Training in progress, epoch 1

.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 sha256:7c442abc71240e6f58675da09e5765d81b7acdb974d4d8c38cb097154b4d6586
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5cb5bd537eb3b77c87382eb60f43274e4ab51fc1b9c9029907759fffd3cc8ea
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ecf9f9daa074d2c0f83c7b2e1ccd3cef5a3090c8c7232c7c4668a7d7be21196
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6a1987aa9794b84d752b1188df73b49628db0fcc16e213866ea03d6bb9d1ff0
+size 1089994880

model.safetensors.index.json

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

special_tokens_map.json

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

tokenizer.json

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

tokenizer_config.json

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

trainer_log.jsonl

@@ -0,0 +1,501 @@
+{"current_steps": 1, "total_steps": 2470, "loss": 0.809, "lr": 1.619433198380567e-07, "epoch": 0.0020242914979757085, "percentage": 0.04, "elapsed_time": "0:00:33", "remaining_time": "22:51:22"}
+{"current_steps": 2, "total_steps": 2470, "loss": 0.8189, "lr": 3.238866396761134e-07, "epoch": 0.004048582995951417, "percentage": 0.08, "elapsed_time": "0:00:59", "remaining_time": "20:17:21"}
+{"current_steps": 3, "total_steps": 2470, "loss": 0.8304, "lr": 4.858299595141701e-07, "epoch": 0.006072874493927126, "percentage": 0.12, "elapsed_time": "0:01:24", "remaining_time": "19:19:19"}
+{"current_steps": 4, "total_steps": 2470, "loss": 0.8276, "lr": 6.477732793522268e-07, "epoch": 0.008097165991902834, "percentage": 0.16, "elapsed_time": "0:01:49", "remaining_time": "18:42:46"}
+{"current_steps": 5, "total_steps": 2470, "loss": 0.8219, "lr": 8.097165991902834e-07, "epoch": 0.010121457489878543, "percentage": 0.2, "elapsed_time": "0:02:14", "remaining_time": "18:26:49"}
+{"current_steps": 6, "total_steps": 2470, "loss": 0.8068, "lr": 9.716599190283402e-07, "epoch": 0.012145748987854251, "percentage": 0.24, "elapsed_time": "0:02:38", "remaining_time": "18:08:10"}
+{"current_steps": 7, "total_steps": 2470, "loss": 0.8092, "lr": 1.133603238866397e-06, "epoch": 0.01417004048582996, "percentage": 0.28, "elapsed_time": "0:03:04", "remaining_time": "18:02:28"}
+{"current_steps": 8, "total_steps": 2470, "loss": 0.7946, "lr": 1.2955465587044536e-06, "epoch": 0.016194331983805668, "percentage": 0.32, "elapsed_time": "0:03:29", "remaining_time": "17:52:33"}
+{"current_steps": 9, "total_steps": 2470, "loss": 0.7684, "lr": 1.4574898785425101e-06, "epoch": 0.018218623481781375, "percentage": 0.36, "elapsed_time": "0:03:54", "remaining_time": "17:50:45"}
+{"current_steps": 10, "total_steps": 2470, "loss": 0.7561, "lr": 1.6194331983805669e-06, "epoch": 0.020242914979757085, "percentage": 0.4, "elapsed_time": "0:04:19", "remaining_time": "17:45:38"}
+{"current_steps": 11, "total_steps": 2470, "loss": 0.7646, "lr": 1.7813765182186236e-06, "epoch": 0.022267206477732792, "percentage": 0.45, "elapsed_time": "0:04:43", "remaining_time": "17:35:20"}
+{"current_steps": 12, "total_steps": 2470, "loss": 0.7338, "lr": 1.9433198380566803e-06, "epoch": 0.024291497975708502, "percentage": 0.49, "elapsed_time": "0:05:08", "remaining_time": "17:32:17"}
+{"current_steps": 13, "total_steps": 2470, "loss": 0.7272, "lr": 2.105263157894737e-06, "epoch": 0.02631578947368421, "percentage": 0.53, "elapsed_time": "0:05:33", "remaining_time": "17:29:21"}
+{"current_steps": 14, "total_steps": 2470, "loss": 0.7221, "lr": 2.267206477732794e-06, "epoch": 0.02834008097165992, "percentage": 0.57, "elapsed_time": "0:05:57", "remaining_time": "17:26:26"}
+{"current_steps": 15, "total_steps": 2470, "loss": 0.7182, "lr": 2.42914979757085e-06, "epoch": 0.030364372469635626, "percentage": 0.61, "elapsed_time": "0:06:23", "remaining_time": "17:27:08"}
+{"current_steps": 16, "total_steps": 2470, "loss": 0.7164, "lr": 2.5910931174089072e-06, "epoch": 0.032388663967611336, "percentage": 0.65, "elapsed_time": "0:06:49", "remaining_time": "17:26:01"}
+{"current_steps": 17, "total_steps": 2470, "loss": 0.7186, "lr": 2.7530364372469636e-06, "epoch": 0.03441295546558704, "percentage": 0.69, "elapsed_time": "0:07:16", "remaining_time": "17:29:17"}
+{"current_steps": 18, "total_steps": 2470, "loss": 0.6935, "lr": 2.9149797570850203e-06, "epoch": 0.03643724696356275, "percentage": 0.73, "elapsed_time": "0:07:41", "remaining_time": "17:27:03"}
+{"current_steps": 19, "total_steps": 2470, "loss": 0.6898, "lr": 3.0769230769230774e-06, "epoch": 0.038461538461538464, "percentage": 0.77, "elapsed_time": "0:08:07", "remaining_time": "17:28:24"}
+{"current_steps": 20, "total_steps": 2470, "loss": 0.6954, "lr": 3.2388663967611337e-06, "epoch": 0.04048582995951417, "percentage": 0.81, "elapsed_time": "0:08:32", "remaining_time": "17:27:20"}
+{"current_steps": 21, "total_steps": 2470, "loss": 0.6695, "lr": 3.4008097165991905e-06, "epoch": 0.04251012145748988, "percentage": 0.85, "elapsed_time": "0:08:56", "remaining_time": "17:23:07"}
+{"current_steps": 22, "total_steps": 2470, "loss": 0.6444, "lr": 3.562753036437247e-06, "epoch": 0.044534412955465584, "percentage": 0.89, "elapsed_time": "0:09:21", "remaining_time": "17:21:23"}
+{"current_steps": 23, "total_steps": 2470, "loss": 0.6452, "lr": 3.724696356275304e-06, "epoch": 0.0465587044534413, "percentage": 0.93, "elapsed_time": "0:09:46", "remaining_time": "17:20:06"}
+{"current_steps": 24, "total_steps": 2470, "loss": 0.6173, "lr": 3.886639676113361e-06, "epoch": 0.048582995951417005, "percentage": 0.97, "elapsed_time": "0:10:11", "remaining_time": "17:17:53"}
+{"current_steps": 25, "total_steps": 2470, "loss": 0.639, "lr": 4.048582995951417e-06, "epoch": 0.05060728744939271, "percentage": 1.01, "elapsed_time": "0:10:35", "remaining_time": "17:15:16"}
+{"current_steps": 26, "total_steps": 2470, "loss": 0.6154, "lr": 4.210526315789474e-06, "epoch": 0.05263157894736842, "percentage": 1.05, "elapsed_time": "0:11:00", "remaining_time": "17:14:44"}
+{"current_steps": 27, "total_steps": 2470, "loss": 0.6204, "lr": 4.372469635627531e-06, "epoch": 0.05465587044534413, "percentage": 1.09, "elapsed_time": "0:11:25", "remaining_time": "17:13:14"}
+{"current_steps": 28, "total_steps": 2470, "loss": 0.6075, "lr": 4.534412955465588e-06, "epoch": 0.05668016194331984, "percentage": 1.13, "elapsed_time": "0:11:50", "remaining_time": "17:12:09"}
+{"current_steps": 29, "total_steps": 2470, "loss": 0.5806, "lr": 4.696356275303644e-06, "epoch": 0.058704453441295545, "percentage": 1.17, "elapsed_time": "0:12:16", "remaining_time": "17:12:37"}
+{"current_steps": 30, "total_steps": 2470, "loss": 0.6002, "lr": 4.8582995951417e-06, "epoch": 0.06072874493927125, "percentage": 1.21, "elapsed_time": "0:12:42", "remaining_time": "17:13:07"}
+{"current_steps": 31, "total_steps": 2470, "loss": 0.574, "lr": 5.020242914979757e-06, "epoch": 0.06275303643724696, "percentage": 1.26, "elapsed_time": "0:13:07", "remaining_time": "17:12:11"}
+{"current_steps": 32, "total_steps": 2470, "loss": 0.5872, "lr": 5.1821862348178145e-06, "epoch": 0.06477732793522267, "percentage": 1.3, "elapsed_time": "0:13:31", "remaining_time": "17:10:58"}
+{"current_steps": 33, "total_steps": 2470, "loss": 0.5584, "lr": 5.344129554655872e-06, "epoch": 0.06680161943319839, "percentage": 1.34, "elapsed_time": "0:13:57", "remaining_time": "17:10:23"}
+{"current_steps": 34, "total_steps": 2470, "loss": 0.5591, "lr": 5.506072874493927e-06, "epoch": 0.06882591093117409, "percentage": 1.38, "elapsed_time": "0:14:21", "remaining_time": "17:09:14"}
+{"current_steps": 35, "total_steps": 2470, "loss": 0.5557, "lr": 5.668016194331984e-06, "epoch": 0.0708502024291498, "percentage": 1.42, "elapsed_time": "0:14:46", "remaining_time": "17:08:19"}
+{"current_steps": 36, "total_steps": 2470, "loss": 0.5584, "lr": 5.8299595141700406e-06, "epoch": 0.0728744939271255, "percentage": 1.46, "elapsed_time": "0:15:11", "remaining_time": "17:07:01"}
+{"current_steps": 37, "total_steps": 2470, "loss": 0.5744, "lr": 5.991902834008098e-06, "epoch": 0.07489878542510121, "percentage": 1.5, "elapsed_time": "0:15:36", "remaining_time": "17:05:53"}
+{"current_steps": 38, "total_steps": 2470, "loss": 0.5473, "lr": 6.153846153846155e-06, "epoch": 0.07692307692307693, "percentage": 1.54, "elapsed_time": "0:16:01", "remaining_time": "17:05:30"}
+{"current_steps": 39, "total_steps": 2470, "loss": 0.5449, "lr": 6.31578947368421e-06, "epoch": 0.07894736842105263, "percentage": 1.58, "elapsed_time": "0:16:25", "remaining_time": "17:04:12"}
+{"current_steps": 40, "total_steps": 2470, "loss": 0.547, "lr": 6.4777327935222675e-06, "epoch": 0.08097165991902834, "percentage": 1.62, "elapsed_time": "0:16:53", "remaining_time": "17:05:41"}
+{"current_steps": 41, "total_steps": 2470, "loss": 0.5552, "lr": 6.639676113360325e-06, "epoch": 0.08299595141700405, "percentage": 1.66, "elapsed_time": "0:17:17", "remaining_time": "17:04:25"}
+{"current_steps": 42, "total_steps": 2470, "loss": 0.5494, "lr": 6.801619433198381e-06, "epoch": 0.08502024291497975, "percentage": 1.7, "elapsed_time": "0:17:42", "remaining_time": "17:03:29"}
+{"current_steps": 43, "total_steps": 2470, "loss": 0.5312, "lr": 6.963562753036438e-06, "epoch": 0.08704453441295547, "percentage": 1.74, "elapsed_time": "0:18:06", "remaining_time": "17:02:02"}
+{"current_steps": 44, "total_steps": 2470, "loss": 0.551, "lr": 7.125506072874494e-06, "epoch": 0.08906882591093117, "percentage": 1.78, "elapsed_time": "0:18:31", "remaining_time": "17:00:59"}
+{"current_steps": 45, "total_steps": 2470, "loss": 0.5269, "lr": 7.2874493927125516e-06, "epoch": 0.09109311740890688, "percentage": 1.82, "elapsed_time": "0:18:57", "remaining_time": "17:01:11"}
+{"current_steps": 46, "total_steps": 2470, "loss": 0.5277, "lr": 7.449392712550608e-06, "epoch": 0.0931174089068826, "percentage": 1.86, "elapsed_time": "0:19:22", "remaining_time": "17:00:40"}
+{"current_steps": 47, "total_steps": 2470, "loss": 0.5199, "lr": 7.611336032388664e-06, "epoch": 0.0951417004048583, "percentage": 1.9, "elapsed_time": "0:19:46", "remaining_time": "16:59:38"}
+{"current_steps": 48, "total_steps": 2470, "loss": 0.527, "lr": 7.773279352226721e-06, "epoch": 0.09716599190283401, "percentage": 1.94, "elapsed_time": "0:20:12", "remaining_time": "16:59:57"}
+{"current_steps": 49, "total_steps": 2470, "loss": 0.5121, "lr": 7.935222672064778e-06, "epoch": 0.09919028340080972, "percentage": 1.98, "elapsed_time": "0:20:37", "remaining_time": "16:58:59"}
+{"current_steps": 50, "total_steps": 2470, "loss": 0.5286, "lr": 8.097165991902834e-06, "epoch": 0.10121457489878542, "percentage": 2.02, "elapsed_time": "0:21:02", "remaining_time": "16:58:01"}
+{"current_steps": 51, "total_steps": 2470, "loss": 0.5122, "lr": 8.259109311740891e-06, "epoch": 0.10323886639676114, "percentage": 2.06, "elapsed_time": "0:21:26", "remaining_time": "16:57:12"}
+{"current_steps": 52, "total_steps": 2470, "loss": 0.5114, "lr": 8.421052631578948e-06, "epoch": 0.10526315789473684, "percentage": 2.11, "elapsed_time": "0:21:50", "remaining_time": "16:55:56"}
+{"current_steps": 53, "total_steps": 2470, "loss": 0.5177, "lr": 8.582995951417005e-06, "epoch": 0.10728744939271255, "percentage": 2.15, "elapsed_time": "0:22:15", "remaining_time": "16:54:48"}
+{"current_steps": 54, "total_steps": 2470, "loss": 0.4999, "lr": 8.744939271255063e-06, "epoch": 0.10931174089068826, "percentage": 2.19, "elapsed_time": "0:22:41", "remaining_time": "16:54:56"}
+{"current_steps": 55, "total_steps": 2470, "loss": 0.5256, "lr": 8.906882591093118e-06, "epoch": 0.11133603238866396, "percentage": 2.23, "elapsed_time": "0:23:06", "remaining_time": "16:54:54"}
+{"current_steps": 56, "total_steps": 2470, "loss": 0.5259, "lr": 9.068825910931175e-06, "epoch": 0.11336032388663968, "percentage": 2.27, "elapsed_time": "0:23:31", "remaining_time": "16:54:01"}
+{"current_steps": 57, "total_steps": 2470, "loss": 0.5086, "lr": 9.230769230769232e-06, "epoch": 0.11538461538461539, "percentage": 2.31, "elapsed_time": "0:23:55", "remaining_time": "16:53:06"}
+{"current_steps": 58, "total_steps": 2470, "loss": 0.5088, "lr": 9.392712550607288e-06, "epoch": 0.11740890688259109, "percentage": 2.35, "elapsed_time": "0:24:19", "remaining_time": "16:51:55"}
+{"current_steps": 59, "total_steps": 2470, "loss": 0.5036, "lr": 9.554655870445345e-06, "epoch": 0.1194331983805668, "percentage": 2.39, "elapsed_time": "0:24:44", "remaining_time": "16:51:11"}
+{"current_steps": 60, "total_steps": 2470, "loss": 0.4991, "lr": 9.7165991902834e-06, "epoch": 0.1214574898785425, "percentage": 2.43, "elapsed_time": "0:25:09", "remaining_time": "16:50:19"}
+{"current_steps": 61, "total_steps": 2470, "loss": 0.5103, "lr": 9.878542510121458e-06, "epoch": 0.12348178137651822, "percentage": 2.47, "elapsed_time": "0:25:33", "remaining_time": "16:49:14"}
+{"current_steps": 62, "total_steps": 2470, "loss": 0.5138, "lr": 1.0040485829959515e-05, "epoch": 0.12550607287449392, "percentage": 2.51, "elapsed_time": "0:25:58", "remaining_time": "16:48:32"}
+{"current_steps": 63, "total_steps": 2470, "loss": 0.5132, "lr": 1.0202429149797572e-05, "epoch": 0.12753036437246965, "percentage": 2.55, "elapsed_time": "0:26:21", "remaining_time": "16:47:08"}
+{"current_steps": 64, "total_steps": 2470, "loss": 0.5082, "lr": 1.0364372469635629e-05, "epoch": 0.12955465587044535, "percentage": 2.59, "elapsed_time": "0:26:47", "remaining_time": "16:47:06"}
+{"current_steps": 65, "total_steps": 2470, "loss": 0.5005, "lr": 1.0526315789473684e-05, "epoch": 0.13157894736842105, "percentage": 2.63, "elapsed_time": "0:27:10", "remaining_time": "16:45:45"}
+{"current_steps": 66, "total_steps": 2470, "loss": 0.5098, "lr": 1.0688259109311743e-05, "epoch": 0.13360323886639677, "percentage": 2.67, "elapsed_time": "0:27:35", "remaining_time": "16:44:50"}
+{"current_steps": 67, "total_steps": 2470, "loss": 0.5011, "lr": 1.0850202429149799e-05, "epoch": 0.13562753036437247, "percentage": 2.71, "elapsed_time": "0:27:59", "remaining_time": "16:44:01"}
+{"current_steps": 68, "total_steps": 2470, "loss": 0.4981, "lr": 1.1012145748987854e-05, "epoch": 0.13765182186234817, "percentage": 2.75, "elapsed_time": "0:28:24", "remaining_time": "16:43:17"}
+{"current_steps": 69, "total_steps": 2470, "loss": 0.4797, "lr": 1.1174089068825913e-05, "epoch": 0.1396761133603239, "percentage": 2.79, "elapsed_time": "0:28:48", "remaining_time": "16:42:38"}
+{"current_steps": 70, "total_steps": 2470, "loss": 0.4854, "lr": 1.1336032388663969e-05, "epoch": 0.1417004048582996, "percentage": 2.83, "elapsed_time": "0:29:11", "remaining_time": "16:41:03"}
+{"current_steps": 71, "total_steps": 2470, "loss": 0.4938, "lr": 1.1497975708502026e-05, "epoch": 0.1437246963562753, "percentage": 2.87, "elapsed_time": "0:29:36", "remaining_time": "16:40:12"}
+{"current_steps": 72, "total_steps": 2470, "loss": 0.5051, "lr": 1.1659919028340081e-05, "epoch": 0.145748987854251, "percentage": 2.91, "elapsed_time": "0:29:59", "remaining_time": "16:39:04"}
+{"current_steps": 73, "total_steps": 2470, "loss": 0.487, "lr": 1.182186234817814e-05, "epoch": 0.14777327935222673, "percentage": 2.96, "elapsed_time": "0:30:24", "remaining_time": "16:38:15"}
+{"current_steps": 74, "total_steps": 2470, "loss": 0.4884, "lr": 1.1983805668016195e-05, "epoch": 0.14979757085020243, "percentage": 3.0, "elapsed_time": "0:30:48", "remaining_time": "16:37:24"}
+{"current_steps": 75, "total_steps": 2470, "loss": 0.5021, "lr": 1.2145748987854251e-05, "epoch": 0.15182186234817813, "percentage": 3.04, "elapsed_time": "0:31:11", "remaining_time": "16:36:13"}
+{"current_steps": 76, "total_steps": 2470, "loss": 0.4966, "lr": 1.230769230769231e-05, "epoch": 0.15384615384615385, "percentage": 3.08, "elapsed_time": "0:31:36", "remaining_time": "16:35:40"}
+{"current_steps": 77, "total_steps": 2470, "loss": 0.4985, "lr": 1.2469635627530365e-05, "epoch": 0.15587044534412955, "percentage": 3.12, "elapsed_time": "0:32:01", "remaining_time": "16:35:31"}
+{"current_steps": 78, "total_steps": 2470, "loss": 0.4951, "lr": 1.263157894736842e-05, "epoch": 0.15789473684210525, "percentage": 3.16, "elapsed_time": "0:32:25", "remaining_time": "16:34:21"}
+{"current_steps": 79, "total_steps": 2470, "loss": 0.4784, "lr": 1.279352226720648e-05, "epoch": 0.15991902834008098, "percentage": 3.2, "elapsed_time": "0:32:49", "remaining_time": "16:33:29"}
+{"current_steps": 80, "total_steps": 2470, "loss": 0.4794, "lr": 1.2955465587044535e-05, "epoch": 0.16194331983805668, "percentage": 3.24, "elapsed_time": "0:33:14", "remaining_time": "16:32:51"}
+{"current_steps": 81, "total_steps": 2470, "loss": 0.4902, "lr": 1.3117408906882592e-05, "epoch": 0.16396761133603238, "percentage": 3.28, "elapsed_time": "0:33:39", "remaining_time": "16:32:52"}
+{"current_steps": 82, "total_steps": 2470, "loss": 0.4783, "lr": 1.327935222672065e-05, "epoch": 0.1659919028340081, "percentage": 3.32, "elapsed_time": "0:34:03", "remaining_time": "16:31:57"}
+{"current_steps": 83, "total_steps": 2470, "loss": 0.4819, "lr": 1.3441295546558706e-05, "epoch": 0.1680161943319838, "percentage": 3.36, "elapsed_time": "0:34:28", "remaining_time": "16:31:35"}
+{"current_steps": 84, "total_steps": 2470, "loss": 0.4804, "lr": 1.3603238866396762e-05, "epoch": 0.1700404858299595, "percentage": 3.4, "elapsed_time": "0:34:54", "remaining_time": "16:31:43"}
+{"current_steps": 85, "total_steps": 2470, "loss": 0.4768, "lr": 1.3765182186234817e-05, "epoch": 0.1720647773279352, "percentage": 3.44, "elapsed_time": "0:35:19", "remaining_time": "16:31:12"}
+{"current_steps": 86, "total_steps": 2470, "loss": 0.4804, "lr": 1.3927125506072876e-05, "epoch": 0.17408906882591094, "percentage": 3.48, "elapsed_time": "0:35:45", "remaining_time": "16:31:02"}
+{"current_steps": 87, "total_steps": 2470, "loss": 0.4991, "lr": 1.4089068825910932e-05, "epoch": 0.17611336032388664, "percentage": 3.52, "elapsed_time": "0:36:10", "remaining_time": "16:30:40"}
+{"current_steps": 88, "total_steps": 2470, "loss": 0.5008, "lr": 1.4251012145748989e-05, "epoch": 0.17813765182186234, "percentage": 3.56, "elapsed_time": "0:36:35", "remaining_time": "16:30:14"}
+{"current_steps": 89, "total_steps": 2470, "loss": 0.4836, "lr": 1.4412955465587046e-05, "epoch": 0.18016194331983806, "percentage": 3.6, "elapsed_time": "0:36:58", "remaining_time": "16:29:03"}
+{"current_steps": 90, "total_steps": 2470, "loss": 0.4806, "lr": 1.4574898785425103e-05, "epoch": 0.18218623481781376, "percentage": 3.64, "elapsed_time": "0:37:22", "remaining_time": "16:28:11"}
+{"current_steps": 91, "total_steps": 2470, "loss": 0.4832, "lr": 1.4736842105263159e-05, "epoch": 0.18421052631578946, "percentage": 3.68, "elapsed_time": "0:37:47", "remaining_time": "16:27:46"}
+{"current_steps": 92, "total_steps": 2470, "loss": 0.4755, "lr": 1.4898785425101216e-05, "epoch": 0.1862348178137652, "percentage": 3.72, "elapsed_time": "0:38:11", "remaining_time": "16:27:07"}
+{"current_steps": 93, "total_steps": 2470, "loss": 0.471, "lr": 1.5060728744939273e-05, "epoch": 0.1882591093117409, "percentage": 3.77, "elapsed_time": "0:38:36", "remaining_time": "16:26:36"}
+{"current_steps": 94, "total_steps": 2470, "loss": 0.4649, "lr": 1.5222672064777328e-05, "epoch": 0.1902834008097166, "percentage": 3.81, "elapsed_time": "0:39:00", "remaining_time": "16:25:56"}
+{"current_steps": 95, "total_steps": 2470, "loss": 0.4918, "lr": 1.5384615384615387e-05, "epoch": 0.19230769230769232, "percentage": 3.85, "elapsed_time": "0:39:26", "remaining_time": "16:25:52"}
+{"current_steps": 96, "total_steps": 2470, "loss": 0.4528, "lr": 1.5546558704453443e-05, "epoch": 0.19433198380566802, "percentage": 3.89, "elapsed_time": "0:39:50", "remaining_time": "16:25:08"}
+{"current_steps": 97, "total_steps": 2470, "loss": 0.4741, "lr": 1.5708502024291498e-05, "epoch": 0.19635627530364372, "percentage": 3.93, "elapsed_time": "0:40:16", "remaining_time": "16:25:13"}
+{"current_steps": 98, "total_steps": 2470, "loss": 0.4895, "lr": 1.5870445344129557e-05, "epoch": 0.19838056680161945, "percentage": 3.97, "elapsed_time": "0:40:40", "remaining_time": "16:24:38"}
+{"current_steps": 99, "total_steps": 2470, "loss": 0.4826, "lr": 1.6032388663967612e-05, "epoch": 0.20040485829959515, "percentage": 4.01, "elapsed_time": "0:41:05", "remaining_time": "16:24:12"}
+{"current_steps": 100, "total_steps": 2470, "loss": 0.4575, "lr": 1.6194331983805668e-05, "epoch": 0.20242914979757085, "percentage": 4.05, "elapsed_time": "0:41:30", "remaining_time": "16:23:48"}
+{"current_steps": 101, "total_steps": 2470, "loss": 0.4762, "lr": 1.6356275303643723e-05, "epoch": 0.20445344129554655, "percentage": 4.09, "elapsed_time": "0:41:55", "remaining_time": "16:23:13"}
+{"current_steps": 102, "total_steps": 2470, "loss": 0.4798, "lr": 1.6518218623481782e-05, "epoch": 0.20647773279352227, "percentage": 4.13, "elapsed_time": "0:42:19", "remaining_time": "16:22:29"}
+{"current_steps": 103, "total_steps": 2470, "loss": 0.4925, "lr": 1.6680161943319838e-05, "epoch": 0.20850202429149797, "percentage": 4.17, "elapsed_time": "0:42:43", "remaining_time": "16:21:56"}
+{"current_steps": 104, "total_steps": 2470, "loss": 0.4718, "lr": 1.6842105263157896e-05, "epoch": 0.21052631578947367, "percentage": 4.21, "elapsed_time": "0:43:08", "remaining_time": "16:21:26"}
+{"current_steps": 105, "total_steps": 2470, "loss": 0.4758, "lr": 1.7004048582995952e-05, "epoch": 0.2125506072874494, "percentage": 4.25, "elapsed_time": "0:43:33", "remaining_time": "16:20:57"}
+{"current_steps": 106, "total_steps": 2470, "loss": 0.4618, "lr": 1.716599190283401e-05, "epoch": 0.2145748987854251, "percentage": 4.29, "elapsed_time": "0:43:57", "remaining_time": "16:20:29"}
+{"current_steps": 107, "total_steps": 2470, "loss": 0.4731, "lr": 1.7327935222672066e-05, "epoch": 0.2165991902834008, "percentage": 4.33, "elapsed_time": "0:44:22", "remaining_time": "16:19:52"}
+{"current_steps": 108, "total_steps": 2470, "loss": 0.4823, "lr": 1.7489878542510125e-05, "epoch": 0.21862348178137653, "percentage": 4.37, "elapsed_time": "0:44:46", "remaining_time": "16:19:19"}
+{"current_steps": 109, "total_steps": 2470, "loss": 0.4905, "lr": 1.765182186234818e-05, "epoch": 0.22064777327935223, "percentage": 4.41, "elapsed_time": "0:45:11", "remaining_time": "16:18:53"}
+{"current_steps": 110, "total_steps": 2470, "loss": 0.4561, "lr": 1.7813765182186236e-05, "epoch": 0.22267206477732793, "percentage": 4.45, "elapsed_time": "0:45:35", "remaining_time": "16:18:15"}
+{"current_steps": 111, "total_steps": 2470, "loss": 0.4554, "lr": 1.7975708502024295e-05, "epoch": 0.22469635627530365, "percentage": 4.49, "elapsed_time": "0:45:59", "remaining_time": "16:17:32"}
+{"current_steps": 112, "total_steps": 2470, "loss": 0.4736, "lr": 1.813765182186235e-05, "epoch": 0.22672064777327935, "percentage": 4.53, "elapsed_time": "0:46:24", "remaining_time": "16:16:57"}
+{"current_steps": 113, "total_steps": 2470, "loss": 0.4845, "lr": 1.8299595141700406e-05, "epoch": 0.22874493927125505, "percentage": 4.57, "elapsed_time": "0:46:49", "remaining_time": "16:16:35"}
+{"current_steps": 114, "total_steps": 2470, "loss": 0.4721, "lr": 1.8461538461538465e-05, "epoch": 0.23076923076923078, "percentage": 4.62, "elapsed_time": "0:47:14", "remaining_time": "16:16:12"}
+{"current_steps": 115, "total_steps": 2470, "loss": 0.4885, "lr": 1.862348178137652e-05, "epoch": 0.23279352226720648, "percentage": 4.66, "elapsed_time": "0:47:38", "remaining_time": "16:15:42"}
+{"current_steps": 116, "total_steps": 2470, "loss": 0.47, "lr": 1.8785425101214576e-05, "epoch": 0.23481781376518218, "percentage": 4.7, "elapsed_time": "0:48:02", "remaining_time": "16:14:58"}
+{"current_steps": 117, "total_steps": 2470, "loss": 0.4709, "lr": 1.894736842105263e-05, "epoch": 0.23684210526315788, "percentage": 4.74, "elapsed_time": "0:48:27", "remaining_time": "16:14:31"}
+{"current_steps": 118, "total_steps": 2470, "loss": 0.4713, "lr": 1.910931174089069e-05, "epoch": 0.2388663967611336, "percentage": 4.78, "elapsed_time": "0:48:51", "remaining_time": "16:13:42"}
+{"current_steps": 119, "total_steps": 2470, "loss": 0.4625, "lr": 1.9271255060728745e-05, "epoch": 0.2408906882591093, "percentage": 4.82, "elapsed_time": "0:49:15", "remaining_time": "16:13:17"}
+{"current_steps": 120, "total_steps": 2470, "loss": 0.4546, "lr": 1.94331983805668e-05, "epoch": 0.242914979757085, "percentage": 4.86, "elapsed_time": "0:49:40", "remaining_time": "16:12:47"}
+{"current_steps": 121, "total_steps": 2470, "loss": 0.46, "lr": 1.959514170040486e-05, "epoch": 0.24493927125506074, "percentage": 4.9, "elapsed_time": "0:50:04", "remaining_time": "16:12:13"}
+{"current_steps": 122, "total_steps": 2470, "loss": 0.4656, "lr": 1.9757085020242915e-05, "epoch": 0.24696356275303644, "percentage": 4.94, "elapsed_time": "0:50:29", "remaining_time": "16:11:46"}
+{"current_steps": 123, "total_steps": 2470, "loss": 0.4561, "lr": 1.9919028340080974e-05, "epoch": 0.24898785425101214, "percentage": 4.98, "elapsed_time": "0:50:53", "remaining_time": "16:11:10"}
+{"current_steps": 124, "total_steps": 2470, "loss": 0.4439, "lr": 2.008097165991903e-05, "epoch": 0.25101214574898784, "percentage": 5.02, "elapsed_time": "0:51:17", "remaining_time": "16:10:28"}
+{"current_steps": 125, "total_steps": 2470, "loss": 0.4539, "lr": 2.0242914979757088e-05, "epoch": 0.25303643724696356, "percentage": 5.06, "elapsed_time": "0:51:41", "remaining_time": "16:09:49"}
+{"current_steps": 126, "total_steps": 2470, "loss": 0.4613, "lr": 2.0404858299595144e-05, "epoch": 0.2550607287449393, "percentage": 5.1, "elapsed_time": "0:52:06", "remaining_time": "16:09:20"}
+{"current_steps": 127, "total_steps": 2470, "loss": 0.4665, "lr": 2.05668016194332e-05, "epoch": 0.25708502024291496, "percentage": 5.14, "elapsed_time": "0:52:30", "remaining_time": "16:08:52"}
+{"current_steps": 128, "total_steps": 2470, "loss": 0.4623, "lr": 2.0728744939271258e-05, "epoch": 0.2591093117408907, "percentage": 5.18, "elapsed_time": "0:52:54", "remaining_time": "16:08:09"}
+{"current_steps": 129, "total_steps": 2470, "loss": 0.4679, "lr": 2.0890688259109317e-05, "epoch": 0.2611336032388664, "percentage": 5.22, "elapsed_time": "0:53:20", "remaining_time": "16:07:53"}
+{"current_steps": 130, "total_steps": 2470, "loss": 0.4636, "lr": 2.105263157894737e-05, "epoch": 0.2631578947368421, "percentage": 5.26, "elapsed_time": "0:53:44", "remaining_time": "16:07:27"}
+{"current_steps": 131, "total_steps": 2470, "loss": 0.4622, "lr": 2.1214574898785428e-05, "epoch": 0.2651821862348178, "percentage": 5.3, "elapsed_time": "0:54:09", "remaining_time": "16:07:01"}
+{"current_steps": 132, "total_steps": 2470, "loss": 0.4814, "lr": 2.1376518218623487e-05, "epoch": 0.26720647773279355, "percentage": 5.34, "elapsed_time": "0:54:33", "remaining_time": "16:06:25"}
+{"current_steps": 133, "total_steps": 2470, "loss": 0.4447, "lr": 2.153846153846154e-05, "epoch": 0.2692307692307692, "percentage": 5.38, "elapsed_time": "0:54:58", "remaining_time": "16:06:01"}
+{"current_steps": 134, "total_steps": 2470, "loss": 0.4697, "lr": 2.1700404858299598e-05, "epoch": 0.27125506072874495, "percentage": 5.43, "elapsed_time": "0:55:23", "remaining_time": "16:05:37"}
+{"current_steps": 135, "total_steps": 2470, "loss": 0.4544, "lr": 2.1862348178137656e-05, "epoch": 0.2732793522267207, "percentage": 5.47, "elapsed_time": "0:55:48", "remaining_time": "16:05:14"}
+{"current_steps": 136, "total_steps": 2470, "loss": 0.4617, "lr": 2.202429149797571e-05, "epoch": 0.27530364372469635, "percentage": 5.51, "elapsed_time": "0:56:13", "remaining_time": "16:04:49"}
+{"current_steps": 137, "total_steps": 2470, "loss": 0.4573, "lr": 2.2186234817813767e-05, "epoch": 0.2773279352226721, "percentage": 5.55, "elapsed_time": "0:56:37", "remaining_time": "16:04:15"}
+{"current_steps": 138, "total_steps": 2470, "loss": 0.4605, "lr": 2.2348178137651826e-05, "epoch": 0.2793522267206478, "percentage": 5.59, "elapsed_time": "0:57:02", "remaining_time": "16:04:01"}
+{"current_steps": 139, "total_steps": 2470, "loss": 0.4694, "lr": 2.2510121457489878e-05, "epoch": 0.2813765182186235, "percentage": 5.63, "elapsed_time": "0:57:28", "remaining_time": "16:03:43"}
+{"current_steps": 140, "total_steps": 2470, "loss": 0.4571, "lr": 2.2672064777327937e-05, "epoch": 0.2834008097165992, "percentage": 5.67, "elapsed_time": "0:57:54", "remaining_time": "16:03:48"}
+{"current_steps": 141, "total_steps": 2470, "loss": 0.4629, "lr": 2.2834008097165992e-05, "epoch": 0.2854251012145749, "percentage": 5.71, "elapsed_time": "0:58:19", "remaining_time": "16:03:21"}
+{"current_steps": 142, "total_steps": 2470, "loss": 0.4597, "lr": 2.299595141700405e-05, "epoch": 0.2874493927125506, "percentage": 5.75, "elapsed_time": "0:58:42", "remaining_time": "16:02:33"}
+{"current_steps": 143, "total_steps": 2470, "loss": 0.4714, "lr": 2.3157894736842107e-05, "epoch": 0.2894736842105263, "percentage": 5.79, "elapsed_time": "0:59:06", "remaining_time": "16:01:54"}
+{"current_steps": 144, "total_steps": 2470, "loss": 0.4616, "lr": 2.3319838056680162e-05, "epoch": 0.291497975708502, "percentage": 5.83, "elapsed_time": "0:59:30", "remaining_time": "16:01:16"}
+{"current_steps": 145, "total_steps": 2470, "loss": 0.4493, "lr": 2.348178137651822e-05, "epoch": 0.2935222672064777, "percentage": 5.87, "elapsed_time": "0:59:55", "remaining_time": "16:00:44"}
+{"current_steps": 146, "total_steps": 2470, "loss": 0.456, "lr": 2.364372469635628e-05, "epoch": 0.29554655870445345, "percentage": 5.91, "elapsed_time": "1:00:19", "remaining_time": "16:00:10"}
+{"current_steps": 147, "total_steps": 2470, "loss": 0.4644, "lr": 2.3805668016194332e-05, "epoch": 0.2975708502024291, "percentage": 5.95, "elapsed_time": "1:00:43", "remaining_time": "15:59:43"}
+{"current_steps": 148, "total_steps": 2470, "loss": 0.4671, "lr": 2.396761133603239e-05, "epoch": 0.29959514170040485, "percentage": 5.99, "elapsed_time": "1:01:07", "remaining_time": "15:59:07"}
+{"current_steps": 149, "total_steps": 2470, "loss": 0.4336, "lr": 2.412955465587045e-05, "epoch": 0.3016194331983806, "percentage": 6.03, "elapsed_time": "1:01:31", "remaining_time": "15:58:30"}
+{"current_steps": 150, "total_steps": 2470, "loss": 0.4603, "lr": 2.4291497975708502e-05, "epoch": 0.30364372469635625, "percentage": 6.07, "elapsed_time": "1:01:56", "remaining_time": "15:57:56"}
+{"current_steps": 151, "total_steps": 2470, "loss": 0.4486, "lr": 2.445344129554656e-05, "epoch": 0.305668016194332, "percentage": 6.11, "elapsed_time": "1:02:20", "remaining_time": "15:57:18"}
+{"current_steps": 152, "total_steps": 2470, "loss": 0.4623, "lr": 2.461538461538462e-05, "epoch": 0.3076923076923077, "percentage": 6.15, "elapsed_time": "1:02:43", "remaining_time": "15:56:36"}
+{"current_steps": 153, "total_steps": 2470, "loss": 0.4655, "lr": 2.477732793522267e-05, "epoch": 0.3097165991902834, "percentage": 6.19, "elapsed_time": "1:03:08", "remaining_time": "15:56:13"}
+{"current_steps": 154, "total_steps": 2470, "loss": 0.4516, "lr": 2.493927125506073e-05, "epoch": 0.3117408906882591, "percentage": 6.23, "elapsed_time": "1:03:33", "remaining_time": "15:55:52"}
+{"current_steps": 155, "total_steps": 2470, "loss": 0.4592, "lr": 2.510121457489879e-05, "epoch": 0.31376518218623484, "percentage": 6.28, "elapsed_time": "1:03:57", "remaining_time": "15:55:20"}
+{"current_steps": 156, "total_steps": 2470, "loss": 0.4499, "lr": 2.526315789473684e-05, "epoch": 0.3157894736842105, "percentage": 6.32, "elapsed_time": "1:04:22", "remaining_time": "15:54:54"}
+{"current_steps": 157, "total_steps": 2470, "loss": 0.4573, "lr": 2.54251012145749e-05, "epoch": 0.31781376518218624, "percentage": 6.36, "elapsed_time": "1:04:47", "remaining_time": "15:54:39"}
+{"current_steps": 158, "total_steps": 2470, "loss": 0.4563, "lr": 2.558704453441296e-05, "epoch": 0.31983805668016196, "percentage": 6.4, "elapsed_time": "1:05:12", "remaining_time": "15:54:15"}
+{"current_steps": 159, "total_steps": 2470, "loss": 0.4536, "lr": 2.5748987854251014e-05, "epoch": 0.32186234817813764, "percentage": 6.44, "elapsed_time": "1:05:37", "remaining_time": "15:53:50"}
+{"current_steps": 160, "total_steps": 2470, "loss": 0.4457, "lr": 2.591093117408907e-05, "epoch": 0.32388663967611336, "percentage": 6.48, "elapsed_time": "1:06:01", "remaining_time": "15:53:20"}
+{"current_steps": 161, "total_steps": 2470, "loss": 0.4571, "lr": 2.607287449392713e-05, "epoch": 0.3259109311740891, "percentage": 6.52, "elapsed_time": "1:06:26", "remaining_time": "15:52:48"}
+{"current_steps": 162, "total_steps": 2470, "loss": 0.4471, "lr": 2.6234817813765184e-05, "epoch": 0.32793522267206476, "percentage": 6.56, "elapsed_time": "1:06:50", "remaining_time": "15:52:14"}
+{"current_steps": 163, "total_steps": 2470, "loss": 0.4682, "lr": 2.6396761133603243e-05, "epoch": 0.3299595141700405, "percentage": 6.6, "elapsed_time": "1:07:15", "remaining_time": "15:51:51"}
+{"current_steps": 164, "total_steps": 2470, "loss": 0.4461, "lr": 2.65587044534413e-05, "epoch": 0.3319838056680162, "percentage": 6.64, "elapsed_time": "1:07:38", "remaining_time": "15:51:13"}
+{"current_steps": 165, "total_steps": 2470, "loss": 0.4549, "lr": 2.6720647773279354e-05, "epoch": 0.3340080971659919, "percentage": 6.68, "elapsed_time": "1:08:03", "remaining_time": "15:50:48"}
+{"current_steps": 166, "total_steps": 2470, "loss": 0.4522, "lr": 2.6882591093117413e-05, "epoch": 0.3360323886639676, "percentage": 6.72, "elapsed_time": "1:08:28", "remaining_time": "15:50:20"}
+{"current_steps": 167, "total_steps": 2470, "loss": 0.4534, "lr": 2.704453441295547e-05, "epoch": 0.33805668016194335, "percentage": 6.76, "elapsed_time": "1:08:53", "remaining_time": "15:49:57"}
+{"current_steps": 168, "total_steps": 2470, "loss": 0.4502, "lr": 2.7206477732793524e-05, "epoch": 0.340080971659919, "percentage": 6.8, "elapsed_time": "1:09:17", "remaining_time": "15:49:27"}
+{"current_steps": 169, "total_steps": 2470, "loss": 0.4563, "lr": 2.7368421052631583e-05, "epoch": 0.34210526315789475, "percentage": 6.84, "elapsed_time": "1:09:42", "remaining_time": "15:49:08"}
+{"current_steps": 170, "total_steps": 2470, "loss": 0.4622, "lr": 2.7530364372469635e-05, "epoch": 0.3441295546558704, "percentage": 6.88, "elapsed_time": "1:10:06", "remaining_time": "15:48:33"}
+{"current_steps": 171, "total_steps": 2470, "loss": 0.4389, "lr": 2.7692307692307694e-05, "epoch": 0.34615384615384615, "percentage": 6.92, "elapsed_time": "1:10:31", "remaining_time": "15:48:12"}
+{"current_steps": 172, "total_steps": 2470, "loss": 0.4457, "lr": 2.7854251012145752e-05, "epoch": 0.3481781376518219, "percentage": 6.96, "elapsed_time": "1:10:55", "remaining_time": "15:47:38"}
+{"current_steps": 173, "total_steps": 2470, "loss": 0.4395, "lr": 2.8016194331983804e-05, "epoch": 0.35020242914979755, "percentage": 7.0, "elapsed_time": "1:11:20", "remaining_time": "15:47:12"}
+{"current_steps": 174, "total_steps": 2470, "loss": 0.4537, "lr": 2.8178137651821863e-05, "epoch": 0.3522267206477733, "percentage": 7.04, "elapsed_time": "1:11:45", "remaining_time": "15:46:49"}
+{"current_steps": 175, "total_steps": 2470, "loss": 0.4521, "lr": 2.8340080971659922e-05, "epoch": 0.354251012145749, "percentage": 7.09, "elapsed_time": "1:12:09", "remaining_time": "15:46:23"}
+{"current_steps": 176, "total_steps": 2470, "loss": 0.4584, "lr": 2.8502024291497978e-05, "epoch": 0.3562753036437247, "percentage": 7.13, "elapsed_time": "1:12:33", "remaining_time": "15:45:49"}
+{"current_steps": 177, "total_steps": 2470, "loss": 0.4472, "lr": 2.8663967611336033e-05, "epoch": 0.3582995951417004, "percentage": 7.17, "elapsed_time": "1:12:58", "remaining_time": "15:45:20"}
+{"current_steps": 178, "total_steps": 2470, "loss": 0.4346, "lr": 2.8825910931174092e-05, "epoch": 0.3603238866396761, "percentage": 7.21, "elapsed_time": "1:13:22", "remaining_time": "15:44:42"}
+{"current_steps": 179, "total_steps": 2470, "loss": 0.4495, "lr": 2.8987854251012147e-05, "epoch": 0.3623481781376518, "percentage": 7.25, "elapsed_time": "1:13:45", "remaining_time": "15:44:05"}
+{"current_steps": 180, "total_steps": 2470, "loss": 0.4568, "lr": 2.9149797570850206e-05, "epoch": 0.3643724696356275, "percentage": 7.29, "elapsed_time": "1:14:10", "remaining_time": "15:43:35"}
+{"current_steps": 181, "total_steps": 2470, "loss": 0.4641, "lr": 2.931174089068826e-05, "epoch": 0.36639676113360325, "percentage": 7.33, "elapsed_time": "1:14:34", "remaining_time": "15:43:06"}
+{"current_steps": 182, "total_steps": 2470, "loss": 0.4275, "lr": 2.9473684210526317e-05, "epoch": 0.3684210526315789, "percentage": 7.37, "elapsed_time": "1:14:58", "remaining_time": "15:42:30"}
+{"current_steps": 183, "total_steps": 2470, "loss": 0.4404, "lr": 2.9635627530364376e-05, "epoch": 0.37044534412955465, "percentage": 7.41, "elapsed_time": "1:15:23", "remaining_time": "15:42:11"}
+{"current_steps": 184, "total_steps": 2470, "loss": 0.4506, "lr": 2.979757085020243e-05, "epoch": 0.3724696356275304, "percentage": 7.45, "elapsed_time": "1:15:47", "remaining_time": "15:41:41"}
+{"current_steps": 185, "total_steps": 2470, "loss": 0.4448, "lr": 2.9959514170040487e-05, "epoch": 0.37449392712550605, "percentage": 7.49, "elapsed_time": "1:16:11", "remaining_time": "15:41:09"}
+{"current_steps": 186, "total_steps": 2470, "loss": 0.4378, "lr": 3.0121457489878546e-05, "epoch": 0.3765182186234818, "percentage": 7.53, "elapsed_time": "1:16:36", "remaining_time": "15:40:40"}
+{"current_steps": 187, "total_steps": 2470, "loss": 0.4439, "lr": 3.0283400809716605e-05, "epoch": 0.3785425101214575, "percentage": 7.57, "elapsed_time": "1:17:01", "remaining_time": "15:40:26"}
+{"current_steps": 188, "total_steps": 2470, "loss": 0.4412, "lr": 3.0445344129554657e-05, "epoch": 0.3805668016194332, "percentage": 7.61, "elapsed_time": "1:17:26", "remaining_time": "15:40:03"}
+{"current_steps": 189, "total_steps": 2470, "loss": 0.4478, "lr": 3.060728744939271e-05, "epoch": 0.3825910931174089, "percentage": 7.65, "elapsed_time": "1:17:50", "remaining_time": "15:39:30"}
+{"current_steps": 190, "total_steps": 2470, "loss": 0.4494, "lr": 3.0769230769230774e-05, "epoch": 0.38461538461538464, "percentage": 7.69, "elapsed_time": "1:18:14", "remaining_time": "15:38:58"}
+{"current_steps": 191, "total_steps": 2470, "loss": 0.4357, "lr": 3.093117408906883e-05, "epoch": 0.3866396761133603, "percentage": 7.73, "elapsed_time": "1:18:38", "remaining_time": "15:38:20"}
+{"current_steps": 192, "total_steps": 2470, "loss": 0.4462, "lr": 3.1093117408906885e-05, "epoch": 0.38866396761133604, "percentage": 7.77, "elapsed_time": "1:19:03", "remaining_time": "15:37:57"}
+{"current_steps": 193, "total_steps": 2470, "loss": 0.4466, "lr": 3.125506072874494e-05, "epoch": 0.39068825910931176, "percentage": 7.81, "elapsed_time": "1:19:27", "remaining_time": "15:37:30"}
+{"current_steps": 194, "total_steps": 2470, "loss": 0.4409, "lr": 3.1417004048582996e-05, "epoch": 0.39271255060728744, "percentage": 7.85, "elapsed_time": "1:19:51", "remaining_time": "15:36:54"}
+{"current_steps": 195, "total_steps": 2470, "loss": 0.4415, "lr": 3.157894736842106e-05, "epoch": 0.39473684210526316, "percentage": 7.89, "elapsed_time": "1:20:16", "remaining_time": "15:36:28"}
+{"current_steps": 196, "total_steps": 2470, "loss": 0.4486, "lr": 3.1740890688259114e-05, "epoch": 0.3967611336032389, "percentage": 7.94, "elapsed_time": "1:20:40", "remaining_time": "15:36:03"}
+{"current_steps": 197, "total_steps": 2470, "loss": 0.4485, "lr": 3.190283400809717e-05, "epoch": 0.39878542510121456, "percentage": 7.98, "elapsed_time": "1:21:04", "remaining_time": "15:35:30"}
+{"current_steps": 198, "total_steps": 2470, "loss": 0.4531, "lr": 3.2064777327935225e-05, "epoch": 0.4008097165991903, "percentage": 8.02, "elapsed_time": "1:21:29", "remaining_time": "15:35:04"}
+{"current_steps": 199, "total_steps": 2470, "loss": 0.444, "lr": 3.222672064777329e-05, "epoch": 0.402834008097166, "percentage": 8.06, "elapsed_time": "1:21:54", "remaining_time": "15:34:40"}
+{"current_steps": 200, "total_steps": 2470, "loss": 0.4504, "lr": 3.2388663967611336e-05, "epoch": 0.4048582995951417, "percentage": 8.1, "elapsed_time": "1:22:18", "remaining_time": "15:34:14"}
+{"current_steps": 201, "total_steps": 2470, "loss": 0.4384, "lr": 3.25506072874494e-05, "epoch": 0.4068825910931174, "percentage": 8.14, "elapsed_time": "1:22:43", "remaining_time": "15:33:46"}
+{"current_steps": 202, "total_steps": 2470, "loss": 0.4452, "lr": 3.271255060728745e-05, "epoch": 0.4089068825910931, "percentage": 8.18, "elapsed_time": "1:23:07", "remaining_time": "15:33:18"}
+{"current_steps": 203, "total_steps": 2470, "loss": 0.4619, "lr": 3.287449392712551e-05, "epoch": 0.4109311740890688, "percentage": 8.22, "elapsed_time": "1:23:33", "remaining_time": "15:33:10"}
+{"current_steps": 204, "total_steps": 2470, "loss": 0.4525, "lr": 3.3036437246963564e-05, "epoch": 0.41295546558704455, "percentage": 8.26, "elapsed_time": "1:23:59", "remaining_time": "15:32:55"}
+{"current_steps": 205, "total_steps": 2470, "loss": 0.4445, "lr": 3.319838056680162e-05, "epoch": 0.4149797570850202, "percentage": 8.3, "elapsed_time": "1:24:24", "remaining_time": "15:32:33"}
+{"current_steps": 206, "total_steps": 2470, "loss": 0.451, "lr": 3.3360323886639675e-05, "epoch": 0.41700404858299595, "percentage": 8.34, "elapsed_time": "1:24:48", "remaining_time": "15:32:06"}
+{"current_steps": 207, "total_steps": 2470, "loss": 0.4405, "lr": 3.352226720647774e-05, "epoch": 0.4190283400809717, "percentage": 8.38, "elapsed_time": "1:25:12", "remaining_time": "15:31:36"}
+{"current_steps": 208, "total_steps": 2470, "loss": 0.4394, "lr": 3.368421052631579e-05, "epoch": 0.42105263157894735, "percentage": 8.42, "elapsed_time": "1:25:37", "remaining_time": "15:31:05"}
+{"current_steps": 209, "total_steps": 2470, "loss": 0.437, "lr": 3.384615384615385e-05, "epoch": 0.4230769230769231, "percentage": 8.46, "elapsed_time": "1:26:01", "remaining_time": "15:30:40"}
+{"current_steps": 210, "total_steps": 2470, "loss": 0.4505, "lr": 3.4008097165991904e-05, "epoch": 0.4251012145748988, "percentage": 8.5, "elapsed_time": "1:26:26", "remaining_time": "15:30:14"}
+{"current_steps": 211, "total_steps": 2470, "loss": 0.4559, "lr": 3.417004048582996e-05, "epoch": 0.4271255060728745, "percentage": 8.54, "elapsed_time": "1:26:50", "remaining_time": "15:29:40"}
+{"current_steps": 212, "total_steps": 2470, "loss": 0.44, "lr": 3.433198380566802e-05, "epoch": 0.4291497975708502, "percentage": 8.58, "elapsed_time": "1:27:15", "remaining_time": "15:29:25"}
+{"current_steps": 213, "total_steps": 2470, "loss": 0.4338, "lr": 3.449392712550608e-05, "epoch": 0.4311740890688259, "percentage": 8.62, "elapsed_time": "1:27:39", "remaining_time": "15:28:55"}
+{"current_steps": 214, "total_steps": 2470, "loss": 0.4391, "lr": 3.465587044534413e-05, "epoch": 0.4331983805668016, "percentage": 8.66, "elapsed_time": "1:28:05", "remaining_time": "15:28:35"}
+{"current_steps": 215, "total_steps": 2470, "loss": 0.4412, "lr": 3.481781376518219e-05, "epoch": 0.4352226720647773, "percentage": 8.7, "elapsed_time": "1:28:29", "remaining_time": "15:28:09"}
+{"current_steps": 216, "total_steps": 2470, "loss": 0.4398, "lr": 3.497975708502025e-05, "epoch": 0.43724696356275305, "percentage": 8.74, "elapsed_time": "1:28:54", "remaining_time": "15:27:41"}
+{"current_steps": 217, "total_steps": 2470, "loss": 0.4447, "lr": 3.51417004048583e-05, "epoch": 0.4392712550607287, "percentage": 8.79, "elapsed_time": "1:29:18", "remaining_time": "15:27:11"}
+{"current_steps": 218, "total_steps": 2470, "loss": 0.4316, "lr": 3.530364372469636e-05, "epoch": 0.44129554655870445, "percentage": 8.83, "elapsed_time": "1:29:41", "remaining_time": "15:26:30"}
+{"current_steps": 219, "total_steps": 2470, "loss": 0.4416, "lr": 3.5465587044534417e-05, "epoch": 0.4433198380566802, "percentage": 8.87, "elapsed_time": "1:30:06", "remaining_time": "15:26:07"}
+{"current_steps": 220, "total_steps": 2470, "loss": 0.4457, "lr": 3.562753036437247e-05, "epoch": 0.44534412955465585, "percentage": 8.91, "elapsed_time": "1:30:30", "remaining_time": "15:25:42"}
+{"current_steps": 221, "total_steps": 2470, "loss": 0.4424, "lr": 3.578947368421053e-05, "epoch": 0.4473684210526316, "percentage": 8.95, "elapsed_time": "1:30:54", "remaining_time": "15:25:06"}
+{"current_steps": 222, "total_steps": 2470, "loss": 0.4347, "lr": 3.595141700404859e-05, "epoch": 0.4493927125506073, "percentage": 8.99, "elapsed_time": "1:31:19", "remaining_time": "15:24:41"}
+{"current_steps": 223, "total_steps": 2470, "loss": 0.4602, "lr": 3.611336032388664e-05, "epoch": 0.451417004048583, "percentage": 9.03, "elapsed_time": "1:31:43", "remaining_time": "15:24:19"}
+{"current_steps": 224, "total_steps": 2470, "loss": 0.4367, "lr": 3.62753036437247e-05, "epoch": 0.4534412955465587, "percentage": 9.07, "elapsed_time": "1:32:07", "remaining_time": "15:23:44"}
+{"current_steps": 225, "total_steps": 2470, "loss": 0.4423, "lr": 3.6437246963562756e-05, "epoch": 0.45546558704453444, "percentage": 9.11, "elapsed_time": "1:32:32", "remaining_time": "15:23:16"}
+{"current_steps": 226, "total_steps": 2470, "loss": 0.4329, "lr": 3.659919028340081e-05, "epoch": 0.4574898785425101, "percentage": 9.15, "elapsed_time": "1:32:57", "remaining_time": "15:22:57"}
+{"current_steps": 227, "total_steps": 2470, "loss": 0.4329, "lr": 3.676113360323887e-05, "epoch": 0.45951417004048584, "percentage": 9.19, "elapsed_time": "1:33:21", "remaining_time": "15:22:27"}
+{"current_steps": 228, "total_steps": 2470, "loss": 0.4522, "lr": 3.692307692307693e-05, "epoch": 0.46153846153846156, "percentage": 9.23, "elapsed_time": "1:33:46", "remaining_time": "15:22:06"}
+{"current_steps": 229, "total_steps": 2470, "loss": 0.4502, "lr": 3.7085020242914985e-05, "epoch": 0.46356275303643724, "percentage": 9.27, "elapsed_time": "1:34:10", "remaining_time": "15:21:39"}
+{"current_steps": 230, "total_steps": 2470, "loss": 0.4319, "lr": 3.724696356275304e-05, "epoch": 0.46558704453441296, "percentage": 9.31, "elapsed_time": "1:34:35", "remaining_time": "15:21:11"}
+{"current_steps": 231, "total_steps": 2470, "loss": 0.4466, "lr": 3.7408906882591096e-05, "epoch": 0.4676113360323887, "percentage": 9.35, "elapsed_time": "1:35:00", "remaining_time": "15:20:49"}
+{"current_steps": 232, "total_steps": 2470, "loss": 0.4457, "lr": 3.757085020242915e-05, "epoch": 0.46963562753036436, "percentage": 9.39, "elapsed_time": "1:35:24", "remaining_time": "15:20:21"}
+{"current_steps": 233, "total_steps": 2470, "loss": 0.4286, "lr": 3.773279352226721e-05, "epoch": 0.4716599190283401, "percentage": 9.43, "elapsed_time": "1:35:50", "remaining_time": "15:20:12"}
+{"current_steps": 234, "total_steps": 2470, "loss": 0.4377, "lr": 3.789473684210526e-05, "epoch": 0.47368421052631576, "percentage": 9.47, "elapsed_time": "1:36:15", "remaining_time": "15:19:45"}
+{"current_steps": 235, "total_steps": 2470, "loss": 0.4374, "lr": 3.8056680161943324e-05, "epoch": 0.4757085020242915, "percentage": 9.51, "elapsed_time": "1:36:39", "remaining_time": "15:19:18"}
+{"current_steps": 236, "total_steps": 2470, "loss": 0.4311, "lr": 3.821862348178138e-05, "epoch": 0.4777327935222672, "percentage": 9.55, "elapsed_time": "1:37:04", "remaining_time": "15:18:51"}
+{"current_steps": 237, "total_steps": 2470, "loss": 0.4447, "lr": 3.8380566801619435e-05, "epoch": 0.4797570850202429, "percentage": 9.6, "elapsed_time": "1:37:28", "remaining_time": "15:18:24"}
+{"current_steps": 238, "total_steps": 2470, "loss": 0.4342, "lr": 3.854251012145749e-05, "epoch": 0.4817813765182186, "percentage": 9.64, "elapsed_time": "1:37:53", "remaining_time": "15:18:01"}
+{"current_steps": 239, "total_steps": 2470, "loss": 0.4338, "lr": 3.870445344129555e-05, "epoch": 0.48380566801619435, "percentage": 9.68, "elapsed_time": "1:38:17", "remaining_time": "15:17:31"}
+{"current_steps": 240, "total_steps": 2470, "loss": 0.4327, "lr": 3.88663967611336e-05, "epoch": 0.48582995951417, "percentage": 9.72, "elapsed_time": "1:38:42", "remaining_time": "15:17:10"}
+{"current_steps": 241, "total_steps": 2470, "loss": 0.4322, "lr": 3.9028340080971664e-05, "epoch": 0.48785425101214575, "percentage": 9.76, "elapsed_time": "1:39:06", "remaining_time": "15:16:39"}
+{"current_steps": 242, "total_steps": 2470, "loss": 0.45, "lr": 3.919028340080972e-05, "epoch": 0.4898785425101215, "percentage": 9.8, "elapsed_time": "1:39:30", "remaining_time": "15:16:04"}
+{"current_steps": 243, "total_steps": 2470, "loss": 0.4253, "lr": 3.9352226720647775e-05, "epoch": 0.49190283400809715, "percentage": 9.84, "elapsed_time": "1:39:54", "remaining_time": "15:15:40"}
+{"current_steps": 244, "total_steps": 2470, "loss": 0.4546, "lr": 3.951417004048583e-05, "epoch": 0.4939271255060729, "percentage": 9.88, "elapsed_time": "1:40:19", "remaining_time": "15:15:16"}
+{"current_steps": 245, "total_steps": 2470, "loss": 0.4322, "lr": 3.967611336032389e-05, "epoch": 0.4959514170040486, "percentage": 9.92, "elapsed_time": "1:40:44", "remaining_time": "15:14:52"}
+{"current_steps": 246, "total_steps": 2470, "loss": 0.4422, "lr": 3.983805668016195e-05, "epoch": 0.4979757085020243, "percentage": 9.96, "elapsed_time": "1:41:08", "remaining_time": "15:14:25"}
+{"current_steps": 247, "total_steps": 2470, "loss": 0.4374, "lr": 4e-05, "epoch": 0.5, "percentage": 10.0, "elapsed_time": "1:41:33", "remaining_time": "15:14:00"}
+{"current_steps": 248, "total_steps": 2470, "loss": 0.4267, "lr": 3.999998002803723e-05, "epoch": 0.5020242914979757, "percentage": 10.04, "elapsed_time": "1:41:57", "remaining_time": "15:13:31"}
+{"current_steps": 249, "total_steps": 2470, "loss": 0.4375, "lr": 3.999992011218883e-05, "epoch": 0.5040485829959515, "percentage": 10.08, "elapsed_time": "1:42:21", "remaining_time": "15:12:59"}
+{"current_steps": 250, "total_steps": 2470, "loss": 0.4362, "lr": 3.999982025257444e-05, "epoch": 0.5060728744939271, "percentage": 10.12, "elapsed_time": "1:42:46", "remaining_time": "15:12:38"}
+{"current_steps": 251, "total_steps": 2470, "loss": 0.4421, "lr": 3.9999680449393514e-05, "epoch": 0.5080971659919028, "percentage": 10.16, "elapsed_time": "1:43:10", "remaining_time": "15:12:10"}
+{"current_steps": 252, "total_steps": 2470, "loss": 0.4601, "lr": 3.999950070292526e-05, "epoch": 0.5101214574898786, "percentage": 10.2, "elapsed_time": "1:43:35", "remaining_time": "15:11:46"}
+{"current_steps": 253, "total_steps": 2470, "loss": 0.4287, "lr": 3.999928101352868e-05, "epoch": 0.5121457489878543, "percentage": 10.24, "elapsed_time": "1:44:00", "remaining_time": "15:11:23"}
+{"current_steps": 254, "total_steps": 2470, "loss": 0.4382, "lr": 3.999902138164251e-05, "epoch": 0.5141700404858299, "percentage": 10.28, "elapsed_time": "1:44:25", "remaining_time": "15:10:59"}
+{"current_steps": 255, "total_steps": 2470, "loss": 0.4305, "lr": 3.999872180778531e-05, "epoch": 0.5161943319838057, "percentage": 10.32, "elapsed_time": "1:44:48", "remaining_time": "15:10:25"}
+{"current_steps": 256, "total_steps": 2470, "loss": 0.4361, "lr": 3.999838229255538e-05, "epoch": 0.5182186234817814, "percentage": 10.36, "elapsed_time": "1:45:13", "remaining_time": "15:09:58"}
+{"current_steps": 257, "total_steps": 2470, "loss": 0.4458, "lr": 3.99980028366308e-05, "epoch": 0.520242914979757, "percentage": 10.4, "elapsed_time": "1:45:37", "remaining_time": "15:09:34"}
+{"current_steps": 258, "total_steps": 2470, "loss": 0.4406, "lr": 3.9997583440769415e-05, "epoch": 0.5222672064777328, "percentage": 10.45, "elapsed_time": "1:46:02", "remaining_time": "15:09:09"}
+{"current_steps": 259, "total_steps": 2470, "loss": 0.4406, "lr": 3.999712410580884e-05, "epoch": 0.5242914979757085, "percentage": 10.49, "elapsed_time": "1:46:27", "remaining_time": "15:08:46"}
+{"current_steps": 260, "total_steps": 2470, "loss": 0.4386, "lr": 3.999662483266646e-05, "epoch": 0.5263157894736842, "percentage": 10.53, "elapsed_time": "1:46:52", "remaining_time": "15:08:22"}
+{"current_steps": 261, "total_steps": 2470, "loss": 0.4376, "lr": 3.999608562233942e-05, "epoch": 0.52834008097166, "percentage": 10.57, "elapsed_time": "1:47:16", "remaining_time": "15:07:57"}
+{"current_steps": 262, "total_steps": 2470, "loss": 0.439, "lr": 3.999550647590464e-05, "epoch": 0.5303643724696356, "percentage": 10.61, "elapsed_time": "1:47:41", "remaining_time": "15:07:30"}
+{"current_steps": 263, "total_steps": 2470, "loss": 0.4397, "lr": 3.9994887394518765e-05, "epoch": 0.5323886639676113, "percentage": 10.65, "elapsed_time": "1:48:06", "remaining_time": "15:07:09"}
+{"current_steps": 264, "total_steps": 2470, "loss": 0.4265, "lr": 3.999422837941824e-05, "epoch": 0.5344129554655871, "percentage": 10.69, "elapsed_time": "1:48:31", "remaining_time": "15:06:46"}
+{"current_steps": 265, "total_steps": 2470, "loss": 0.4361, "lr": 3.999352943191924e-05, "epoch": 0.5364372469635628, "percentage": 10.73, "elapsed_time": "1:48:55", "remaining_time": "15:06:22"}
+{"current_steps": 266, "total_steps": 2470, "loss": 0.4208, "lr": 3.999279055341771e-05, "epoch": 0.5384615384615384, "percentage": 10.77, "elapsed_time": "1:49:20", "remaining_time": "15:05:55"}
+{"current_steps": 267, "total_steps": 2470, "loss": 0.4302, "lr": 3.9992011745389335e-05, "epoch": 0.5404858299595142, "percentage": 10.81, "elapsed_time": "1:49:44", "remaining_time": "15:05:31"}
+{"current_steps": 268, "total_steps": 2470, "loss": 0.4308, "lr": 3.999119300938953e-05, "epoch": 0.5425101214574899, "percentage": 10.85, "elapsed_time": "1:50:07", "remaining_time": "15:04:52"}
+{"current_steps": 269, "total_steps": 2470, "loss": 0.434, "lr": 3.999033434705348e-05, "epoch": 0.5445344129554656, "percentage": 10.89, "elapsed_time": "1:50:32", "remaining_time": "15:04:25"}
+{"current_steps": 270, "total_steps": 2470, "loss": 0.4421, "lr": 3.998943576009612e-05, "epoch": 0.5465587044534413, "percentage": 10.93, "elapsed_time": "1:50:55", "remaining_time": "15:03:53"}
+{"current_steps": 271, "total_steps": 2470, "loss": 0.4253, "lr": 3.998849725031208e-05, "epoch": 0.548582995951417, "percentage": 10.97, "elapsed_time": "1:51:19", "remaining_time": "15:03:20"}
+{"current_steps": 272, "total_steps": 2470, "loss": 0.4369, "lr": 3.998751881957576e-05, "epoch": 0.5506072874493927, "percentage": 11.01, "elapsed_time": "1:51:44", "remaining_time": "15:02:57"}
+{"current_steps": 273, "total_steps": 2470, "loss": 0.4277, "lr": 3.9986500469841275e-05, "epoch": 0.5526315789473685, "percentage": 11.05, "elapsed_time": "1:52:08", "remaining_time": "15:02:26"}
+{"current_steps": 274, "total_steps": 2470, "loss": 0.4285, "lr": 3.9985442203142475e-05, "epoch": 0.5546558704453441, "percentage": 11.09, "elapsed_time": "1:52:32", "remaining_time": "15:01:59"}
+{"current_steps": 275, "total_steps": 2470, "loss": 0.4374, "lr": 3.998434402159291e-05, "epoch": 0.5566801619433198, "percentage": 11.13, "elapsed_time": "1:52:57", "remaining_time": "15:01:34"}
+{"current_steps": 276, "total_steps": 2470, "loss": 0.4368, "lr": 3.9983205927385895e-05, "epoch": 0.5587044534412956, "percentage": 11.17, "elapsed_time": "1:53:22", "remaining_time": "15:01:12"}
+{"current_steps": 277, "total_steps": 2470, "loss": 0.4283, "lr": 3.99820279227944e-05, "epoch": 0.5607287449392713, "percentage": 11.21, "elapsed_time": "1:53:47", "remaining_time": "15:00:49"}
+{"current_steps": 278, "total_steps": 2470, "loss": 0.4286, "lr": 3.998081001017114e-05, "epoch": 0.562753036437247, "percentage": 11.26, "elapsed_time": "1:54:11", "remaining_time": "15:00:24"}
+{"current_steps": 279, "total_steps": 2470, "loss": 0.4314, "lr": 3.997955219194852e-05, "epoch": 0.5647773279352226, "percentage": 11.3, "elapsed_time": "1:54:36", "remaining_time": "14:59:58"}
+{"current_steps": 280, "total_steps": 2470, "loss": 0.4237, "lr": 3.997825447063866e-05, "epoch": 0.5668016194331984, "percentage": 11.34, "elapsed_time": "1:54:59", "remaining_time": "14:59:27"}
+{"current_steps": 281, "total_steps": 2470, "loss": 0.4326, "lr": 3.997691684883336e-05, "epoch": 0.5688259109311741, "percentage": 11.38, "elapsed_time": "1:55:24", "remaining_time": "14:59:02"}
+{"current_steps": 282, "total_steps": 2470, "loss": 0.4256, "lr": 3.9975539329204116e-05, "epoch": 0.5708502024291497, "percentage": 11.42, "elapsed_time": "1:55:50", "remaining_time": "14:58:48"}
+{"current_steps": 283, "total_steps": 2470, "loss": 0.4307, "lr": 3.9974121914502094e-05, "epoch": 0.5728744939271255, "percentage": 11.46, "elapsed_time": "1:56:15", "remaining_time": "14:58:26"}
+{"current_steps": 284, "total_steps": 2470, "loss": 0.4285, "lr": 3.997266460755816e-05, "epoch": 0.5748987854251012, "percentage": 11.5, "elapsed_time": "1:56:39", "remaining_time": "14:57:57"}
+{"current_steps": 285, "total_steps": 2470, "loss": 0.428, "lr": 3.9971167411282835e-05, "epoch": 0.5769230769230769, "percentage": 11.54, "elapsed_time": "1:57:04", "remaining_time": "14:57:30"}
+{"current_steps": 286, "total_steps": 2470, "loss": 0.4323, "lr": 3.996963032866633e-05, "epoch": 0.5789473684210527, "percentage": 11.58, "elapsed_time": "1:57:28", "remaining_time": "14:57:07"}
+{"current_steps": 287, "total_steps": 2470, "loss": 0.4114, "lr": 3.996805336277848e-05, "epoch": 0.5809716599190283, "percentage": 11.62, "elapsed_time": "1:57:53", "remaining_time": "14:56:44"}
+{"current_steps": 288, "total_steps": 2470, "loss": 0.4205, "lr": 3.996643651676881e-05, "epoch": 0.582995951417004, "percentage": 11.66, "elapsed_time": "1:58:18", "remaining_time": "14:56:20"}
+{"current_steps": 289, "total_steps": 2470, "loss": 0.4352, "lr": 3.996477979386646e-05, "epoch": 0.5850202429149798, "percentage": 11.7, "elapsed_time": "1:58:42", "remaining_time": "14:55:54"}
+{"current_steps": 290, "total_steps": 2470, "loss": 0.4292, "lr": 3.9963083197380256e-05, "epoch": 0.5870445344129555, "percentage": 11.74, "elapsed_time": "1:59:07", "remaining_time": "14:55:31"}
+{"current_steps": 291, "total_steps": 2470, "loss": 0.431, "lr": 3.996134673069862e-05, "epoch": 0.5890688259109311, "percentage": 11.78, "elapsed_time": "1:59:32", "remaining_time": "14:55:05"}
+{"current_steps": 292, "total_steps": 2470, "loss": 0.4413, "lr": 3.995957039728962e-05, "epoch": 0.5910931174089069, "percentage": 11.82, "elapsed_time": "1:59:58", "remaining_time": "14:54:50"}
+{"current_steps": 293, "total_steps": 2470, "loss": 0.4301, "lr": 3.995775420070094e-05, "epoch": 0.5931174089068826, "percentage": 11.86, "elapsed_time": "2:00:22", "remaining_time": "14:54:21"}
+{"current_steps": 294, "total_steps": 2470, "loss": 0.4389, "lr": 3.995589814455988e-05, "epoch": 0.5951417004048583, "percentage": 11.9, "elapsed_time": "2:00:46", "remaining_time": "14:53:56"}
+{"current_steps": 295, "total_steps": 2470, "loss": 0.4143, "lr": 3.995400223257335e-05, "epoch": 0.597165991902834, "percentage": 11.94, "elapsed_time": "2:01:11", "remaining_time": "14:53:31"}
+{"current_steps": 296, "total_steps": 2470, "loss": 0.4167, "lr": 3.9952066468527866e-05, "epoch": 0.5991902834008097, "percentage": 11.98, "elapsed_time": "2:01:36", "remaining_time": "14:53:09"}
+{"current_steps": 297, "total_steps": 2470, "loss": 0.4229, "lr": 3.995009085628951e-05, "epoch": 0.6012145748987854, "percentage": 12.02, "elapsed_time": "2:01:59", "remaining_time": "14:52:35"}
+{"current_steps": 298, "total_steps": 2470, "loss": 0.4363, "lr": 3.9948075399803986e-05, "epoch": 0.6032388663967612, "percentage": 12.06, "elapsed_time": "2:02:24", "remaining_time": "14:52:08"}
+{"current_steps": 299, "total_steps": 2470, "loss": 0.424, "lr": 3.994602010309655e-05, "epoch": 0.6052631578947368, "percentage": 12.11, "elapsed_time": "2:02:48", "remaining_time": "14:51:41"}
+{"current_steps": 300, "total_steps": 2470, "loss": 0.4225, "lr": 3.9943924970272034e-05, "epoch": 0.6072874493927125, "percentage": 12.15, "elapsed_time": "2:03:12", "remaining_time": "14:51:13"}
+{"current_steps": 301, "total_steps": 2470, "loss": 0.4325, "lr": 3.9941790005514825e-05, "epoch": 0.6093117408906883, "percentage": 12.19, "elapsed_time": "2:03:37", "remaining_time": "14:50:49"}
+{"current_steps": 302, "total_steps": 2470, "loss": 0.4327, "lr": 3.9939615213088865e-05, "epoch": 0.611336032388664, "percentage": 12.23, "elapsed_time": "2:04:01", "remaining_time": "14:50:17"}
+{"current_steps": 303, "total_steps": 2470, "loss": 0.4156, "lr": 3.993740059733766e-05, "epoch": 0.6133603238866396, "percentage": 12.27, "elapsed_time": "2:04:25", "remaining_time": "14:49:48"}
+{"current_steps": 304, "total_steps": 2470, "loss": 0.4236, "lr": 3.9935146162684206e-05, "epoch": 0.6153846153846154, "percentage": 12.31, "elapsed_time": "2:04:49", "remaining_time": "14:49:21"}
+{"current_steps": 305, "total_steps": 2470, "loss": 0.4314, "lr": 3.9932851913631066e-05, "epoch": 0.6174089068825911, "percentage": 12.35, "elapsed_time": "2:05:13", "remaining_time": "14:48:51"}
+{"current_steps": 306, "total_steps": 2470, "loss": 0.419, "lr": 3.99305178547603e-05, "epoch": 0.6194331983805668, "percentage": 12.39, "elapsed_time": "2:05:38", "remaining_time": "14:48:29"}
+{"current_steps": 307, "total_steps": 2470, "loss": 0.4218, "lr": 3.992814399073349e-05, "epoch": 0.6214574898785425, "percentage": 12.43, "elapsed_time": "2:06:02", "remaining_time": "14:48:04"}
+{"current_steps": 308, "total_steps": 2470, "loss": 0.4252, "lr": 3.99257303262917e-05, "epoch": 0.6234817813765182, "percentage": 12.47, "elapsed_time": "2:06:27", "remaining_time": "14:47:39"}
+{"current_steps": 309, "total_steps": 2470, "loss": 0.4394, "lr": 3.99232768662555e-05, "epoch": 0.6255060728744939, "percentage": 12.51, "elapsed_time": "2:06:51", "remaining_time": "14:47:13"}
+{"current_steps": 310, "total_steps": 2470, "loss": 0.4435, "lr": 3.9920783615524924e-05, "epoch": 0.6275303643724697, "percentage": 12.55, "elapsed_time": "2:07:16", "remaining_time": "14:46:49"}
+{"current_steps": 311, "total_steps": 2470, "loss": 0.4263, "lr": 3.9918250579079485e-05, "epoch": 0.6295546558704453, "percentage": 12.59, "elapsed_time": "2:07:41", "remaining_time": "14:46:25"}
+{"current_steps": 312, "total_steps": 2470, "loss": 0.4347, "lr": 3.991567776197815e-05, "epoch": 0.631578947368421, "percentage": 12.63, "elapsed_time": "2:08:06", "remaining_time": "14:46:02"}
+{"current_steps": 313, "total_steps": 2470, "loss": 0.4319, "lr": 3.991306516935935e-05, "epoch": 0.6336032388663968, "percentage": 12.67, "elapsed_time": "2:08:30", "remaining_time": "14:45:36"}
+{"current_steps": 314, "total_steps": 2470, "loss": 0.4233, "lr": 3.991041280644094e-05, "epoch": 0.6356275303643725, "percentage": 12.71, "elapsed_time": "2:08:55", "remaining_time": "14:45:15"}
+{"current_steps": 315, "total_steps": 2470, "loss": 0.4236, "lr": 3.990772067852021e-05, "epoch": 0.6376518218623481, "percentage": 12.75, "elapsed_time": "2:09:21", "remaining_time": "14:44:56"}
+{"current_steps": 316, "total_steps": 2470, "loss": 0.4387, "lr": 3.990498879097386e-05, "epoch": 0.6396761133603239, "percentage": 12.79, "elapsed_time": "2:09:45", "remaining_time": "14:44:31"}
+{"current_steps": 317, "total_steps": 2470, "loss": 0.4261, "lr": 3.990221714925802e-05, "epoch": 0.6417004048582996, "percentage": 12.83, "elapsed_time": "2:10:10", "remaining_time": "14:44:06"}
+{"current_steps": 318, "total_steps": 2470, "loss": 0.4142, "lr": 3.989940575890818e-05, "epoch": 0.6437246963562753, "percentage": 12.87, "elapsed_time": "2:10:35", "remaining_time": "14:43:45"}
+{"current_steps": 319, "total_steps": 2470, "loss": 0.4185, "lr": 3.989655462553926e-05, "epoch": 0.645748987854251, "percentage": 12.91, "elapsed_time": "2:11:00", "remaining_time": "14:43:23"}
+{"current_steps": 320, "total_steps": 2470, "loss": 0.4326, "lr": 3.989366375484553e-05, "epoch": 0.6477732793522267, "percentage": 12.96, "elapsed_time": "2:11:25", "remaining_time": "14:43:02"}
+{"current_steps": 321, "total_steps": 2470, "loss": 0.4328, "lr": 3.989073315260062e-05, "epoch": 0.6497975708502024, "percentage": 13.0, "elapsed_time": "2:11:50", "remaining_time": "14:42:39"}
+{"current_steps": 322, "total_steps": 2470, "loss": 0.4194, "lr": 3.988776282465752e-05, "epoch": 0.6518218623481782, "percentage": 13.04, "elapsed_time": "2:12:15", "remaining_time": "14:42:18"}
+{"current_steps": 323, "total_steps": 2470, "loss": 0.429, "lr": 3.9884752776948564e-05, "epoch": 0.6538461538461539, "percentage": 13.08, "elapsed_time": "2:12:40", "remaining_time": "14:41:56"}
+{"current_steps": 324, "total_steps": 2470, "loss": 0.4228, "lr": 3.98817030154854e-05, "epoch": 0.6558704453441295, "percentage": 13.12, "elapsed_time": "2:13:05", "remaining_time": "14:41:29"}
+{"current_steps": 325, "total_steps": 2470, "loss": 0.4368, "lr": 3.9878613546358996e-05, "epoch": 0.6578947368421053, "percentage": 13.16, "elapsed_time": "2:13:29", "remaining_time": "14:41:02"}
+{"current_steps": 326, "total_steps": 2470, "loss": 0.4308, "lr": 3.987548437573963e-05, "epoch": 0.659919028340081, "percentage": 13.2, "elapsed_time": "2:13:53", "remaining_time": "14:40:36"}
+{"current_steps": 327, "total_steps": 2470, "loss": 0.4143, "lr": 3.9872315509876885e-05, "epoch": 0.6619433198380567, "percentage": 13.24, "elapsed_time": "2:14:18", "remaining_time": "14:40:11"}
+{"current_steps": 328, "total_steps": 2470, "loss": 0.4278, "lr": 3.986910695509959e-05, "epoch": 0.6639676113360324, "percentage": 13.28, "elapsed_time": "2:14:43", "remaining_time": "14:39:49"}
+{"current_steps": 329, "total_steps": 2470, "loss": 0.4232, "lr": 3.9865858717815875e-05, "epoch": 0.6659919028340081, "percentage": 13.32, "elapsed_time": "2:15:08", "remaining_time": "14:39:26"}
+{"current_steps": 330, "total_steps": 2470, "loss": 0.4291, "lr": 3.986257080451309e-05, "epoch": 0.6680161943319838, "percentage": 13.36, "elapsed_time": "2:15:33", "remaining_time": "14:39:01"}
+{"current_steps": 331, "total_steps": 2470, "loss": 0.4197, "lr": 3.985924322175786e-05, "epoch": 0.6700404858299596, "percentage": 13.4, "elapsed_time": "2:15:57", "remaining_time": "14:38:37"}
+{"current_steps": 332, "total_steps": 2470, "loss": 0.4169, "lr": 3.9855875976196e-05, "epoch": 0.6720647773279352, "percentage": 13.44, "elapsed_time": "2:16:21", "remaining_time": "14:38:09"}
+{"current_steps": 333, "total_steps": 2470, "loss": 0.4326, "lr": 3.985246907455259e-05, "epoch": 0.6740890688259109, "percentage": 13.48, "elapsed_time": "2:16:47", "remaining_time": "14:37:51"}
+{"current_steps": 334, "total_steps": 2470, "loss": 0.4263, "lr": 3.9849022523631856e-05, "epoch": 0.6761133603238867, "percentage": 13.52, "elapsed_time": "2:17:11", "remaining_time": "14:37:24"}
+{"current_steps": 335, "total_steps": 2470, "loss": 0.4163, "lr": 3.9845536330317244e-05, "epoch": 0.6781376518218624, "percentage": 13.56, "elapsed_time": "2:17:36", "remaining_time": "14:37:00"}
+{"current_steps": 336, "total_steps": 2470, "loss": 0.4302, "lr": 3.984201050157136e-05, "epoch": 0.680161943319838, "percentage": 13.6, "elapsed_time": "2:18:01", "remaining_time": "14:36:36"}
+{"current_steps": 337, "total_steps": 2470, "loss": 0.4327, "lr": 3.9838445044435997e-05, "epoch": 0.6821862348178138, "percentage": 13.64, "elapsed_time": "2:18:25", "remaining_time": "14:36:10"}
+{"current_steps": 338, "total_steps": 2470, "loss": 0.4272, "lr": 3.983483996603205e-05, "epoch": 0.6842105263157895, "percentage": 13.68, "elapsed_time": "2:18:49", "remaining_time": "14:35:41"}
+{"current_steps": 339, "total_steps": 2470, "loss": 0.4233, "lr": 3.9831195273559575e-05, "epoch": 0.6862348178137652, "percentage": 13.72, "elapsed_time": "2:19:14", "remaining_time": "14:35:14"}
+{"current_steps": 340, "total_steps": 2470, "loss": 0.4169, "lr": 3.982751097429775e-05, "epoch": 0.6882591093117408, "percentage": 13.77, "elapsed_time": "2:19:38", "remaining_time": "14:34:49"}
+{"current_steps": 341, "total_steps": 2470, "loss": 0.4239, "lr": 3.982378707560483e-05, "epoch": 0.6902834008097166, "percentage": 13.81, "elapsed_time": "2:20:02", "remaining_time": "14:34:21"}
+{"current_steps": 342, "total_steps": 2470, "loss": 0.415, "lr": 3.982002358491817e-05, "epoch": 0.6923076923076923, "percentage": 13.85, "elapsed_time": "2:20:27", "remaining_time": "14:33:55"}
+{"current_steps": 343, "total_steps": 2470, "loss": 0.4265, "lr": 3.9816220509754214e-05, "epoch": 0.694331983805668, "percentage": 13.89, "elapsed_time": "2:20:52", "remaining_time": "14:33:36"}
+{"current_steps": 344, "total_steps": 2470, "loss": 0.4201, "lr": 3.981237785770844e-05, "epoch": 0.6963562753036437, "percentage": 13.93, "elapsed_time": "2:21:17", "remaining_time": "14:33:13"}
+{"current_steps": 345, "total_steps": 2470, "loss": 0.4127, "lr": 3.980849563645537e-05, "epoch": 0.6983805668016194, "percentage": 13.97, "elapsed_time": "2:21:41", "remaining_time": "14:32:42"}
+{"current_steps": 346, "total_steps": 2470, "loss": 0.4137, "lr": 3.980457385374858e-05, "epoch": 0.7004048582995951, "percentage": 14.01, "elapsed_time": "2:22:05", "remaining_time": "14:32:17"}
+{"current_steps": 347, "total_steps": 2470, "loss": 0.4218, "lr": 3.9800612517420626e-05, "epoch": 0.7024291497975709, "percentage": 14.05, "elapsed_time": "2:22:30", "remaining_time": "14:31:51"}
+{"current_steps": 348, "total_steps": 2470, "loss": 0.4267, "lr": 3.9796611635383084e-05, "epoch": 0.7044534412955465, "percentage": 14.09, "elapsed_time": "2:22:55", "remaining_time": "14:31:27"}
+{"current_steps": 349, "total_steps": 2470, "loss": 0.4157, "lr": 3.979257121562649e-05, "epoch": 0.7064777327935222, "percentage": 14.13, "elapsed_time": "2:23:19", "remaining_time": "14:31:04"}
+{"current_steps": 350, "total_steps": 2470, "loss": 0.4354, "lr": 3.9788491266220364e-05, "epoch": 0.708502024291498, "percentage": 14.17, "elapsed_time": "2:23:44", "remaining_time": "14:30:39"}
+{"current_steps": 351, "total_steps": 2470, "loss": 0.4201, "lr": 3.978437179531316e-05, "epoch": 0.7105263157894737, "percentage": 14.21, "elapsed_time": "2:24:09", "remaining_time": "14:30:15"}
+{"current_steps": 352, "total_steps": 2470, "loss": 0.4173, "lr": 3.978021281113228e-05, "epoch": 0.7125506072874493, "percentage": 14.25, "elapsed_time": "2:24:33", "remaining_time": "14:29:47"}
+{"current_steps": 353, "total_steps": 2470, "loss": 0.422, "lr": 3.9776014321984015e-05, "epoch": 0.7145748987854251, "percentage": 14.29, "elapsed_time": "2:24:57", "remaining_time": "14:29:20"}
+{"current_steps": 354, "total_steps": 2470, "loss": 0.4236, "lr": 3.9771776336253586e-05, "epoch": 0.7165991902834008, "percentage": 14.33, "elapsed_time": "2:25:21", "remaining_time": "14:28:50"}
+{"current_steps": 355, "total_steps": 2470, "loss": 0.4362, "lr": 3.976749886240508e-05, "epoch": 0.7186234817813765, "percentage": 14.37, "elapsed_time": "2:25:45", "remaining_time": "14:28:25"}
+{"current_steps": 356, "total_steps": 2470, "loss": 0.4359, "lr": 3.976318190898145e-05, "epoch": 0.7206477732793523, "percentage": 14.41, "elapsed_time": "2:26:10", "remaining_time": "14:28:00"}
+{"current_steps": 357, "total_steps": 2470, "loss": 0.4288, "lr": 3.9758825484604494e-05, "epoch": 0.7226720647773279, "percentage": 14.45, "elapsed_time": "2:26:35", "remaining_time": "14:27:37"}
+{"current_steps": 358, "total_steps": 2470, "loss": 0.4231, "lr": 3.9754429597974855e-05, "epoch": 0.7246963562753036, "percentage": 14.49, "elapsed_time": "2:27:00", "remaining_time": "14:27:14"}
+{"current_steps": 359, "total_steps": 2470, "loss": 0.4242, "lr": 3.974999425787197e-05, "epoch": 0.7267206477732794, "percentage": 14.53, "elapsed_time": "2:27:24", "remaining_time": "14:26:45"}
+{"current_steps": 360, "total_steps": 2470, "loss": 0.4324, "lr": 3.97455194731541e-05, "epoch": 0.728744939271255, "percentage": 14.57, "elapsed_time": "2:27:49", "remaining_time": "14:26:22"}
+{"current_steps": 361, "total_steps": 2470, "loss": 0.4146, "lr": 3.9741005252758255e-05, "epoch": 0.7307692307692307, "percentage": 14.62, "elapsed_time": "2:28:13", "remaining_time": "14:25:56"}
+{"current_steps": 362, "total_steps": 2470, "loss": 0.4382, "lr": 3.973645160570023e-05, "epoch": 0.7327935222672065, "percentage": 14.66, "elapsed_time": "2:28:37", "remaining_time": "14:25:26"}
+{"current_steps": 363, "total_steps": 2470, "loss": 0.4239, "lr": 3.973185854107455e-05, "epoch": 0.7348178137651822, "percentage": 14.7, "elapsed_time": "2:29:00", "remaining_time": "14:24:56"}
+{"current_steps": 364, "total_steps": 2470, "loss": 0.414, "lr": 3.972722606805445e-05, "epoch": 0.7368421052631579, "percentage": 14.74, "elapsed_time": "2:29:25", "remaining_time": "14:24:30"}
+{"current_steps": 365, "total_steps": 2470, "loss": 0.4188, "lr": 3.9722554195891906e-05, "epoch": 0.7388663967611336, "percentage": 14.78, "elapsed_time": "2:29:49", "remaining_time": "14:24:00"}
+{"current_steps": 366, "total_steps": 2470, "loss": 0.4193, "lr": 3.971784293391756e-05, "epoch": 0.7408906882591093, "percentage": 14.82, "elapsed_time": "2:30:13", "remaining_time": "14:23:36"}
+{"current_steps": 367, "total_steps": 2470, "loss": 0.4371, "lr": 3.971309229154072e-05, "epoch": 0.742914979757085, "percentage": 14.86, "elapsed_time": "2:30:38", "remaining_time": "14:23:12"}
+{"current_steps": 368, "total_steps": 2470, "loss": 0.4108, "lr": 3.970830227824937e-05, "epoch": 0.7449392712550608, "percentage": 14.9, "elapsed_time": "2:31:03", "remaining_time": "14:22:48"}
+{"current_steps": 369, "total_steps": 2470, "loss": 0.4177, "lr": 3.970347290361007e-05, "epoch": 0.7469635627530364, "percentage": 14.94, "elapsed_time": "2:31:27", "remaining_time": "14:22:24"}
+{"current_steps": 370, "total_steps": 2470, "loss": 0.4224, "lr": 3.969860417726807e-05, "epoch": 0.7489878542510121, "percentage": 14.98, "elapsed_time": "2:31:51", "remaining_time": "14:21:53"}
+{"current_steps": 371, "total_steps": 2470, "loss": 0.405, "lr": 3.9693696108947144e-05, "epoch": 0.7510121457489879, "percentage": 15.02, "elapsed_time": "2:32:15", "remaining_time": "14:21:28"}
+{"current_steps": 372, "total_steps": 2470, "loss": 0.4094, "lr": 3.9688748708449686e-05, "epoch": 0.7530364372469636, "percentage": 15.06, "elapsed_time": "2:32:40", "remaining_time": "14:21:04"}
+{"current_steps": 373, "total_steps": 2470, "loss": 0.4144, "lr": 3.968376198565661e-05, "epoch": 0.7550607287449392, "percentage": 15.1, "elapsed_time": "2:33:04", "remaining_time": "14:20:36"}
+{"current_steps": 374, "total_steps": 2470, "loss": 0.4175, "lr": 3.9678735950527396e-05, "epoch": 0.757085020242915, "percentage": 15.14, "elapsed_time": "2:33:28", "remaining_time": "14:20:06"}
+{"current_steps": 375, "total_steps": 2470, "loss": 0.4094, "lr": 3.9673670613100015e-05, "epoch": 0.7591093117408907, "percentage": 15.18, "elapsed_time": "2:33:52", "remaining_time": "14:19:41"}
+{"current_steps": 376, "total_steps": 2470, "loss": 0.4221, "lr": 3.966856598349093e-05, "epoch": 0.7611336032388664, "percentage": 15.22, "elapsed_time": "2:34:17", "remaining_time": "14:19:14"}
+{"current_steps": 377, "total_steps": 2470, "loss": 0.4273, "lr": 3.9663422071895103e-05, "epoch": 0.7631578947368421, "percentage": 15.26, "elapsed_time": "2:34:41", "remaining_time": "14:18:49"}
+{"current_steps": 378, "total_steps": 2470, "loss": 0.4131, "lr": 3.9658238888585934e-05, "epoch": 0.7651821862348178, "percentage": 15.3, "elapsed_time": "2:35:05", "remaining_time": "14:18:19"}
+{"current_steps": 379, "total_steps": 2470, "loss": 0.424, "lr": 3.965301644391525e-05, "epoch": 0.7672064777327935, "percentage": 15.34, "elapsed_time": "2:35:29", "remaining_time": "14:17:54"}
+{"current_steps": 380, "total_steps": 2470, "loss": 0.4208, "lr": 3.9647754748313294e-05, "epoch": 0.7692307692307693, "percentage": 15.38, "elapsed_time": "2:35:54", "remaining_time": "14:17:30"}
+{"current_steps": 381, "total_steps": 2470, "loss": 0.4236, "lr": 3.964245381228872e-05, "epoch": 0.771255060728745, "percentage": 15.43, "elapsed_time": "2:36:19", "remaining_time": "14:17:07"}
+{"current_steps": 382, "total_steps": 2470, "loss": 0.4208, "lr": 3.963711364642854e-05, "epoch": 0.7732793522267206, "percentage": 15.47, "elapsed_time": "2:36:44", "remaining_time": "14:16:42"}
+{"current_steps": 383, "total_steps": 2470, "loss": 0.4223, "lr": 3.963173426139808e-05, "epoch": 0.7753036437246964, "percentage": 15.51, "elapsed_time": "2:37:08", "remaining_time": "14:16:14"}
+{"current_steps": 384, "total_steps": 2470, "loss": 0.4308, "lr": 3.962631566794107e-05, "epoch": 0.7773279352226721, "percentage": 15.55, "elapsed_time": "2:37:31", "remaining_time": "14:15:45"}
+{"current_steps": 385, "total_steps": 2470, "loss": 0.4316, "lr": 3.962085787687948e-05, "epoch": 0.7793522267206477, "percentage": 15.59, "elapsed_time": "2:37:56", "remaining_time": "14:15:20"}
+{"current_steps": 386, "total_steps": 2470, "loss": 0.4149, "lr": 3.96153608991136e-05, "epoch": 0.7813765182186235, "percentage": 15.63, "elapsed_time": "2:38:21", "remaining_time": "14:15:00"}
+{"current_steps": 387, "total_steps": 2470, "loss": 0.4132, "lr": 3.960982474562196e-05, "epoch": 0.7834008097165992, "percentage": 15.67, "elapsed_time": "2:38:46", "remaining_time": "14:14:35"}
+{"current_steps": 388, "total_steps": 2470, "loss": 0.4001, "lr": 3.960424942746136e-05, "epoch": 0.7854251012145749, "percentage": 15.71, "elapsed_time": "2:39:10", "remaining_time": "14:14:07"}
+{"current_steps": 389, "total_steps": 2470, "loss": 0.4195, "lr": 3.95986349557668e-05, "epoch": 0.7874493927125507, "percentage": 15.75, "elapsed_time": "2:39:35", "remaining_time": "14:13:43"}
+{"current_steps": 390, "total_steps": 2470, "loss": 0.4113, "lr": 3.959298134175148e-05, "epoch": 0.7894736842105263, "percentage": 15.79, "elapsed_time": "2:40:00", "remaining_time": "14:13:20"}
+{"current_steps": 391, "total_steps": 2470, "loss": 0.4347, "lr": 3.958728859670678e-05, "epoch": 0.791497975708502, "percentage": 15.83, "elapsed_time": "2:40:25", "remaining_time": "14:12:59"}
+{"current_steps": 392, "total_steps": 2470, "loss": 0.4201, "lr": 3.958155673200223e-05, "epoch": 0.7935222672064778, "percentage": 15.87, "elapsed_time": "2:40:50", "remaining_time": "14:12:35"}
+{"current_steps": 393, "total_steps": 2470, "loss": 0.4125, "lr": 3.9575785759085476e-05, "epoch": 0.7955465587044535, "percentage": 15.91, "elapsed_time": "2:41:14", "remaining_time": "14:12:11"}
+{"current_steps": 394, "total_steps": 2470, "loss": 0.4246, "lr": 3.95699756894823e-05, "epoch": 0.7975708502024291, "percentage": 15.95, "elapsed_time": "2:41:39", "remaining_time": "14:11:44"}
+{"current_steps": 395, "total_steps": 2470, "loss": 0.4098, "lr": 3.956412653479654e-05, "epoch": 0.7995951417004049, "percentage": 15.99, "elapsed_time": "2:42:02", "remaining_time": "14:11:13"}
+{"current_steps": 396, "total_steps": 2470, "loss": 0.4141, "lr": 3.955823830671012e-05, "epoch": 0.8016194331983806, "percentage": 16.03, "elapsed_time": "2:42:27", "remaining_time": "14:10:49"}
+{"current_steps": 397, "total_steps": 2470, "loss": 0.4068, "lr": 3.9552311016982964e-05, "epoch": 0.8036437246963563, "percentage": 16.07, "elapsed_time": "2:42:52", "remaining_time": "14:10:26"}
+{"current_steps": 398, "total_steps": 2470, "loss": 0.4086, "lr": 3.9546344677453067e-05, "epoch": 0.805668016194332, "percentage": 16.11, "elapsed_time": "2:43:16", "remaining_time": "14:10:01"}
+{"current_steps": 399, "total_steps": 2470, "loss": 0.4241, "lr": 3.954033930003634e-05, "epoch": 0.8076923076923077, "percentage": 16.15, "elapsed_time": "2:43:41", "remaining_time": "14:09:37"}
+{"current_steps": 400, "total_steps": 2470, "loss": 0.4161, "lr": 3.953429489672673e-05, "epoch": 0.8097165991902834, "percentage": 16.19, "elapsed_time": "2:44:06", "remaining_time": "14:09:13"}
+{"current_steps": 401, "total_steps": 2470, "loss": 0.4156, "lr": 3.9528211479596095e-05, "epoch": 0.8117408906882592, "percentage": 16.23, "elapsed_time": "2:44:30", "remaining_time": "14:08:46"}
+{"current_steps": 402, "total_steps": 2470, "loss": 0.4134, "lr": 3.952208906079419e-05, "epoch": 0.8137651821862348, "percentage": 16.28, "elapsed_time": "2:44:54", "remaining_time": "14:08:19"}
+{"current_steps": 403, "total_steps": 2470, "loss": 0.4088, "lr": 3.9515927652548714e-05, "epoch": 0.8157894736842105, "percentage": 16.32, "elapsed_time": "2:45:19", "remaining_time": "14:07:55"}
+{"current_steps": 404, "total_steps": 2470, "loss": 0.4202, "lr": 3.9509727267165196e-05, "epoch": 0.8178137651821862, "percentage": 16.36, "elapsed_time": "2:45:42", "remaining_time": "14:07:26"}
+{"current_steps": 405, "total_steps": 2470, "loss": 0.4127, "lr": 3.950348791702702e-05, "epoch": 0.819838056680162, "percentage": 16.4, "elapsed_time": "2:46:07", "remaining_time": "14:07:00"}
+{"current_steps": 406, "total_steps": 2470, "loss": 0.4318, "lr": 3.94972096145954e-05, "epoch": 0.8218623481781376, "percentage": 16.44, "elapsed_time": "2:46:31", "remaining_time": "14:06:33"}
+{"current_steps": 407, "total_steps": 2470, "loss": 0.412, "lr": 3.949089237240933e-05, "epoch": 0.8238866396761133, "percentage": 16.48, "elapsed_time": "2:46:56", "remaining_time": "14:06:10"}
+{"current_steps": 408, "total_steps": 2470, "loss": 0.4085, "lr": 3.948453620308559e-05, "epoch": 0.8259109311740891, "percentage": 16.52, "elapsed_time": "2:47:20", "remaining_time": "14:05:41"}
+{"current_steps": 409, "total_steps": 2470, "loss": 0.419, "lr": 3.9478141119318695e-05, "epoch": 0.8279352226720648, "percentage": 16.56, "elapsed_time": "2:47:44", "remaining_time": "14:05:18"}
+{"current_steps": 410, "total_steps": 2470, "loss": 0.4125, "lr": 3.9471707133880885e-05, "epoch": 0.8299595141700404, "percentage": 16.6, "elapsed_time": "2:48:09", "remaining_time": "14:04:55"}
+{"current_steps": 411, "total_steps": 2470, "loss": 0.4161, "lr": 3.946523425962209e-05, "epoch": 0.8319838056680162, "percentage": 16.64, "elapsed_time": "2:48:34", "remaining_time": "14:04:29"}
+{"current_steps": 412, "total_steps": 2470, "loss": 0.4173, "lr": 3.94587225094699e-05, "epoch": 0.8340080971659919, "percentage": 16.68, "elapsed_time": "2:48:59", "remaining_time": "14:04:06"}
+{"current_steps": 413, "total_steps": 2470, "loss": 0.4109, "lr": 3.945217189642958e-05, "epoch": 0.8360323886639676, "percentage": 16.72, "elapsed_time": "2:49:24", "remaining_time": "14:03:43"}
+{"current_steps": 414, "total_steps": 2470, "loss": 0.4108, "lr": 3.944558243358397e-05, "epoch": 0.8380566801619433, "percentage": 16.76, "elapsed_time": "2:49:49", "remaining_time": "14:03:20"}
+{"current_steps": 415, "total_steps": 2470, "loss": 0.3969, "lr": 3.9438954134093536e-05, "epoch": 0.840080971659919, "percentage": 16.8, "elapsed_time": "2:50:13", "remaining_time": "14:02:55"}
+{"current_steps": 416, "total_steps": 2470, "loss": 0.4174, "lr": 3.943228701119628e-05, "epoch": 0.8421052631578947, "percentage": 16.84, "elapsed_time": "2:50:37", "remaining_time": "14:02:28"}
+{"current_steps": 417, "total_steps": 2470, "loss": 0.4176, "lr": 3.9425581078207764e-05, "epoch": 0.8441295546558705, "percentage": 16.88, "elapsed_time": "2:51:02", "remaining_time": "14:02:04"}
+{"current_steps": 418, "total_steps": 2470, "loss": 0.4065, "lr": 3.9418836348521045e-05, "epoch": 0.8461538461538461, "percentage": 16.92, "elapsed_time": "2:51:27", "remaining_time": "14:01:41"}
+{"current_steps": 419, "total_steps": 2470, "loss": 0.4211, "lr": 3.941205283560668e-05, "epoch": 0.8481781376518218, "percentage": 16.96, "elapsed_time": "2:51:51", "remaining_time": "14:01:16"}
+{"current_steps": 420, "total_steps": 2470, "loss": 0.4254, "lr": 3.940523055301267e-05, "epoch": 0.8502024291497976, "percentage": 17.0, "elapsed_time": "2:52:16", "remaining_time": "14:00:49"}
+{"current_steps": 421, "total_steps": 2470, "loss": 0.4245, "lr": 3.939836951436445e-05, "epoch": 0.8522267206477733, "percentage": 17.04, "elapsed_time": "2:52:41", "remaining_time": "14:00:29"}
+{"current_steps": 422, "total_steps": 2470, "loss": 0.4232, "lr": 3.939146973336487e-05, "epoch": 0.854251012145749, "percentage": 17.09, "elapsed_time": "2:53:06", "remaining_time": "14:00:06"}
+{"current_steps": 423, "total_steps": 2470, "loss": 0.407, "lr": 3.9384531223794135e-05, "epoch": 0.8562753036437247, "percentage": 17.13, "elapsed_time": "2:53:30", "remaining_time": "13:59:40"}
+{"current_steps": 424, "total_steps": 2470, "loss": 0.4211, "lr": 3.937755399950983e-05, "epoch": 0.8582995951417004, "percentage": 17.17, "elapsed_time": "2:53:55", "remaining_time": "13:59:15"}
+{"current_steps": 425, "total_steps": 2470, "loss": 0.4171, "lr": 3.937053807444682e-05, "epoch": 0.8603238866396761, "percentage": 17.21, "elapsed_time": "2:54:20", "remaining_time": "13:58:55"}
+{"current_steps": 426, "total_steps": 2470, "loss": 0.417, "lr": 3.9363483462617294e-05, "epoch": 0.8623481781376519, "percentage": 17.25, "elapsed_time": "2:54:45", "remaining_time": "13:58:30"}
+{"current_steps": 427, "total_steps": 2470, "loss": 0.4138, "lr": 3.9356390178110694e-05, "epoch": 0.8643724696356275, "percentage": 17.29, "elapsed_time": "2:55:10", "remaining_time": "13:58:07"}
+{"current_steps": 428, "total_steps": 2470, "loss": 0.4226, "lr": 3.934925823509371e-05, "epoch": 0.8663967611336032, "percentage": 17.33, "elapsed_time": "2:55:34", "remaining_time": "13:57:41"}
+{"current_steps": 429, "total_steps": 2470, "loss": 0.4101, "lr": 3.934208764781022e-05, "epoch": 0.868421052631579, "percentage": 17.37, "elapsed_time": "2:56:00", "remaining_time": "13:57:21"}
+{"current_steps": 430, "total_steps": 2470, "loss": 0.4199, "lr": 3.93348784305813e-05, "epoch": 0.8704453441295547, "percentage": 17.41, "elapsed_time": "2:56:24", "remaining_time": "13:56:52"}
+{"current_steps": 431, "total_steps": 2470, "loss": 0.398, "lr": 3.9327630597805166e-05, "epoch": 0.8724696356275303, "percentage": 17.45, "elapsed_time": "2:56:48", "remaining_time": "13:56:26"}
+{"current_steps": 432, "total_steps": 2470, "loss": 0.4079, "lr": 3.9320344163957177e-05, "epoch": 0.8744939271255061, "percentage": 17.49, "elapsed_time": "2:57:12", "remaining_time": "13:55:59"}
+{"current_steps": 433, "total_steps": 2470, "loss": 0.4265, "lr": 3.931301914358976e-05, "epoch": 0.8765182186234818, "percentage": 17.53, "elapsed_time": "2:57:37", "remaining_time": "13:55:35"}
+{"current_steps": 434, "total_steps": 2470, "loss": 0.4243, "lr": 3.930565555133241e-05, "epoch": 0.8785425101214575, "percentage": 17.57, "elapsed_time": "2:58:01", "remaining_time": "13:55:11"}
+{"current_steps": 435, "total_steps": 2470, "loss": 0.4077, "lr": 3.929825340189169e-05, "epoch": 0.8805668016194332, "percentage": 17.61, "elapsed_time": "2:58:26", "remaining_time": "13:54:45"}
+{"current_steps": 436, "total_steps": 2470, "loss": 0.4097, "lr": 3.929081271005113e-05, "epoch": 0.8825910931174089, "percentage": 17.65, "elapsed_time": "2:58:51", "remaining_time": "13:54:22"}
+{"current_steps": 437, "total_steps": 2470, "loss": 0.4085, "lr": 3.928333349067125e-05, "epoch": 0.8846153846153846, "percentage": 17.69, "elapsed_time": "2:59:15", "remaining_time": "13:53:58"}
+{"current_steps": 438, "total_steps": 2470, "loss": 0.3928, "lr": 3.927581575868953e-05, "epoch": 0.8866396761133604, "percentage": 17.73, "elapsed_time": "2:59:40", "remaining_time": "13:53:31"}
+{"current_steps": 439, "total_steps": 2470, "loss": 0.4155, "lr": 3.926825952912034e-05, "epoch": 0.888663967611336, "percentage": 17.77, "elapsed_time": "3:00:03", "remaining_time": "13:53:03"}
+{"current_steps": 440, "total_steps": 2470, "loss": 0.4173, "lr": 3.926066481705497e-05, "epoch": 0.8906882591093117, "percentage": 17.81, "elapsed_time": "3:00:28", "remaining_time": "13:52:37"}
+{"current_steps": 441, "total_steps": 2470, "loss": 0.4163, "lr": 3.925303163766154e-05, "epoch": 0.8927125506072875, "percentage": 17.85, "elapsed_time": "3:00:52", "remaining_time": "13:52:09"}
+{"current_steps": 442, "total_steps": 2470, "loss": 0.4156, "lr": 3.924536000618501e-05, "epoch": 0.8947368421052632, "percentage": 17.89, "elapsed_time": "3:01:16", "remaining_time": "13:51:41"}
+{"current_steps": 443, "total_steps": 2470, "loss": 0.4085, "lr": 3.923764993794714e-05, "epoch": 0.8967611336032388, "percentage": 17.94, "elapsed_time": "3:01:40", "remaining_time": "13:51:17"}
+{"current_steps": 444, "total_steps": 2470, "loss": 0.4105, "lr": 3.922990144834644e-05, "epoch": 0.8987854251012146, "percentage": 17.98, "elapsed_time": "3:02:04", "remaining_time": "13:50:51"}
+{"current_steps": 445, "total_steps": 2470, "loss": 0.4144, "lr": 3.922211455285817e-05, "epoch": 0.9008097165991903, "percentage": 18.02, "elapsed_time": "3:02:28", "remaining_time": "13:50:24"}
+{"current_steps": 446, "total_steps": 2470, "loss": 0.4185, "lr": 3.921428926703429e-05, "epoch": 0.902834008097166, "percentage": 18.06, "elapsed_time": "3:02:53", "remaining_time": "13:49:57"}
+{"current_steps": 447, "total_steps": 2470, "loss": 0.4139, "lr": 3.920642560650343e-05, "epoch": 0.9048582995951417, "percentage": 18.1, "elapsed_time": "3:03:17", "remaining_time": "13:49:32"}
+{"current_steps": 448, "total_steps": 2470, "loss": 0.4056, "lr": 3.9198523586970865e-05, "epoch": 0.9068825910931174, "percentage": 18.14, "elapsed_time": "3:03:41", "remaining_time": "13:49:04"}
+{"current_steps": 449, "total_steps": 2470, "loss": 0.409, "lr": 3.919058322421847e-05, "epoch": 0.9089068825910931, "percentage": 18.18, "elapsed_time": "3:04:04", "remaining_time": "13:48:34"}
+{"current_steps": 450, "total_steps": 2470, "loss": 0.4211, "lr": 3.918260453410472e-05, "epoch": 0.9109311740890689, "percentage": 18.22, "elapsed_time": "3:04:29", "remaining_time": "13:48:10"}
+{"current_steps": 451, "total_steps": 2470, "loss": 0.4181, "lr": 3.917458753256462e-05, "epoch": 0.9129554655870445, "percentage": 18.26, "elapsed_time": "3:04:54", "remaining_time": "13:47:44"}
+{"current_steps": 452, "total_steps": 2470, "loss": 0.4159, "lr": 3.9166532235609695e-05, "epoch": 0.9149797570850202, "percentage": 18.3, "elapsed_time": "3:05:19", "remaining_time": "13:47:22"}
+{"current_steps": 453, "total_steps": 2470, "loss": 0.415, "lr": 3.915843865932795e-05, "epoch": 0.917004048582996, "percentage": 18.34, "elapsed_time": "3:05:44", "remaining_time": "13:47:02"}
+{"current_steps": 454, "total_steps": 2470, "loss": 0.4228, "lr": 3.915030681988386e-05, "epoch": 0.9190283400809717, "percentage": 18.38, "elapsed_time": "3:06:08", "remaining_time": "13:46:34"}
+{"current_steps": 455, "total_steps": 2470, "loss": 0.4015, "lr": 3.9142136733518285e-05, "epoch": 0.9210526315789473, "percentage": 18.42, "elapsed_time": "3:06:32", "remaining_time": "13:46:07"}
+{"current_steps": 456, "total_steps": 2470, "loss": 0.4173, "lr": 3.913392841654851e-05, "epoch": 0.9230769230769231, "percentage": 18.46, "elapsed_time": "3:06:56", "remaining_time": "13:45:39"}
+{"current_steps": 457, "total_steps": 2470, "loss": 0.4149, "lr": 3.912568188536814e-05, "epoch": 0.9251012145748988, "percentage": 18.5, "elapsed_time": "3:07:21", "remaining_time": "13:45:16"}
+{"current_steps": 458, "total_steps": 2470, "loss": 0.4067, "lr": 3.9117397156447125e-05, "epoch": 0.9271255060728745, "percentage": 18.54, "elapsed_time": "3:07:44", "remaining_time": "13:44:44"}
+{"current_steps": 459, "total_steps": 2470, "loss": 0.4143, "lr": 3.9109074246331686e-05, "epoch": 0.9291497975708503, "percentage": 18.58, "elapsed_time": "3:08:09", "remaining_time": "13:44:21"}
+{"current_steps": 460, "total_steps": 2470, "loss": 0.4106, "lr": 3.910071317164432e-05, "epoch": 0.9311740890688259, "percentage": 18.62, "elapsed_time": "3:08:33", "remaining_time": "13:43:56"}
+{"current_steps": 461, "total_steps": 2470, "loss": 0.4154, "lr": 3.909231394908373e-05, "epoch": 0.9331983805668016, "percentage": 18.66, "elapsed_time": "3:08:58", "remaining_time": "13:43:30"}
+{"current_steps": 462, "total_steps": 2470, "loss": 0.4271, "lr": 3.908387659542481e-05, "epoch": 0.9352226720647774, "percentage": 18.7, "elapsed_time": "3:09:22", "remaining_time": "13:43:04"}
+{"current_steps": 463, "total_steps": 2470, "loss": 0.409, "lr": 3.9075401127518606e-05, "epoch": 0.937246963562753, "percentage": 18.74, "elapsed_time": "3:09:46", "remaining_time": "13:42:39"}
+{"current_steps": 464, "total_steps": 2470, "loss": 0.4329, "lr": 3.90668875622923e-05, "epoch": 0.9392712550607287, "percentage": 18.79, "elapsed_time": "3:10:11", "remaining_time": "13:42:17"}
+{"current_steps": 465, "total_steps": 2470, "loss": 0.4109, "lr": 3.9058335916749154e-05, "epoch": 0.9412955465587044, "percentage": 18.83, "elapsed_time": "3:10:37", "remaining_time": "13:41:58"}
+{"current_steps": 466, "total_steps": 2470, "loss": 0.4133, "lr": 3.904974620796847e-05, "epoch": 0.9433198380566802, "percentage": 18.87, "elapsed_time": "3:11:02", "remaining_time": "13:41:34"}
+{"current_steps": 467, "total_steps": 2470, "loss": 0.4075, "lr": 3.904111845310559e-05, "epoch": 0.9453441295546559, "percentage": 18.91, "elapsed_time": "3:11:27", "remaining_time": "13:41:10"}
+{"current_steps": 468, "total_steps": 2470, "loss": 0.4125, "lr": 3.903245266939184e-05, "epoch": 0.9473684210526315, "percentage": 18.95, "elapsed_time": "3:11:51", "remaining_time": "13:40:43"}
+{"current_steps": 469, "total_steps": 2470, "loss": 0.4124, "lr": 3.9023748874134486e-05, "epoch": 0.9493927125506073, "percentage": 18.99, "elapsed_time": "3:12:16", "remaining_time": "13:40:19"}
+{"current_steps": 470, "total_steps": 2470, "loss": 0.4136, "lr": 3.901500708471671e-05, "epoch": 0.951417004048583, "percentage": 19.03, "elapsed_time": "3:12:39", "remaining_time": "13:39:51"}
+{"current_steps": 471, "total_steps": 2470, "loss": 0.4202, "lr": 3.900622731859759e-05, "epoch": 0.9534412955465587, "percentage": 19.07, "elapsed_time": "3:13:04", "remaining_time": "13:39:27"}
+{"current_steps": 472, "total_steps": 2470, "loss": 0.4139, "lr": 3.899740959331204e-05, "epoch": 0.9554655870445344, "percentage": 19.11, "elapsed_time": "3:13:29", "remaining_time": "13:39:01"}
+{"current_steps": 473, "total_steps": 2470, "loss": 0.3991, "lr": 3.898855392647078e-05, "epoch": 0.9574898785425101, "percentage": 19.15, "elapsed_time": "3:13:53", "remaining_time": "13:38:37"}
+{"current_steps": 474, "total_steps": 2470, "loss": 0.4148, "lr": 3.897966033576032e-05, "epoch": 0.9595141700404858, "percentage": 19.19, "elapsed_time": "3:14:18", "remaining_time": "13:38:13"}
+{"current_steps": 475, "total_steps": 2470, "loss": 0.4061, "lr": 3.897072883894291e-05, "epoch": 0.9615384615384616, "percentage": 19.23, "elapsed_time": "3:14:43", "remaining_time": "13:37:49"}
+{"current_steps": 476, "total_steps": 2470, "loss": 0.4086, "lr": 3.896175945385651e-05, "epoch": 0.9635627530364372, "percentage": 19.27, "elapsed_time": "3:15:07", "remaining_time": "13:37:24"}
+{"current_steps": 477, "total_steps": 2470, "loss": 0.41, "lr": 3.8952752198414716e-05, "epoch": 0.9655870445344129, "percentage": 19.31, "elapsed_time": "3:15:32", "remaining_time": "13:36:59"}
+{"current_steps": 478, "total_steps": 2470, "loss": 0.4305, "lr": 3.894370709060681e-05, "epoch": 0.9676113360323887, "percentage": 19.35, "elapsed_time": "3:15:57", "remaining_time": "13:36:36"}
+{"current_steps": 479, "total_steps": 2470, "loss": 0.412, "lr": 3.893462414849764e-05, "epoch": 0.9696356275303644, "percentage": 19.39, "elapsed_time": "3:16:22", "remaining_time": "13:36:13"}
+{"current_steps": 480, "total_steps": 2470, "loss": 0.4256, "lr": 3.892550339022762e-05, "epoch": 0.97165991902834, "percentage": 19.43, "elapsed_time": "3:16:46", "remaining_time": "13:35:48"}
+{"current_steps": 481, "total_steps": 2470, "loss": 0.4118, "lr": 3.8916344834012695e-05, "epoch": 0.9736842105263158, "percentage": 19.47, "elapsed_time": "3:17:10", "remaining_time": "13:35:18"}
+{"current_steps": 482, "total_steps": 2470, "loss": 0.4069, "lr": 3.890714849814431e-05, "epoch": 0.9757085020242915, "percentage": 19.51, "elapsed_time": "3:17:34", "remaining_time": "13:34:54"}
+{"current_steps": 483, "total_steps": 2470, "loss": 0.4017, "lr": 3.889791440098933e-05, "epoch": 0.9777327935222672, "percentage": 19.55, "elapsed_time": "3:17:59", "remaining_time": "13:34:29"}
+{"current_steps": 484, "total_steps": 2470, "loss": 0.4132, "lr": 3.888864256099009e-05, "epoch": 0.979757085020243, "percentage": 19.6, "elapsed_time": "3:18:22", "remaining_time": "13:33:59"}
+{"current_steps": 485, "total_steps": 2470, "loss": 0.4156, "lr": 3.8879332996664245e-05, "epoch": 0.9817813765182186, "percentage": 19.64, "elapsed_time": "3:18:47", "remaining_time": "13:33:35"}
+{"current_steps": 486, "total_steps": 2470, "loss": 0.4094, "lr": 3.886998572660485e-05, "epoch": 0.9838056680161943, "percentage": 19.68, "elapsed_time": "3:19:11", "remaining_time": "13:33:10"}
+{"current_steps": 487, "total_steps": 2470, "loss": 0.4118, "lr": 3.8860600769480215e-05, "epoch": 0.9858299595141701, "percentage": 19.72, "elapsed_time": "3:19:35", "remaining_time": "13:32:42"}
+{"current_steps": 488, "total_steps": 2470, "loss": 0.4165, "lr": 3.885117814403395e-05, "epoch": 0.9878542510121457, "percentage": 19.76, "elapsed_time": "3:20:00", "remaining_time": "13:32:18"}
+{"current_steps": 489, "total_steps": 2470, "loss": 0.4098, "lr": 3.8841717869084885e-05, "epoch": 0.9898785425101214, "percentage": 19.8, "elapsed_time": "3:20:23", "remaining_time": "13:31:50"}
+{"current_steps": 490, "total_steps": 2470, "loss": 0.4079, "lr": 3.883221996352705e-05, "epoch": 0.9919028340080972, "percentage": 19.84, "elapsed_time": "3:20:48", "remaining_time": "13:31:25"}
+{"current_steps": 491, "total_steps": 2470, "loss": 0.4112, "lr": 3.8822684446329625e-05, "epoch": 0.9939271255060729, "percentage": 19.88, "elapsed_time": "3:21:12", "remaining_time": "13:30:58"}
+{"current_steps": 492, "total_steps": 2470, "loss": 0.3885, "lr": 3.881311133653691e-05, "epoch": 0.9959514170040485, "percentage": 19.92, "elapsed_time": "3:21:36", "remaining_time": "13:30:31"}
+{"current_steps": 493, "total_steps": 2470, "loss": 0.4099, "lr": 3.880350065326829e-05, "epoch": 0.9979757085020243, "percentage": 19.96, "elapsed_time": "3:22:01", "remaining_time": "13:30:10"}
+{"current_steps": 494, "total_steps": 2470, "loss": 0.4244, "lr": 3.879385241571817e-05, "epoch": 1.0, "percentage": 20.0, "elapsed_time": "3:22:26", "remaining_time": "13:29:46"}
+{"current_steps": 495, "total_steps": 2470, "loss": 0.3799, "lr": 3.878416664315599e-05, "epoch": 1.0020242914979758, "percentage": 20.04, "elapsed_time": "3:23:35", "remaining_time": "13:32:19"}
+{"current_steps": 496, "total_steps": 2470, "loss": 0.3697, "lr": 3.877444335492613e-05, "epoch": 1.0040485829959513, "percentage": 20.08, "elapsed_time": "3:24:00", "remaining_time": "13:31:56"}
+{"current_steps": 497, "total_steps": 2470, "loss": 0.3734, "lr": 3.87646825704479e-05, "epoch": 1.0060728744939271, "percentage": 20.12, "elapsed_time": "3:24:25", "remaining_time": "13:31:31"}
+{"current_steps": 498, "total_steps": 2470, "loss": 0.3928, "lr": 3.875488430921552e-05, "epoch": 1.008097165991903, "percentage": 20.16, "elapsed_time": "3:24:49", "remaining_time": "13:31:05"}
+{"current_steps": 499, "total_steps": 2470, "loss": 0.3701, "lr": 3.874504859079803e-05, "epoch": 1.0101214574898785, "percentage": 20.2, "elapsed_time": "3:25:14", "remaining_time": "13:30:41"}
+{"current_steps": 500, "total_steps": 2470, "loss": 0.378, "lr": 3.8735175434839284e-05, "epoch": 1.0121457489878543, "percentage": 20.24, "elapsed_time": "3:25:38", "remaining_time": "13:30:12"}
+{"current_steps": 501, "total_steps": 2470, "loss": 0.374, "lr": 3.8725264861057924e-05, "epoch": 1.01417004048583, "percentage": 20.28, "elapsed_time": "3:26:02", "remaining_time": "13:29:47"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6510817e2f012acf8d3f639adcbe75f2c1b997b4c80d91eac80d54b7623d6d93
+size 7160