": 1.2995155314904532, "percentage": 43.38, "elapsed_time": "4:26:34", "remaining_time": "5:47:56"}
+{"current_steps": 381, "total_steps": 876, "loss": 0.5801, "lr": 6.959362812479342e-06, "epoch": 1.3029353092049016, "percentage": 43.49, "elapsed_time": "4:27:18", "remaining_time": "5:47:16"}
+{"current_steps": 382, "total_steps": 876, "loss": 0.5767, "lr": 6.941007654198254e-06, "epoch": 1.3063550869193503, "percentage": 43.61, "elapsed_time": "4:28:00", "remaining_time": "5:46:34"}
+{"current_steps": 383, "total_steps": 876, "loss": 0.541, "lr": 6.9226216445806315e-06, "epoch": 1.3097748646337988, "percentage": 43.72, "elapsed_time": "4:28:46", "remaining_time": "5:45:57"}
+{"current_steps": 384, "total_steps": 876, "loss": 0.565, "lr": 6.904205075862816e-06, "epoch": 1.3131946423482472, "percentage": 43.84, "elapsed_time": "4:29:27", "remaining_time": "5:45:14"}
+{"current_steps": 385, "total_steps": 876, "loss": 0.5296, "lr": 6.885758240766867e-06, "epoch": 1.316614420062696, "percentage": 43.95, "elapsed_time": "4:30:10", "remaining_time": "5:44:33"}
+{"current_steps": 386, "total_steps": 876, "loss": 0.5457, "lr": 6.867281432495911e-06, "epoch": 1.3200341977771446, "percentage": 44.06, "elapsed_time": "4:30:52", "remaining_time": "5:43:50"}
+{"current_steps": 387, "total_steps": 876, "loss": 0.5599, "lr": 6.848774944729487e-06, "epoch": 1.323453975491593, "percentage": 44.18, "elapsed_time": "4:31:34", "remaining_time": "5:43:09"}
+{"current_steps": 388, "total_steps": 876, "loss": 0.5674, "lr": 6.830239071618874e-06, "epoch": 1.3268737532060415, "percentage": 44.29, "elapsed_time": "4:32:17", "remaining_time": "5:42:27"}
+{"current_steps": 389, "total_steps": 876, "loss": 0.5413, "lr": 6.811674107782413e-06, "epoch": 1.3302935309204902, "percentage": 44.41, "elapsed_time": "4:32:58", "remaining_time": "5:41:45"}
+{"current_steps": 390, "total_steps": 876, "loss": 0.5622, "lr": 6.793080348300834e-06, "epoch": 1.3337133086349389, "percentage": 44.52, "elapsed_time": "4:33:40", "remaining_time": "5:41:02"}
+{"current_steps": 391, "total_steps": 876, "loss": 0.6199, "lr": 6.7744580887125475e-06, "epoch": 1.3371330863493873, "percentage": 44.63, "elapsed_time": "4:34:22", "remaining_time": "5:40:19"}
+{"current_steps": 392, "total_steps": 876, "loss": 0.5598, "lr": 6.755807625008974e-06, "epoch": 1.3405528640638358, "percentage": 44.75, "elapsed_time": "4:35:03", "remaining_time": "5:39:37"}
+{"current_steps": 393, "total_steps": 876, "loss": 0.5812, "lr": 6.737129253629812e-06, "epoch": 1.3439726417782845, "percentage": 44.86, "elapsed_time": "4:35:46", "remaining_time": "5:38:56"}
+{"current_steps": 394, "total_steps": 876, "loss": 0.5606, "lr": 6.718423271458343e-06, "epoch": 1.347392419492733, "percentage": 44.98, "elapsed_time": "4:36:29", "remaining_time": "5:38:14"}
+{"current_steps": 395, "total_steps": 876, "loss": 0.5564, "lr": 6.69968997581671e-06, "epoch": 1.3508121972071816, "percentage": 45.09, "elapsed_time": "4:37:10", "remaining_time": "5:37:31"}
+{"current_steps": 396, "total_steps": 876, "loss": 0.5576, "lr": 6.680929664461184e-06, "epoch": 1.35423197492163, "percentage": 45.21, "elapsed_time": "4:37:53", "remaining_time": "5:36:49"}
+{"current_steps": 397, "total_steps": 876, "loss": 0.5488, "lr": 6.662142635577444e-06, "epoch": 1.3576517526360787, "percentage": 45.32, "elapsed_time": "4:38:35", "remaining_time": "5:36:08"}
+{"current_steps": 398, "total_steps": 876, "loss": 0.5575, "lr": 6.643329187775827e-06, "epoch": 1.3610715303505272, "percentage": 45.43, "elapsed_time": "4:39:18", "remaining_time": "5:35:27"}
+{"current_steps": 399, "total_steps": 876, "loss": 0.5951, "lr": 6.624489620086584e-06, "epoch": 1.3644913080649759, "percentage": 45.55, "elapsed_time": "4:40:00", "remaining_time": "5:34:44"}
+{"current_steps": 400, "total_steps": 876, "loss": 0.5908, "lr": 6.6056242319551315e-06, "epoch": 1.3679110857794243, "percentage": 45.66, "elapsed_time": "4:40:42", "remaining_time": "5:34:02"}
+{"current_steps": 401, "total_steps": 876, "loss": 0.5759, "lr": 6.586733323237288e-06, "epoch": 1.3713308634938728, "percentage": 45.78, "elapsed_time": "4:42:47", "remaining_time": "5:34:59"}
+{"current_steps": 402, "total_steps": 876, "loss": 0.606, "lr": 6.567817194194508e-06, "epoch": 1.3747506412083215, "percentage": 45.89, "elapsed_time": "4:43:30", "remaining_time": "5:34:17"}
+{"current_steps": 403, "total_steps": 876, "loss": 0.5796, "lr": 6.548876145489109e-06, "epoch": 1.3781704189227701, "percentage": 46.0, "elapsed_time": "4:44:14", "remaining_time": "5:33:36"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:794baa3f0f497bd0f6557b141214264caa99ba25f2f36023cda7ede59b132514
+size 7736