Training in progress, epoch 0

.gitattributes

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

added_tokens.json

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

config.json

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

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcf8479372330b9cae859cbc6424e808c38100865d8b569c3aa22a2909a15371
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:766b619b61fd8fc8ff149de763408b0539269b88d09c4c0364683c7e404f38a4
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e05018c6db44d34ee8138c1106d44c1034113992b872e1e4152f72892d50d975
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81c943065fb7a868fe8c917289653d26fdd930b0e7c7c0af93729c90635c82b6
+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 +1,496 @@
+{"current_steps": 1, "total_steps": 2440, "loss": 0.89, "lr": 1.639344262295082e-07, "epoch": 0.0020470829068577278, "percentage": 0.04, "elapsed_time": "0:00:22", "remaining_time": "15:10:07"}
+{"current_steps": 2, "total_steps": 2440, "loss": 0.8602, "lr": 3.278688524590164e-07, "epoch": 0.0040941658137154556, "percentage": 0.08, "elapsed_time": "0:00:39", "remaining_time": "13:20:38"}
+{"current_steps": 3, "total_steps": 2440, "loss": 0.8902, "lr": 4.918032786885246e-07, "epoch": 0.006141248720573183, "percentage": 0.12, "elapsed_time": "0:00:58", "remaining_time": "13:11:23"}
+{"current_steps": 4, "total_steps": 2440, "loss": 0.9194, "lr": 6.557377049180328e-07, "epoch": 0.008188331627430911, "percentage": 0.16, "elapsed_time": "0:01:21", "remaining_time": "13:49:02"}
+{"current_steps": 5, "total_steps": 2440, "loss": 0.8523, "lr": 8.196721311475409e-07, "epoch": 0.01023541453428864, "percentage": 0.2, "elapsed_time": "0:01:38", "remaining_time": "13:20:12"}
+{"current_steps": 6, "total_steps": 2440, "loss": 0.9102, "lr": 9.836065573770493e-07, "epoch": 0.012282497441146366, "percentage": 0.25, "elapsed_time": "0:01:58", "remaining_time": "13:21:05"}
+{"current_steps": 7, "total_steps": 2440, "loss": 0.8682, "lr": 1.1475409836065575e-06, "epoch": 0.014329580348004094, "percentage": 0.29, "elapsed_time": "0:02:14", "remaining_time": "13:01:38"}
+{"current_steps": 8, "total_steps": 2440, "loss": 0.8689, "lr": 1.3114754098360657e-06, "epoch": 0.016376663254861822, "percentage": 0.33, "elapsed_time": "0:02:33", "remaining_time": "12:56:57"}
+{"current_steps": 9, "total_steps": 2440, "loss": 0.852, "lr": 1.4754098360655739e-06, "epoch": 0.01842374616171955, "percentage": 0.37, "elapsed_time": "0:02:52", "remaining_time": "12:56:48"}
+{"current_steps": 10, "total_steps": 2440, "loss": 0.8102, "lr": 1.6393442622950819e-06, "epoch": 0.02047082906857728, "percentage": 0.41, "elapsed_time": "0:03:11", "remaining_time": "12:56:09"}
+{"current_steps": 11, "total_steps": 2440, "loss": 0.8102, "lr": 1.8032786885245903e-06, "epoch": 0.022517911975435005, "percentage": 0.45, "elapsed_time": "0:03:27", "remaining_time": "12:43:02"}
+{"current_steps": 12, "total_steps": 2440, "loss": 0.7481, "lr": 1.9672131147540985e-06, "epoch": 0.02456499488229273, "percentage": 0.49, "elapsed_time": "0:03:45", "remaining_time": "12:39:09"}
+{"current_steps": 13, "total_steps": 2440, "loss": 0.7631, "lr": 2.1311475409836067e-06, "epoch": 0.02661207778915046, "percentage": 0.53, "elapsed_time": "0:04:04", "remaining_time": "12:40:51"}
+{"current_steps": 14, "total_steps": 2440, "loss": 0.7506, "lr": 2.295081967213115e-06, "epoch": 0.028659160696008188, "percentage": 0.57, "elapsed_time": "0:04:23", "remaining_time": "12:40:12"}
+{"current_steps": 15, "total_steps": 2440, "loss": 0.7781, "lr": 2.459016393442623e-06, "epoch": 0.030706243602865915, "percentage": 0.61, "elapsed_time": "0:04:42", "remaining_time": "12:40:54"}
+{"current_steps": 16, "total_steps": 2440, "loss": 0.7488, "lr": 2.6229508196721314e-06, "epoch": 0.032753326509723645, "percentage": 0.66, "elapsed_time": "0:05:03", "remaining_time": "12:45:10"}
+{"current_steps": 17, "total_steps": 2440, "loss": 0.7746, "lr": 2.786885245901639e-06, "epoch": 0.03480040941658137, "percentage": 0.7, "elapsed_time": "0:05:22", "remaining_time": "12:46:23"}
+{"current_steps": 18, "total_steps": 2440, "loss": 0.7618, "lr": 2.9508196721311478e-06, "epoch": 0.0368474923234391, "percentage": 0.74, "elapsed_time": "0:05:39", "remaining_time": "12:41:54"}
+{"current_steps": 19, "total_steps": 2440, "loss": 0.7207, "lr": 3.114754098360656e-06, "epoch": 0.038894575230296824, "percentage": 0.78, "elapsed_time": "0:05:57", "remaining_time": "12:39:44"}
+{"current_steps": 20, "total_steps": 2440, "loss": 0.7352, "lr": 3.2786885245901638e-06, "epoch": 0.04094165813715456, "percentage": 0.82, "elapsed_time": "0:06:17", "remaining_time": "12:41:55"}
+{"current_steps": 21, "total_steps": 2440, "loss": 0.7432, "lr": 3.4426229508196724e-06, "epoch": 0.042988741044012284, "percentage": 0.86, "elapsed_time": "0:06:34", "remaining_time": "12:37:45"}
+{"current_steps": 22, "total_steps": 2440, "loss": 0.7444, "lr": 3.6065573770491806e-06, "epoch": 0.04503582395087001, "percentage": 0.9, "elapsed_time": "0:06:57", "remaining_time": "12:44:13"}
+{"current_steps": 23, "total_steps": 2440, "loss": 0.6646, "lr": 3.7704918032786884e-06, "epoch": 0.04708290685772774, "percentage": 0.94, "elapsed_time": "0:07:15", "remaining_time": "12:42:50"}
+{"current_steps": 24, "total_steps": 2440, "loss": 0.6388, "lr": 3.934426229508197e-06, "epoch": 0.04912998976458546, "percentage": 0.98, "elapsed_time": "0:07:35", "remaining_time": "12:43:56"}
+{"current_steps": 25, "total_steps": 2440, "loss": 0.6575, "lr": 4.098360655737705e-06, "epoch": 0.0511770726714432, "percentage": 1.02, "elapsed_time": "0:07:56", "remaining_time": "12:46:54"}
+{"current_steps": 26, "total_steps": 2440, "loss": 0.671, "lr": 4.2622950819672135e-06, "epoch": 0.05322415557830092, "percentage": 1.07, "elapsed_time": "0:08:15", "remaining_time": "12:46:17"}
+{"current_steps": 27, "total_steps": 2440, "loss": 0.6526, "lr": 4.426229508196722e-06, "epoch": 0.05527123848515865, "percentage": 1.11, "elapsed_time": "0:08:33", "remaining_time": "12:45:22"}
+{"current_steps": 28, "total_steps": 2440, "loss": 0.6533, "lr": 4.59016393442623e-06, "epoch": 0.057318321392016376, "percentage": 1.15, "elapsed_time": "0:08:47", "remaining_time": "12:37:33"}
+{"current_steps": 29, "total_steps": 2440, "loss": 0.6525, "lr": 4.754098360655738e-06, "epoch": 0.0593654042988741, "percentage": 1.19, "elapsed_time": "0:09:14", "remaining_time": "12:48:07"}
+{"current_steps": 30, "total_steps": 2440, "loss": 0.6513, "lr": 4.918032786885246e-06, "epoch": 0.06141248720573183, "percentage": 1.23, "elapsed_time": "0:09:30", "remaining_time": "12:43:36"}
+{"current_steps": 31, "total_steps": 2440, "loss": 0.6517, "lr": 5.0819672131147545e-06, "epoch": 0.06345957011258956, "percentage": 1.27, "elapsed_time": "0:09:50", "remaining_time": "12:44:40"}
+{"current_steps": 32, "total_steps": 2440, "loss": 0.6562, "lr": 5.245901639344263e-06, "epoch": 0.06550665301944729, "percentage": 1.31, "elapsed_time": "0:10:10", "remaining_time": "12:45:41"}
+{"current_steps": 33, "total_steps": 2440, "loss": 0.6318, "lr": 5.409836065573772e-06, "epoch": 0.06755373592630501, "percentage": 1.35, "elapsed_time": "0:10:31", "remaining_time": "12:47:44"}
+{"current_steps": 34, "total_steps": 2440, "loss": 0.6528, "lr": 5.573770491803278e-06, "epoch": 0.06960081883316274, "percentage": 1.39, "elapsed_time": "0:10:51", "remaining_time": "12:47:48"}
+{"current_steps": 35, "total_steps": 2440, "loss": 0.6233, "lr": 5.737704918032787e-06, "epoch": 0.07164790174002048, "percentage": 1.43, "elapsed_time": "0:11:05", "remaining_time": "12:41:44"}
+{"current_steps": 36, "total_steps": 2440, "loss": 0.5959, "lr": 5.9016393442622956e-06, "epoch": 0.0736949846468782, "percentage": 1.48, "elapsed_time": "0:11:19", "remaining_time": "12:36:31"}
+{"current_steps": 37, "total_steps": 2440, "loss": 0.6031, "lr": 6.065573770491804e-06, "epoch": 0.07574206755373593, "percentage": 1.52, "elapsed_time": "0:11:38", "remaining_time": "12:36:14"}
+{"current_steps": 38, "total_steps": 2440, "loss": 0.6252, "lr": 6.229508196721312e-06, "epoch": 0.07778915046059365, "percentage": 1.56, "elapsed_time": "0:11:58", "remaining_time": "12:36:40"}
+{"current_steps": 39, "total_steps": 2440, "loss": 0.5891, "lr": 6.393442622950821e-06, "epoch": 0.07983623336745138, "percentage": 1.6, "elapsed_time": "0:12:19", "remaining_time": "12:38:18"}
+{"current_steps": 40, "total_steps": 2440, "loss": 0.6151, "lr": 6.5573770491803276e-06, "epoch": 0.08188331627430911, "percentage": 1.64, "elapsed_time": "0:12:40", "remaining_time": "12:40:40"}
+{"current_steps": 41, "total_steps": 2440, "loss": 0.5832, "lr": 6.721311475409837e-06, "epoch": 0.08393039918116683, "percentage": 1.68, "elapsed_time": "0:13:02", "remaining_time": "12:42:48"}
+{"current_steps": 42, "total_steps": 2440, "loss": 0.629, "lr": 6.885245901639345e-06, "epoch": 0.08597748208802457, "percentage": 1.72, "elapsed_time": "0:13:22", "remaining_time": "12:43:55"}
+{"current_steps": 43, "total_steps": 2440, "loss": 0.5531, "lr": 7.049180327868853e-06, "epoch": 0.08802456499488229, "percentage": 1.76, "elapsed_time": "0:13:41", "remaining_time": "12:43:28"}
+{"current_steps": 44, "total_steps": 2440, "loss": 0.612, "lr": 7.213114754098361e-06, "epoch": 0.09007164790174002, "percentage": 1.8, "elapsed_time": "0:14:02", "remaining_time": "12:44:28"}
+{"current_steps": 45, "total_steps": 2440, "loss": 0.5675, "lr": 7.3770491803278695e-06, "epoch": 0.09211873080859775, "percentage": 1.84, "elapsed_time": "0:14:18", "remaining_time": "12:41:27"}
+{"current_steps": 46, "total_steps": 2440, "loss": 0.6391, "lr": 7.540983606557377e-06, "epoch": 0.09416581371545547, "percentage": 1.89, "elapsed_time": "0:14:36", "remaining_time": "12:40:09"}
+{"current_steps": 47, "total_steps": 2440, "loss": 0.5773, "lr": 7.704918032786886e-06, "epoch": 0.09621289662231321, "percentage": 1.93, "elapsed_time": "0:14:47", "remaining_time": "12:33:31"}
+{"current_steps": 48, "total_steps": 2440, "loss": 0.5928, "lr": 7.868852459016394e-06, "epoch": 0.09825997952917093, "percentage": 1.97, "elapsed_time": "0:15:07", "remaining_time": "12:33:50"}
+{"current_steps": 49, "total_steps": 2440, "loss": 0.5747, "lr": 8.032786885245902e-06, "epoch": 0.10030706243602866, "percentage": 2.01, "elapsed_time": "0:15:26", "remaining_time": "12:33:26"}
+{"current_steps": 50, "total_steps": 2440, "loss": 0.5506, "lr": 8.19672131147541e-06, "epoch": 0.1023541453428864, "percentage": 2.05, "elapsed_time": "0:15:39", "remaining_time": "12:28:45"}
+{"current_steps": 51, "total_steps": 2440, "loss": 0.5974, "lr": 8.360655737704919e-06, "epoch": 0.10440122824974411, "percentage": 2.09, "elapsed_time": "0:15:58", "remaining_time": "12:28:07"}
+{"current_steps": 52, "total_steps": 2440, "loss": 0.6027, "lr": 8.524590163934427e-06, "epoch": 0.10644831115660185, "percentage": 2.13, "elapsed_time": "0:16:18", "remaining_time": "12:28:44"}
+{"current_steps": 53, "total_steps": 2440, "loss": 0.5997, "lr": 8.688524590163935e-06, "epoch": 0.10849539406345957, "percentage": 2.17, "elapsed_time": "0:16:36", "remaining_time": "12:27:44"}
+{"current_steps": 54, "total_steps": 2440, "loss": 0.5633, "lr": 8.852459016393443e-06, "epoch": 0.1105424769703173, "percentage": 2.21, "elapsed_time": "0:16:53", "remaining_time": "12:26:43"}
+{"current_steps": 55, "total_steps": 2440, "loss": 0.5842, "lr": 9.016393442622952e-06, "epoch": 0.11258955987717502, "percentage": 2.25, "elapsed_time": "0:17:14", "remaining_time": "12:27:24"}
+{"current_steps": 56, "total_steps": 2440, "loss": 0.5645, "lr": 9.18032786885246e-06, "epoch": 0.11463664278403275, "percentage": 2.3, "elapsed_time": "0:17:32", "remaining_time": "12:26:27"}
+{"current_steps": 57, "total_steps": 2440, "loss": 0.5635, "lr": 9.344262295081968e-06, "epoch": 0.11668372569089049, "percentage": 2.34, "elapsed_time": "0:17:51", "remaining_time": "12:26:36"}
+{"current_steps": 58, "total_steps": 2440, "loss": 0.5658, "lr": 9.508196721311476e-06, "epoch": 0.1187308085977482, "percentage": 2.38, "elapsed_time": "0:18:09", "remaining_time": "12:25:44"}
+{"current_steps": 59, "total_steps": 2440, "loss": 0.5559, "lr": 9.672131147540984e-06, "epoch": 0.12077789150460594, "percentage": 2.42, "elapsed_time": "0:18:27", "remaining_time": "12:24:41"}
+{"current_steps": 60, "total_steps": 2440, "loss": 0.5755, "lr": 9.836065573770493e-06, "epoch": 0.12282497441146366, "percentage": 2.46, "elapsed_time": "0:18:45", "remaining_time": "12:23:49"}
+{"current_steps": 61, "total_steps": 2440, "loss": 0.5422, "lr": 1e-05, "epoch": 0.12487205731832139, "percentage": 2.5, "elapsed_time": "0:19:01", "remaining_time": "12:22:10"}
+{"current_steps": 62, "total_steps": 2440, "loss": 0.5638, "lr": 1.0163934426229509e-05, "epoch": 0.1269191402251791, "percentage": 2.54, "elapsed_time": "0:19:17", "remaining_time": "12:19:51"}
+{"current_steps": 63, "total_steps": 2440, "loss": 0.5874, "lr": 1.0327868852459017e-05, "epoch": 0.12896622313203684, "percentage": 2.58, "elapsed_time": "0:19:33", "remaining_time": "12:17:41"}
+{"current_steps": 64, "total_steps": 2440, "loss": 0.5549, "lr": 1.0491803278688525e-05, "epoch": 0.13101330603889458, "percentage": 2.62, "elapsed_time": "0:19:51", "remaining_time": "12:17:27"}
+{"current_steps": 65, "total_steps": 2440, "loss": 0.5287, "lr": 1.0655737704918034e-05, "epoch": 0.1330603889457523, "percentage": 2.66, "elapsed_time": "0:20:10", "remaining_time": "12:16:53"}
+{"current_steps": 66, "total_steps": 2440, "loss": 0.5726, "lr": 1.0819672131147544e-05, "epoch": 0.13510747185261002, "percentage": 2.7, "elapsed_time": "0:20:26", "remaining_time": "12:15:19"}
+{"current_steps": 67, "total_steps": 2440, "loss": 0.641, "lr": 1.0983606557377052e-05, "epoch": 0.13715455475946775, "percentage": 2.75, "elapsed_time": "0:20:48", "remaining_time": "12:16:49"}
+{"current_steps": 68, "total_steps": 2440, "loss": 0.5493, "lr": 1.1147540983606557e-05, "epoch": 0.13920163766632548, "percentage": 2.79, "elapsed_time": "0:21:08", "remaining_time": "12:17:13"}
+{"current_steps": 69, "total_steps": 2440, "loss": 0.5231, "lr": 1.1311475409836066e-05, "epoch": 0.14124872057318322, "percentage": 2.83, "elapsed_time": "0:21:27", "remaining_time": "12:17:07"}
+{"current_steps": 70, "total_steps": 2440, "loss": 0.5221, "lr": 1.1475409836065575e-05, "epoch": 0.14329580348004095, "percentage": 2.87, "elapsed_time": "0:21:46", "remaining_time": "12:17:11"}
+{"current_steps": 71, "total_steps": 2440, "loss": 0.5254, "lr": 1.1639344262295083e-05, "epoch": 0.14534288638689866, "percentage": 2.91, "elapsed_time": "0:22:03", "remaining_time": "12:15:43"}
+{"current_steps": 72, "total_steps": 2440, "loss": 0.5855, "lr": 1.1803278688524591e-05, "epoch": 0.1473899692937564, "percentage": 2.95, "elapsed_time": "0:22:19", "remaining_time": "12:14:28"}
+{"current_steps": 73, "total_steps": 2440, "loss": 0.5277, "lr": 1.19672131147541e-05, "epoch": 0.14943705220061412, "percentage": 2.99, "elapsed_time": "0:22:35", "remaining_time": "12:12:41"}
+{"current_steps": 74, "total_steps": 2440, "loss": 0.5226, "lr": 1.2131147540983608e-05, "epoch": 0.15148413510747186, "percentage": 3.03, "elapsed_time": "0:22:56", "remaining_time": "12:13:29"}
+{"current_steps": 75, "total_steps": 2440, "loss": 0.5588, "lr": 1.2295081967213116e-05, "epoch": 0.1535312180143296, "percentage": 3.07, "elapsed_time": "0:23:15", "remaining_time": "12:13:34"}
+{"current_steps": 76, "total_steps": 2440, "loss": 0.5601, "lr": 1.2459016393442624e-05, "epoch": 0.1555783009211873, "percentage": 3.11, "elapsed_time": "0:23:30", "remaining_time": "12:11:27"}
+{"current_steps": 77, "total_steps": 2440, "loss": 0.5208, "lr": 1.2622950819672132e-05, "epoch": 0.15762538382804503, "percentage": 3.16, "elapsed_time": "0:23:46", "remaining_time": "12:09:49"}
+{"current_steps": 78, "total_steps": 2440, "loss": 0.5646, "lr": 1.2786885245901642e-05, "epoch": 0.15967246673490276, "percentage": 3.2, "elapsed_time": "0:24:03", "remaining_time": "12:08:43"}
+{"current_steps": 79, "total_steps": 2440, "loss": 0.5345, "lr": 1.295081967213115e-05, "epoch": 0.1617195496417605, "percentage": 3.24, "elapsed_time": "0:24:25", "remaining_time": "12:10:02"}
+{"current_steps": 80, "total_steps": 2440, "loss": 0.5516, "lr": 1.3114754098360655e-05, "epoch": 0.16376663254861823, "percentage": 3.28, "elapsed_time": "0:24:43", "remaining_time": "12:09:28"}
+{"current_steps": 81, "total_steps": 2440, "loss": 0.5494, "lr": 1.3278688524590165e-05, "epoch": 0.16581371545547594, "percentage": 3.32, "elapsed_time": "0:25:01", "remaining_time": "12:08:43"}
+{"current_steps": 82, "total_steps": 2440, "loss": 0.5374, "lr": 1.3442622950819673e-05, "epoch": 0.16786079836233367, "percentage": 3.36, "elapsed_time": "0:25:17", "remaining_time": "12:07:04"}
+{"current_steps": 83, "total_steps": 2440, "loss": 0.546, "lr": 1.3606557377049181e-05, "epoch": 0.1699078812691914, "percentage": 3.4, "elapsed_time": "0:25:36", "remaining_time": "12:07:17"}
+{"current_steps": 84, "total_steps": 2440, "loss": 0.5761, "lr": 1.377049180327869e-05, "epoch": 0.17195496417604914, "percentage": 3.44, "elapsed_time": "0:25:53", "remaining_time": "12:06:11"}
+{"current_steps": 85, "total_steps": 2440, "loss": 0.5326, "lr": 1.3934426229508198e-05, "epoch": 0.17400204708290687, "percentage": 3.48, "elapsed_time": "0:26:15", "remaining_time": "12:07:19"}
+{"current_steps": 86, "total_steps": 2440, "loss": 0.5311, "lr": 1.4098360655737706e-05, "epoch": 0.17604912998976457, "percentage": 3.52, "elapsed_time": "0:26:35", "remaining_time": "12:07:58"}
+{"current_steps": 87, "total_steps": 2440, "loss": 0.5206, "lr": 1.4262295081967214e-05, "epoch": 0.1780962128966223, "percentage": 3.57, "elapsed_time": "0:26:51", "remaining_time": "12:06:13"}
+{"current_steps": 88, "total_steps": 2440, "loss": 0.5575, "lr": 1.4426229508196722e-05, "epoch": 0.18014329580348004, "percentage": 3.61, "elapsed_time": "0:27:09", "remaining_time": "12:05:56"}
+{"current_steps": 89, "total_steps": 2440, "loss": 0.5243, "lr": 1.459016393442623e-05, "epoch": 0.18219037871033777, "percentage": 3.65, "elapsed_time": "0:27:28", "remaining_time": "12:05:46"}
+{"current_steps": 90, "total_steps": 2440, "loss": 0.5375, "lr": 1.4754098360655739e-05, "epoch": 0.1842374616171955, "percentage": 3.69, "elapsed_time": "0:27:46", "remaining_time": "12:05:04"}
+{"current_steps": 91, "total_steps": 2440, "loss": 0.5678, "lr": 1.4918032786885249e-05, "epoch": 0.1862845445240532, "percentage": 3.73, "elapsed_time": "0:28:02", "remaining_time": "12:03:55"}
+{"current_steps": 92, "total_steps": 2440, "loss": 0.5317, "lr": 1.5081967213114754e-05, "epoch": 0.18833162743091095, "percentage": 3.77, "elapsed_time": "0:28:15", "remaining_time": "12:01:01"}
+{"current_steps": 93, "total_steps": 2440, "loss": 0.5208, "lr": 1.5245901639344264e-05, "epoch": 0.19037871033776868, "percentage": 3.81, "elapsed_time": "0:28:35", "remaining_time": "12:01:31"}
+{"current_steps": 94, "total_steps": 2440, "loss": 0.5538, "lr": 1.5409836065573772e-05, "epoch": 0.19242579324462641, "percentage": 3.85, "elapsed_time": "0:28:52", "remaining_time": "12:00:32"}
+{"current_steps": 95, "total_steps": 2440, "loss": 0.4945, "lr": 1.5573770491803278e-05, "epoch": 0.19447287615148415, "percentage": 3.89, "elapsed_time": "0:29:06", "remaining_time": "11:58:28"}
+{"current_steps": 96, "total_steps": 2440, "loss": 0.5542, "lr": 1.5737704918032788e-05, "epoch": 0.19651995905834185, "percentage": 3.93, "elapsed_time": "0:29:21", "remaining_time": "11:56:49"}
+{"current_steps": 97, "total_steps": 2440, "loss": 0.5604, "lr": 1.5901639344262295e-05, "epoch": 0.1985670419651996, "percentage": 3.98, "elapsed_time": "0:29:41", "remaining_time": "11:57:19"}
+{"current_steps": 98, "total_steps": 2440, "loss": 0.492, "lr": 1.6065573770491805e-05, "epoch": 0.20061412487205732, "percentage": 4.02, "elapsed_time": "0:29:58", "remaining_time": "11:56:16"}
+{"current_steps": 99, "total_steps": 2440, "loss": 0.5561, "lr": 1.6229508196721314e-05, "epoch": 0.20266120777891505, "percentage": 4.06, "elapsed_time": "0:30:18", "remaining_time": "11:56:43"}
+{"current_steps": 100, "total_steps": 2440, "loss": 0.5319, "lr": 1.639344262295082e-05, "epoch": 0.2047082906857728, "percentage": 4.1, "elapsed_time": "0:30:37", "remaining_time": "11:56:38"}
+{"current_steps": 101, "total_steps": 2440, "loss": 0.5087, "lr": 1.655737704918033e-05, "epoch": 0.2067553735926305, "percentage": 4.14, "elapsed_time": "0:30:57", "remaining_time": "11:56:48"}
+{"current_steps": 102, "total_steps": 2440, "loss": 0.5187, "lr": 1.6721311475409837e-05, "epoch": 0.20880245649948823, "percentage": 4.18, "elapsed_time": "0:31:15", "remaining_time": "11:56:22"}
+{"current_steps": 103, "total_steps": 2440, "loss": 0.5831, "lr": 1.6885245901639347e-05, "epoch": 0.21084953940634596, "percentage": 4.22, "elapsed_time": "0:31:34", "remaining_time": "11:56:17"}
+{"current_steps": 104, "total_steps": 2440, "loss": 0.5979, "lr": 1.7049180327868854e-05, "epoch": 0.2128966223132037, "percentage": 4.26, "elapsed_time": "0:31:52", "remaining_time": "11:55:49"}
+{"current_steps": 105, "total_steps": 2440, "loss": 0.5337, "lr": 1.721311475409836e-05, "epoch": 0.21494370522006143, "percentage": 4.3, "elapsed_time": "0:32:09", "remaining_time": "11:55:08"}
+{"current_steps": 106, "total_steps": 2440, "loss": 0.5014, "lr": 1.737704918032787e-05, "epoch": 0.21699078812691913, "percentage": 4.34, "elapsed_time": "0:32:29", "remaining_time": "11:55:26"}
+{"current_steps": 107, "total_steps": 2440, "loss": 0.5298, "lr": 1.7540983606557377e-05, "epoch": 0.21903787103377687, "percentage": 4.39, "elapsed_time": "0:32:43", "remaining_time": "11:53:33"}
+{"current_steps": 108, "total_steps": 2440, "loss": 0.5597, "lr": 1.7704918032786887e-05, "epoch": 0.2210849539406346, "percentage": 4.43, "elapsed_time": "0:33:01", "remaining_time": "11:53:10"}
+{"current_steps": 109, "total_steps": 2440, "loss": 0.5671, "lr": 1.7868852459016393e-05, "epoch": 0.22313203684749233, "percentage": 4.47, "elapsed_time": "0:33:19", "remaining_time": "11:52:29"}
+{"current_steps": 110, "total_steps": 2440, "loss": 0.5608, "lr": 1.8032786885245903e-05, "epoch": 0.22517911975435004, "percentage": 4.51, "elapsed_time": "0:33:41", "remaining_time": "11:53:32"}
+{"current_steps": 111, "total_steps": 2440, "loss": 0.536, "lr": 1.8196721311475413e-05, "epoch": 0.22722620266120777, "percentage": 4.55, "elapsed_time": "0:34:00", "remaining_time": "11:53:29"}
+{"current_steps": 112, "total_steps": 2440, "loss": 0.5107, "lr": 1.836065573770492e-05, "epoch": 0.2292732855680655, "percentage": 4.59, "elapsed_time": "0:34:20", "remaining_time": "11:53:45"}
+{"current_steps": 113, "total_steps": 2440, "loss": 0.5216, "lr": 1.852459016393443e-05, "epoch": 0.23132036847492324, "percentage": 4.63, "elapsed_time": "0:34:42", "remaining_time": "11:54:52"}
+{"current_steps": 114, "total_steps": 2440, "loss": 0.5673, "lr": 1.8688524590163936e-05, "epoch": 0.23336745138178097, "percentage": 4.67, "elapsed_time": "0:35:02", "remaining_time": "11:54:54"}
+{"current_steps": 115, "total_steps": 2440, "loss": 0.5173, "lr": 1.8852459016393446e-05, "epoch": 0.23541453428863868, "percentage": 4.71, "elapsed_time": "0:35:19", "remaining_time": "11:54:13"}
+{"current_steps": 116, "total_steps": 2440, "loss": 0.526, "lr": 1.9016393442622952e-05, "epoch": 0.2374616171954964, "percentage": 4.75, "elapsed_time": "0:35:40", "remaining_time": "11:54:46"}
+{"current_steps": 117, "total_steps": 2440, "loss": 0.5501, "lr": 1.918032786885246e-05, "epoch": 0.23950870010235414, "percentage": 4.8, "elapsed_time": "0:35:59", "remaining_time": "11:54:39"}
+{"current_steps": 118, "total_steps": 2440, "loss": 0.5564, "lr": 1.934426229508197e-05, "epoch": 0.24155578300921188, "percentage": 4.84, "elapsed_time": "0:36:19", "remaining_time": "11:54:54"}
+{"current_steps": 119, "total_steps": 2440, "loss": 0.5307, "lr": 1.9508196721311475e-05, "epoch": 0.2436028659160696, "percentage": 4.88, "elapsed_time": "0:36:37", "remaining_time": "11:54:25"}
+{"current_steps": 120, "total_steps": 2440, "loss": 0.4867, "lr": 1.9672131147540985e-05, "epoch": 0.24564994882292732, "percentage": 4.92, "elapsed_time": "0:36:58", "remaining_time": "11:54:48"}
+{"current_steps": 121, "total_steps": 2440, "loss": 0.5235, "lr": 1.9836065573770492e-05, "epoch": 0.24769703172978505, "percentage": 4.96, "elapsed_time": "0:37:17", "remaining_time": "11:54:33"}
+{"current_steps": 122, "total_steps": 2440, "loss": 0.524, "lr": 2e-05, "epoch": 0.24974411463664278, "percentage": 5.0, "elapsed_time": "0:37:38", "remaining_time": "11:55:16"}
+{"current_steps": 123, "total_steps": 2440, "loss": 0.4759, "lr": 2.0163934426229508e-05, "epoch": 0.2517911975435005, "percentage": 5.04, "elapsed_time": "0:37:58", "remaining_time": "11:55:15"}
+{"current_steps": 124, "total_steps": 2440, "loss": 0.4925, "lr": 2.0327868852459018e-05, "epoch": 0.2538382804503582, "percentage": 5.08, "elapsed_time": "0:38:18", "remaining_time": "11:55:22"}
+{"current_steps": 125, "total_steps": 2440, "loss": 0.536, "lr": 2.0491803278688525e-05, "epoch": 0.25588536335721596, "percentage": 5.12, "elapsed_time": "0:38:36", "remaining_time": "11:54:56"}
+{"current_steps": 126, "total_steps": 2440, "loss": 0.5937, "lr": 2.0655737704918034e-05, "epoch": 0.2579324462640737, "percentage": 5.16, "elapsed_time": "0:38:58", "remaining_time": "11:55:39"}
+{"current_steps": 127, "total_steps": 2440, "loss": 0.5019, "lr": 2.081967213114754e-05, "epoch": 0.2599795291709314, "percentage": 5.2, "elapsed_time": "0:39:13", "remaining_time": "11:54:29"}
+{"current_steps": 128, "total_steps": 2440, "loss": 0.5643, "lr": 2.098360655737705e-05, "epoch": 0.26202661207778916, "percentage": 5.25, "elapsed_time": "0:39:34", "remaining_time": "11:54:47"}
+{"current_steps": 129, "total_steps": 2440, "loss": 0.5449, "lr": 2.1147540983606557e-05, "epoch": 0.2640736949846469, "percentage": 5.29, "elapsed_time": "0:39:52", "remaining_time": "11:54:15"}
+{"current_steps": 130, "total_steps": 2440, "loss": 0.5377, "lr": 2.1311475409836067e-05, "epoch": 0.2661207778915046, "percentage": 5.33, "elapsed_time": "0:40:13", "remaining_time": "11:54:42"}
+{"current_steps": 131, "total_steps": 2440, "loss": 0.5846, "lr": 2.1475409836065574e-05, "epoch": 0.26816786079836236, "percentage": 5.37, "elapsed_time": "0:40:31", "remaining_time": "11:54:17"}
+{"current_steps": 132, "total_steps": 2440, "loss": 0.5283, "lr": 2.1639344262295087e-05, "epoch": 0.27021494370522003, "percentage": 5.41, "elapsed_time": "0:40:54", "remaining_time": "11:55:21"}
+{"current_steps": 133, "total_steps": 2440, "loss": 0.557, "lr": 2.180327868852459e-05, "epoch": 0.27226202661207777, "percentage": 5.45, "elapsed_time": "0:41:13", "remaining_time": "11:55:12"}
+{"current_steps": 134, "total_steps": 2440, "loss": 0.511, "lr": 2.1967213114754104e-05, "epoch": 0.2743091095189355, "percentage": 5.49, "elapsed_time": "0:41:29", "remaining_time": "11:53:58"}
+{"current_steps": 135, "total_steps": 2440, "loss": 0.5182, "lr": 2.213114754098361e-05, "epoch": 0.27635619242579323, "percentage": 5.53, "elapsed_time": "0:41:44", "remaining_time": "11:52:41"}
+{"current_steps": 136, "total_steps": 2440, "loss": 0.5146, "lr": 2.2295081967213113e-05, "epoch": 0.27840327533265097, "percentage": 5.57, "elapsed_time": "0:42:01", "remaining_time": "11:51:50"}
+{"current_steps": 137, "total_steps": 2440, "loss": 0.5083, "lr": 2.2459016393442626e-05, "epoch": 0.2804503582395087, "percentage": 5.61, "elapsed_time": "0:42:22", "remaining_time": "11:52:13"}
+{"current_steps": 138, "total_steps": 2440, "loss": 0.5313, "lr": 2.2622950819672133e-05, "epoch": 0.28249744114636643, "percentage": 5.66, "elapsed_time": "0:42:43", "remaining_time": "11:52:48"}
+{"current_steps": 139, "total_steps": 2440, "loss": 0.5052, "lr": 2.2786885245901643e-05, "epoch": 0.28454452405322417, "percentage": 5.7, "elapsed_time": "0:43:03", "remaining_time": "11:52:52"}
+{"current_steps": 140, "total_steps": 2440, "loss": 0.5371, "lr": 2.295081967213115e-05, "epoch": 0.2865916069600819, "percentage": 5.74, "elapsed_time": "0:43:19", "remaining_time": "11:51:53"}
+{"current_steps": 141, "total_steps": 2440, "loss": 0.4964, "lr": 2.311475409836066e-05, "epoch": 0.28863868986693964, "percentage": 5.78, "elapsed_time": "0:43:40", "remaining_time": "11:52:05"}
+{"current_steps": 142, "total_steps": 2440, "loss": 0.524, "lr": 2.3278688524590166e-05, "epoch": 0.2906857727737973, "percentage": 5.82, "elapsed_time": "0:43:59", "remaining_time": "11:52:02"}
+{"current_steps": 143, "total_steps": 2440, "loss": 0.5347, "lr": 2.3442622950819676e-05, "epoch": 0.29273285568065505, "percentage": 5.86, "elapsed_time": "0:44:17", "remaining_time": "11:51:30"}
+{"current_steps": 144, "total_steps": 2440, "loss": 0.5212, "lr": 2.3606557377049182e-05, "epoch": 0.2947799385875128, "percentage": 5.9, "elapsed_time": "0:44:38", "remaining_time": "11:51:46"}
+{"current_steps": 145, "total_steps": 2440, "loss": 0.536, "lr": 2.3770491803278692e-05, "epoch": 0.2968270214943705, "percentage": 5.94, "elapsed_time": "0:44:57", "remaining_time": "11:51:35"}
+{"current_steps": 146, "total_steps": 2440, "loss": 0.506, "lr": 2.39344262295082e-05, "epoch": 0.29887410440122825, "percentage": 5.98, "elapsed_time": "0:45:13", "remaining_time": "11:50:30"}
+{"current_steps": 147, "total_steps": 2440, "loss": 0.5165, "lr": 2.4098360655737705e-05, "epoch": 0.300921187308086, "percentage": 6.02, "elapsed_time": "0:45:29", "remaining_time": "11:49:40"}
+{"current_steps": 148, "total_steps": 2440, "loss": 0.5353, "lr": 2.4262295081967215e-05, "epoch": 0.3029682702149437, "percentage": 6.07, "elapsed_time": "0:45:47", "remaining_time": "11:49:06"}
+{"current_steps": 149, "total_steps": 2440, "loss": 0.5378, "lr": 2.442622950819672e-05, "epoch": 0.30501535312180145, "percentage": 6.11, "elapsed_time": "0:46:09", "remaining_time": "11:49:41"}
+{"current_steps": 150, "total_steps": 2440, "loss": 0.5094, "lr": 2.459016393442623e-05, "epoch": 0.3070624360286592, "percentage": 6.15, "elapsed_time": "0:46:27", "remaining_time": "11:49:20"}
+{"current_steps": 151, "total_steps": 2440, "loss": 0.5196, "lr": 2.4754098360655738e-05, "epoch": 0.3091095189355169, "percentage": 6.19, "elapsed_time": "0:46:43", "remaining_time": "11:48:23"}
+{"current_steps": 152, "total_steps": 2440, "loss": 0.4937, "lr": 2.4918032786885248e-05, "epoch": 0.3111566018423746, "percentage": 6.23, "elapsed_time": "0:47:02", "remaining_time": "11:48:11"}
+{"current_steps": 153, "total_steps": 2440, "loss": 0.4788, "lr": 2.5081967213114754e-05, "epoch": 0.3132036847492323, "percentage": 6.27, "elapsed_time": "0:47:24", "remaining_time": "11:48:36"}
+{"current_steps": 154, "total_steps": 2440, "loss": 0.5275, "lr": 2.5245901639344264e-05, "epoch": 0.31525076765609006, "percentage": 6.31, "elapsed_time": "0:47:42", "remaining_time": "11:48:14"}
+{"current_steps": 155, "total_steps": 2440, "loss": 0.4861, "lr": 2.540983606557377e-05, "epoch": 0.3172978505629478, "percentage": 6.35, "elapsed_time": "0:48:00", "remaining_time": "11:47:41"}
+{"current_steps": 156, "total_steps": 2440, "loss": 0.5071, "lr": 2.5573770491803284e-05, "epoch": 0.3193449334698055, "percentage": 6.39, "elapsed_time": "0:48:18", "remaining_time": "11:47:10"}
+{"current_steps": 157, "total_steps": 2440, "loss": 0.5195, "lr": 2.5737704918032787e-05, "epoch": 0.32139201637666326, "percentage": 6.43, "elapsed_time": "0:48:34", "remaining_time": "11:46:27"}
+{"current_steps": 158, "total_steps": 2440, "loss": 0.5523, "lr": 2.59016393442623e-05, "epoch": 0.323439099283521, "percentage": 6.48, "elapsed_time": "0:48:51", "remaining_time": "11:45:43"}
+{"current_steps": 159, "total_steps": 2440, "loss": 0.5451, "lr": 2.6065573770491807e-05, "epoch": 0.3254861821903787, "percentage": 6.52, "elapsed_time": "0:49:08", "remaining_time": "11:44:54"}
+{"current_steps": 160, "total_steps": 2440, "loss": 0.5309, "lr": 2.622950819672131e-05, "epoch": 0.32753326509723646, "percentage": 6.56, "elapsed_time": "0:49:23", "remaining_time": "11:43:52"}
+{"current_steps": 161, "total_steps": 2440, "loss": 0.5342, "lr": 2.6393442622950824e-05, "epoch": 0.3295803480040942, "percentage": 6.6, "elapsed_time": "0:49:37", "remaining_time": "11:42:28"}
+{"current_steps": 162, "total_steps": 2440, "loss": 0.5181, "lr": 2.655737704918033e-05, "epoch": 0.33162743091095187, "percentage": 6.64, "elapsed_time": "0:49:56", "remaining_time": "11:42:12"}
+{"current_steps": 163, "total_steps": 2440, "loss": 0.5074, "lr": 2.672131147540984e-05, "epoch": 0.3336745138178096, "percentage": 6.68, "elapsed_time": "0:50:12", "remaining_time": "11:41:17"}
+{"current_steps": 164, "total_steps": 2440, "loss": 0.5603, "lr": 2.6885245901639346e-05, "epoch": 0.33572159672466734, "percentage": 6.72, "elapsed_time": "0:50:31", "remaining_time": "11:41:09"}
+{"current_steps": 165, "total_steps": 2440, "loss": 0.5272, "lr": 2.7049180327868856e-05, "epoch": 0.33776867963152507, "percentage": 6.76, "elapsed_time": "0:50:48", "remaining_time": "11:40:38"}
+{"current_steps": 166, "total_steps": 2440, "loss": 0.5196, "lr": 2.7213114754098363e-05, "epoch": 0.3398157625383828, "percentage": 6.8, "elapsed_time": "0:51:05", "remaining_time": "11:40:00"}
+{"current_steps": 167, "total_steps": 2440, "loss": 0.5439, "lr": 2.7377049180327873e-05, "epoch": 0.34186284544524054, "percentage": 6.84, "elapsed_time": "0:51:26", "remaining_time": "11:40:06"}
+{"current_steps": 168, "total_steps": 2440, "loss": 0.5375, "lr": 2.754098360655738e-05, "epoch": 0.34390992835209827, "percentage": 6.89, "elapsed_time": "0:51:47", "remaining_time": "11:40:26"}
+{"current_steps": 169, "total_steps": 2440, "loss": 0.5014, "lr": 2.770491803278689e-05, "epoch": 0.345957011258956, "percentage": 6.93, "elapsed_time": "0:52:03", "remaining_time": "11:39:36"}
+{"current_steps": 170, "total_steps": 2440, "loss": 0.5262, "lr": 2.7868852459016396e-05, "epoch": 0.34800409416581374, "percentage": 6.97, "elapsed_time": "0:52:21", "remaining_time": "11:39:06"}
+{"current_steps": 171, "total_steps": 2440, "loss": 0.5454, "lr": 2.8032786885245902e-05, "epoch": 0.3500511770726714, "percentage": 7.01, "elapsed_time": "0:52:38", "remaining_time": "11:38:29"}
+{"current_steps": 172, "total_steps": 2440, "loss": 0.562, "lr": 2.8196721311475412e-05, "epoch": 0.35209825997952915, "percentage": 7.05, "elapsed_time": "0:52:58", "remaining_time": "11:38:38"}
+{"current_steps": 173, "total_steps": 2440, "loss": 0.5147, "lr": 2.836065573770492e-05, "epoch": 0.3541453428863869, "percentage": 7.09, "elapsed_time": "0:53:18", "remaining_time": "11:38:39"}
+{"current_steps": 174, "total_steps": 2440, "loss": 0.561, "lr": 2.852459016393443e-05, "epoch": 0.3561924257932446, "percentage": 7.13, "elapsed_time": "0:53:36", "remaining_time": "11:38:14"}
+{"current_steps": 175, "total_steps": 2440, "loss": 0.5294, "lr": 2.8688524590163935e-05, "epoch": 0.35823950870010235, "percentage": 7.17, "elapsed_time": "0:53:56", "remaining_time": "11:38:14"}
+{"current_steps": 176, "total_steps": 2440, "loss": 0.5086, "lr": 2.8852459016393445e-05, "epoch": 0.3602865916069601, "percentage": 7.21, "elapsed_time": "0:54:13", "remaining_time": "11:37:27"}
+{"current_steps": 177, "total_steps": 2440, "loss": 0.5021, "lr": 2.901639344262295e-05, "epoch": 0.3623336745138178, "percentage": 7.25, "elapsed_time": "0:54:32", "remaining_time": "11:37:20"}
+{"current_steps": 178, "total_steps": 2440, "loss": 0.5007, "lr": 2.918032786885246e-05, "epoch": 0.36438075742067555, "percentage": 7.3, "elapsed_time": "0:54:51", "remaining_time": "11:37:13"}
+{"current_steps": 179, "total_steps": 2440, "loss": 0.5524, "lr": 2.9344262295081968e-05, "epoch": 0.3664278403275333, "percentage": 7.34, "elapsed_time": "0:55:09", "remaining_time": "11:36:46"}
+{"current_steps": 180, "total_steps": 2440, "loss": 0.5142, "lr": 2.9508196721311478e-05, "epoch": 0.368474923234391, "percentage": 7.38, "elapsed_time": "0:55:29", "remaining_time": "11:36:42"}
+{"current_steps": 181, "total_steps": 2440, "loss": 0.5187, "lr": 2.9672131147540984e-05, "epoch": 0.3705220061412487, "percentage": 7.42, "elapsed_time": "0:55:49", "remaining_time": "11:36:40"}
+{"current_steps": 182, "total_steps": 2440, "loss": 0.5508, "lr": 2.9836065573770498e-05, "epoch": 0.3725690890481064, "percentage": 7.46, "elapsed_time": "0:56:09", "remaining_time": "11:36:38"}
+{"current_steps": 183, "total_steps": 2440, "loss": 0.5348, "lr": 3.0000000000000004e-05, "epoch": 0.37461617195496416, "percentage": 7.5, "elapsed_time": "0:56:25", "remaining_time": "11:35:59"}
+{"current_steps": 184, "total_steps": 2440, "loss": 0.5447, "lr": 3.0163934426229507e-05, "epoch": 0.3766632548618219, "percentage": 7.54, "elapsed_time": "0:56:43", "remaining_time": "11:35:29"}
+{"current_steps": 185, "total_steps": 2440, "loss": 0.5474, "lr": 3.032786885245902e-05, "epoch": 0.37871033776867963, "percentage": 7.58, "elapsed_time": "0:57:03", "remaining_time": "11:35:27"}
+{"current_steps": 186, "total_steps": 2440, "loss": 0.5559, "lr": 3.0491803278688527e-05, "epoch": 0.38075742067553736, "percentage": 7.62, "elapsed_time": "0:57:21", "remaining_time": "11:35:03"}
+{"current_steps": 187, "total_steps": 2440, "loss": 0.4875, "lr": 3.065573770491804e-05, "epoch": 0.3828045035823951, "percentage": 7.66, "elapsed_time": "0:57:36", "remaining_time": "11:33:59"}
+{"current_steps": 188, "total_steps": 2440, "loss": 0.5499, "lr": 3.0819672131147544e-05, "epoch": 0.38485158648925283, "percentage": 7.7, "elapsed_time": "0:57:57", "remaining_time": "11:34:17"}
+{"current_steps": 189, "total_steps": 2440, "loss": 0.5144, "lr": 3.098360655737705e-05, "epoch": 0.38689866939611056, "percentage": 7.75, "elapsed_time": "0:58:17", "remaining_time": "11:34:19"}
+{"current_steps": 190, "total_steps": 2440, "loss": 0.6274, "lr": 3.1147540983606557e-05, "epoch": 0.3889457523029683, "percentage": 7.79, "elapsed_time": "0:58:36", "remaining_time": "11:34:00"}
+{"current_steps": 191, "total_steps": 2440, "loss": 0.5261, "lr": 3.131147540983607e-05, "epoch": 0.390992835209826, "percentage": 7.83, "elapsed_time": "0:58:53", "remaining_time": "11:33:22"}
+{"current_steps": 192, "total_steps": 2440, "loss": 0.5284, "lr": 3.1475409836065576e-05, "epoch": 0.3930399181166837, "percentage": 7.87, "elapsed_time": "0:59:07", "remaining_time": "11:32:13"}
+{"current_steps": 193, "total_steps": 2440, "loss": 0.4981, "lr": 3.163934426229509e-05, "epoch": 0.39508700102354144, "percentage": 7.91, "elapsed_time": "0:59:25", "remaining_time": "11:31:56"}
+{"current_steps": 194, "total_steps": 2440, "loss": 0.5184, "lr": 3.180327868852459e-05, "epoch": 0.3971340839303992, "percentage": 7.95, "elapsed_time": "0:59:40", "remaining_time": "11:30:53"}
+{"current_steps": 195, "total_steps": 2440, "loss": 0.5286, "lr": 3.19672131147541e-05, "epoch": 0.3991811668372569, "percentage": 7.99, "elapsed_time": "0:59:57", "remaining_time": "11:30:20"}
+{"current_steps": 196, "total_steps": 2440, "loss": 0.5048, "lr": 3.213114754098361e-05, "epoch": 0.40122824974411464, "percentage": 8.03, "elapsed_time": "1:00:16", "remaining_time": "11:30:04"}
+{"current_steps": 197, "total_steps": 2440, "loss": 0.5696, "lr": 3.2295081967213116e-05, "epoch": 0.4032753326509724, "percentage": 8.07, "elapsed_time": "1:00:31", "remaining_time": "11:29:08"}
+{"current_steps": 198, "total_steps": 2440, "loss": 0.5175, "lr": 3.245901639344263e-05, "epoch": 0.4053224155578301, "percentage": 8.11, "elapsed_time": "1:00:48", "remaining_time": "11:28:35"}
+{"current_steps": 199, "total_steps": 2440, "loss": 0.5426, "lr": 3.2622950819672136e-05, "epoch": 0.40736949846468784, "percentage": 8.16, "elapsed_time": "1:01:04", "remaining_time": "11:27:48"}
+{"current_steps": 200, "total_steps": 2440, "loss": 0.5087, "lr": 3.278688524590164e-05, "epoch": 0.4094165813715456, "percentage": 8.2, "elapsed_time": "1:01:28", "remaining_time": "11:28:32"}
+{"current_steps": 201, "total_steps": 2440, "loss": 0.509, "lr": 3.295081967213115e-05, "epoch": 0.41146366427840325, "percentage": 8.24, "elapsed_time": "1:01:48", "remaining_time": "11:28:33"}
+{"current_steps": 202, "total_steps": 2440, "loss": 0.5193, "lr": 3.311475409836066e-05, "epoch": 0.413510747185261, "percentage": 8.28, "elapsed_time": "1:02:07", "remaining_time": "11:28:17"}
+{"current_steps": 203, "total_steps": 2440, "loss": 0.5197, "lr": 3.327868852459017e-05, "epoch": 0.4155578300921187, "percentage": 8.32, "elapsed_time": "1:02:28", "remaining_time": "11:28:30"}
+{"current_steps": 204, "total_steps": 2440, "loss": 0.5044, "lr": 3.3442622950819675e-05, "epoch": 0.41760491299897645, "percentage": 8.36, "elapsed_time": "1:02:45", "remaining_time": "11:27:49"}
+{"current_steps": 205, "total_steps": 2440, "loss": 0.4974, "lr": 3.360655737704918e-05, "epoch": 0.4196519959058342, "percentage": 8.4, "elapsed_time": "1:03:07", "remaining_time": "11:28:13"}
+{"current_steps": 206, "total_steps": 2440, "loss": 0.5017, "lr": 3.3770491803278695e-05, "epoch": 0.4216990788126919, "percentage": 8.44, "elapsed_time": "1:03:28", "remaining_time": "11:28:24"}
+{"current_steps": 207, "total_steps": 2440, "loss": 0.4983, "lr": 3.39344262295082e-05, "epoch": 0.42374616171954965, "percentage": 8.48, "elapsed_time": "1:03:45", "remaining_time": "11:27:52"}
+{"current_steps": 208, "total_steps": 2440, "loss": 0.5484, "lr": 3.409836065573771e-05, "epoch": 0.4257932446264074, "percentage": 8.52, "elapsed_time": "1:04:04", "remaining_time": "11:27:34"}
+{"current_steps": 209, "total_steps": 2440, "loss": 0.5152, "lr": 3.4262295081967214e-05, "epoch": 0.4278403275332651, "percentage": 8.57, "elapsed_time": "1:04:18", "remaining_time": "11:26:24"}
+{"current_steps": 210, "total_steps": 2440, "loss": 0.529, "lr": 3.442622950819672e-05, "epoch": 0.42988741044012285, "percentage": 8.61, "elapsed_time": "1:04:37", "remaining_time": "11:26:18"}
+{"current_steps": 211, "total_steps": 2440, "loss": 0.547, "lr": 3.4590163934426234e-05, "epoch": 0.43193449334698053, "percentage": 8.65, "elapsed_time": "1:04:56", "remaining_time": "11:26:06"}
+{"current_steps": 212, "total_steps": 2440, "loss": 0.5255, "lr": 3.475409836065574e-05, "epoch": 0.43398157625383826, "percentage": 8.69, "elapsed_time": "1:05:17", "remaining_time": "11:26:09"}
+{"current_steps": 213, "total_steps": 2440, "loss": 0.5407, "lr": 3.491803278688525e-05, "epoch": 0.436028659160696, "percentage": 8.73, "elapsed_time": "1:05:38", "remaining_time": "11:26:15"}
+{"current_steps": 214, "total_steps": 2440, "loss": 0.5467, "lr": 3.5081967213114754e-05, "epoch": 0.43807574206755373, "percentage": 8.77, "elapsed_time": "1:05:55", "remaining_time": "11:25:48"}
+{"current_steps": 215, "total_steps": 2440, "loss": 0.5599, "lr": 3.524590163934427e-05, "epoch": 0.44012282497441146, "percentage": 8.81, "elapsed_time": "1:06:13", "remaining_time": "11:25:26"}
+{"current_steps": 216, "total_steps": 2440, "loss": 0.5127, "lr": 3.5409836065573773e-05, "epoch": 0.4421699078812692, "percentage": 8.85, "elapsed_time": "1:06:32", "remaining_time": "11:25:12"}
+{"current_steps": 217, "total_steps": 2440, "loss": 0.5186, "lr": 3.557377049180329e-05, "epoch": 0.44421699078812693, "percentage": 8.89, "elapsed_time": "1:06:49", "remaining_time": "11:24:36"}
+{"current_steps": 218, "total_steps": 2440, "loss": 0.4989, "lr": 3.5737704918032786e-05, "epoch": 0.44626407369498466, "percentage": 8.93, "elapsed_time": "1:07:08", "remaining_time": "11:24:19"}
+{"current_steps": 219, "total_steps": 2440, "loss": 0.5008, "lr": 3.59016393442623e-05, "epoch": 0.4483111566018424, "percentage": 8.98, "elapsed_time": "1:07:28", "remaining_time": "11:24:21"}
+{"current_steps": 220, "total_steps": 2440, "loss": 0.5122, "lr": 3.6065573770491806e-05, "epoch": 0.4503582395087001, "percentage": 9.02, "elapsed_time": "1:07:48", "remaining_time": "11:24:19"}
+{"current_steps": 221, "total_steps": 2440, "loss": 0.5018, "lr": 3.622950819672131e-05, "epoch": 0.4524053224155578, "percentage": 9.06, "elapsed_time": "1:08:10", "remaining_time": "11:24:34"}
+{"current_steps": 222, "total_steps": 2440, "loss": 0.4921, "lr": 3.6393442622950826e-05, "epoch": 0.45445240532241554, "percentage": 9.1, "elapsed_time": "1:08:27", "remaining_time": "11:24:00"}
+{"current_steps": 223, "total_steps": 2440, "loss": 0.5041, "lr": 3.655737704918033e-05, "epoch": 0.4564994882292733, "percentage": 9.14, "elapsed_time": "1:08:44", "remaining_time": "11:23:24"}
+{"current_steps": 224, "total_steps": 2440, "loss": 0.5281, "lr": 3.672131147540984e-05, "epoch": 0.458546571136131, "percentage": 9.18, "elapsed_time": "1:09:03", "remaining_time": "11:23:10"}
+{"current_steps": 225, "total_steps": 2440, "loss": 0.4862, "lr": 3.6885245901639346e-05, "epoch": 0.46059365404298874, "percentage": 9.22, "elapsed_time": "1:09:18", "remaining_time": "11:22:21"}
+{"current_steps": 226, "total_steps": 2440, "loss": 0.5313, "lr": 3.704918032786886e-05, "epoch": 0.4626407369498465, "percentage": 9.26, "elapsed_time": "1:09:40", "remaining_time": "11:22:32"}
+{"current_steps": 227, "total_steps": 2440, "loss": 0.5611, "lr": 3.7213114754098365e-05, "epoch": 0.4646878198567042, "percentage": 9.3, "elapsed_time": "1:10:00", "remaining_time": "11:22:27"}
+{"current_steps": 228, "total_steps": 2440, "loss": 0.5017, "lr": 3.737704918032787e-05, "epoch": 0.46673490276356194, "percentage": 9.34, "elapsed_time": "1:10:20", "remaining_time": "11:22:26"}
+{"current_steps": 229, "total_steps": 2440, "loss": 0.4801, "lr": 3.754098360655738e-05, "epoch": 0.4687819856704197, "percentage": 9.39, "elapsed_time": "1:10:38", "remaining_time": "11:22:06"}
+{"current_steps": 230, "total_steps": 2440, "loss": 0.5265, "lr": 3.770491803278689e-05, "epoch": 0.47082906857727735, "percentage": 9.43, "elapsed_time": "1:10:58", "remaining_time": "11:22:02"}
+{"current_steps": 231, "total_steps": 2440, "loss": 0.533, "lr": 3.78688524590164e-05, "epoch": 0.4728761514841351, "percentage": 9.47, "elapsed_time": "1:11:21", "remaining_time": "11:22:18"}
+{"current_steps": 232, "total_steps": 2440, "loss": 0.5067, "lr": 3.8032786885245905e-05, "epoch": 0.4749232343909928, "percentage": 9.51, "elapsed_time": "1:11:37", "remaining_time": "11:21:44"}
+{"current_steps": 233, "total_steps": 2440, "loss": 0.5572, "lr": 3.819672131147541e-05, "epoch": 0.47697031729785055, "percentage": 9.55, "elapsed_time": "1:11:57", "remaining_time": "11:21:37"}
+{"current_steps": 234, "total_steps": 2440, "loss": 0.5267, "lr": 3.836065573770492e-05, "epoch": 0.4790174002047083, "percentage": 9.59, "elapsed_time": "1:12:16", "remaining_time": "11:21:17"}
+{"current_steps": 235, "total_steps": 2440, "loss": 0.5007, "lr": 3.852459016393443e-05, "epoch": 0.481064483111566, "percentage": 9.63, "elapsed_time": "1:12:39", "remaining_time": "11:21:40"}
+{"current_steps": 236, "total_steps": 2440, "loss": 0.5465, "lr": 3.868852459016394e-05, "epoch": 0.48311156601842375, "percentage": 9.67, "elapsed_time": "1:13:00", "remaining_time": "11:21:49"}
+{"current_steps": 237, "total_steps": 2440, "loss": 0.5498, "lr": 3.8852459016393444e-05, "epoch": 0.4851586489252815, "percentage": 9.71, "elapsed_time": "1:13:18", "remaining_time": "11:21:23"}
+{"current_steps": 238, "total_steps": 2440, "loss": 0.5032, "lr": 3.901639344262295e-05, "epoch": 0.4872057318321392, "percentage": 9.75, "elapsed_time": "1:13:35", "remaining_time": "11:20:56"}
+{"current_steps": 239, "total_steps": 2440, "loss": 0.5351, "lr": 3.9180327868852464e-05, "epoch": 0.48925281473899696, "percentage": 9.8, "elapsed_time": "1:13:53", "remaining_time": "11:20:32"}
+{"current_steps": 240, "total_steps": 2440, "loss": 0.5511, "lr": 3.934426229508197e-05, "epoch": 0.49129989764585463, "percentage": 9.84, "elapsed_time": "1:14:10", "remaining_time": "11:19:58"}
+{"current_steps": 241, "total_steps": 2440, "loss": 0.5095, "lr": 3.950819672131148e-05, "epoch": 0.49334698055271237, "percentage": 9.88, "elapsed_time": "1:14:31", "remaining_time": "11:19:59"}
+{"current_steps": 242, "total_steps": 2440, "loss": 0.4903, "lr": 3.9672131147540983e-05, "epoch": 0.4953940634595701, "percentage": 9.92, "elapsed_time": "1:14:48", "remaining_time": "11:19:29"}
+{"current_steps": 243, "total_steps": 2440, "loss": 0.5246, "lr": 3.98360655737705e-05, "epoch": 0.49744114636642783, "percentage": 9.96, "elapsed_time": "1:15:04", "remaining_time": "11:18:44"}
+{"current_steps": 244, "total_steps": 2440, "loss": 0.527, "lr": 4e-05, "epoch": 0.49948822927328557, "percentage": 10.0, "elapsed_time": "1:15:24", "remaining_time": "11:18:38"}
+{"current_steps": 245, "total_steps": 2440, "loss": 0.4698, "lr": 3.999997953390434e-05, "epoch": 0.5015353121801432, "percentage": 10.04, "elapsed_time": "1:15:42", "remaining_time": "11:18:19"}
+{"current_steps": 246, "total_steps": 2440, "loss": 0.5021, "lr": 3.999991813565924e-05, "epoch": 0.503582395087001, "percentage": 10.08, "elapsed_time": "1:16:02", "remaining_time": "11:18:15"}
+{"current_steps": 247, "total_steps": 2440, "loss": 0.5195, "lr": 3.999981580539036e-05, "epoch": 0.5056294779938587, "percentage": 10.12, "elapsed_time": "1:16:18", "remaining_time": "11:17:26"}
+{"current_steps": 248, "total_steps": 2440, "loss": 0.4915, "lr": 3.999967254330713e-05, "epoch": 0.5076765609007164, "percentage": 10.16, "elapsed_time": "1:16:38", "remaining_time": "11:17:23"}
+{"current_steps": 249, "total_steps": 2440, "loss": 0.5395, "lr": 3.999948834970275e-05, "epoch": 0.5097236438075742, "percentage": 10.2, "elapsed_time": "1:16:57", "remaining_time": "11:17:09"}
+{"current_steps": 250, "total_steps": 2440, "loss": 0.5156, "lr": 3.9999263224954204e-05, "epoch": 0.5117707267144319, "percentage": 10.25, "elapsed_time": "1:17:15", "remaining_time": "11:16:51"}
+{"current_steps": 251, "total_steps": 2440, "loss": 0.505, "lr": 3.999899716952221e-05, "epoch": 0.5138178096212896, "percentage": 10.29, "elapsed_time": "1:17:33", "remaining_time": "11:16:21"}
+{"current_steps": 252, "total_steps": 2440, "loss": 0.517, "lr": 3.9998690183951304e-05, "epoch": 0.5158648925281474, "percentage": 10.33, "elapsed_time": "1:17:50", "remaining_time": "11:15:50"}
+{"current_steps": 253, "total_steps": 2440, "loss": 0.5209, "lr": 3.999834226886976e-05, "epoch": 0.5179119754350051, "percentage": 10.37, "elapsed_time": "1:18:09", "remaining_time": "11:15:33"}
+{"current_steps": 254, "total_steps": 2440, "loss": 0.5144, "lr": 3.999795342498961e-05, "epoch": 0.5199590583418628, "percentage": 10.41, "elapsed_time": "1:18:23", "remaining_time": "11:14:40"}
+{"current_steps": 255, "total_steps": 2440, "loss": 0.5285, "lr": 3.999752365310668e-05, "epoch": 0.5220061412487206, "percentage": 10.45, "elapsed_time": "1:18:42", "remaining_time": "11:14:21"}
+{"current_steps": 256, "total_steps": 2440, "loss": 0.493, "lr": 3.999705295410054e-05, "epoch": 0.5240532241555783, "percentage": 10.49, "elapsed_time": "1:19:00", "remaining_time": "11:14:02"}
+{"current_steps": 257, "total_steps": 2440, "loss": 0.5257, "lr": 3.999654132893453e-05, "epoch": 0.526100307062436, "percentage": 10.53, "elapsed_time": "1:19:19", "remaining_time": "11:13:47"}
+{"current_steps": 258, "total_steps": 2440, "loss": 0.4947, "lr": 3.999598877865575e-05, "epoch": 0.5281473899692938, "percentage": 10.57, "elapsed_time": "1:19:37", "remaining_time": "11:13:23"}
+{"current_steps": 259, "total_steps": 2440, "loss": 0.5319, "lr": 3.999539530439504e-05, "epoch": 0.5301944728761515, "percentage": 10.61, "elapsed_time": "1:19:57", "remaining_time": "11:13:22"}
+{"current_steps": 260, "total_steps": 2440, "loss": 0.5239, "lr": 3.9994760907367025e-05, "epoch": 0.5322415557830092, "percentage": 10.66, "elapsed_time": "1:20:17", "remaining_time": "11:13:15"}
+{"current_steps": 261, "total_steps": 2440, "loss": 0.5182, "lr": 3.999408558887006e-05, "epoch": 0.534288638689867, "percentage": 10.7, "elapsed_time": "1:20:35", "remaining_time": "11:12:53"}
+{"current_steps": 262, "total_steps": 2440, "loss": 0.5211, "lr": 3.9993369350286265e-05, "epoch": 0.5363357215967247, "percentage": 10.74, "elapsed_time": "1:20:57", "remaining_time": "11:13:00"}
+{"current_steps": 263, "total_steps": 2440, "loss": 0.4922, "lr": 3.999261219308149e-05, "epoch": 0.5383828045035824, "percentage": 10.78, "elapsed_time": "1:21:16", "remaining_time": "11:12:49"}
+{"current_steps": 264, "total_steps": 2440, "loss": 0.4833, "lr": 3.999181411880536e-05, "epoch": 0.5404298874104401, "percentage": 10.82, "elapsed_time": "1:21:33", "remaining_time": "11:12:13"}
+{"current_steps": 265, "total_steps": 2440, "loss": 0.5156, "lr": 3.99909751290912e-05, "epoch": 0.5424769703172978, "percentage": 10.86, "elapsed_time": "1:21:54", "remaining_time": "11:12:12"}
+{"current_steps": 266, "total_steps": 2440, "loss": 0.4918, "lr": 3.9990095225656104e-05, "epoch": 0.5445240532241555, "percentage": 10.9, "elapsed_time": "1:22:12", "remaining_time": "11:11:55"}
+{"current_steps": 267, "total_steps": 2440, "loss": 0.534, "lr": 3.998917441030089e-05, "epoch": 0.5465711361310133, "percentage": 10.94, "elapsed_time": "1:22:32", "remaining_time": "11:11:48"}
+{"current_steps": 268, "total_steps": 2440, "loss": 0.4919, "lr": 3.9988212684910107e-05, "epoch": 0.548618219037871, "percentage": 10.98, "elapsed_time": "1:22:47", "remaining_time": "11:11:00"}
+{"current_steps": 269, "total_steps": 2440, "loss": 0.5023, "lr": 3.998721005145204e-05, "epoch": 0.5506653019447287, "percentage": 11.02, "elapsed_time": "1:23:06", "remaining_time": "11:10:47"}
+{"current_steps": 270, "total_steps": 2440, "loss": 0.5123, "lr": 3.998616651197867e-05, "epoch": 0.5527123848515865, "percentage": 11.07, "elapsed_time": "1:23:23", "remaining_time": "11:10:15"}
+{"current_steps": 271, "total_steps": 2440, "loss": 0.5192, "lr": 3.9985082068625724e-05, "epoch": 0.5547594677584442, "percentage": 11.11, "elapsed_time": "1:23:39", "remaining_time": "11:09:37"}
+{"current_steps": 272, "total_steps": 2440, "loss": 0.5159, "lr": 3.998395672361264e-05, "epoch": 0.5568065506653019, "percentage": 11.15, "elapsed_time": "1:23:58", "remaining_time": "11:09:20"}
+{"current_steps": 273, "total_steps": 2440, "loss": 0.4571, "lr": 3.998279047924255e-05, "epoch": 0.5588536335721597, "percentage": 11.19, "elapsed_time": "1:24:16", "remaining_time": "11:09:00"}
+{"current_steps": 274, "total_steps": 2440, "loss": 0.5093, "lr": 3.998158333790231e-05, "epoch": 0.5609007164790174, "percentage": 11.23, "elapsed_time": "1:24:37", "remaining_time": "11:08:54"}
+{"current_steps": 275, "total_steps": 2440, "loss": 0.5174, "lr": 3.998033530206246e-05, "epoch": 0.5629477993858751, "percentage": 11.27, "elapsed_time": "1:24:55", "remaining_time": "11:08:32"}
+{"current_steps": 276, "total_steps": 2440, "loss": 0.5292, "lr": 3.9979046374277246e-05, "epoch": 0.5649948822927329, "percentage": 11.31, "elapsed_time": "1:25:15", "remaining_time": "11:08:29"}
+{"current_steps": 277, "total_steps": 2440, "loss": 0.5202, "lr": 3.99777165571846e-05, "epoch": 0.5670419651995906, "percentage": 11.35, "elapsed_time": "1:25:35", "remaining_time": "11:08:20"}
+{"current_steps": 278, "total_steps": 2440, "loss": 0.515, "lr": 3.997634585350614e-05, "epoch": 0.5690890481064483, "percentage": 11.39, "elapsed_time": "1:25:53", "remaining_time": "11:07:55"}
+{"current_steps": 279, "total_steps": 2440, "loss": 0.4827, "lr": 3.997493426604715e-05, "epoch": 0.5711361310133061, "percentage": 11.43, "elapsed_time": "1:26:09", "remaining_time": "11:07:22"}
+{"current_steps": 280, "total_steps": 2440, "loss": 0.4984, "lr": 3.997348179769661e-05, "epoch": 0.5731832139201638, "percentage": 11.48, "elapsed_time": "1:26:26", "remaining_time": "11:06:50"}
+{"current_steps": 281, "total_steps": 2440, "loss": 0.4795, "lr": 3.9971988451427155e-05, "epoch": 0.5752302968270215, "percentage": 11.52, "elapsed_time": "1:26:43", "remaining_time": "11:06:18"}
+{"current_steps": 282, "total_steps": 2440, "loss": 0.5288, "lr": 3.997045423029508e-05, "epoch": 0.5772773797338793, "percentage": 11.56, "elapsed_time": "1:26:59", "remaining_time": "11:05:45"}
+{"current_steps": 283, "total_steps": 2440, "loss": 0.5412, "lr": 3.996887913744033e-05, "epoch": 0.579324462640737, "percentage": 11.6, "elapsed_time": "1:27:20", "remaining_time": "11:05:41"}
+{"current_steps": 284, "total_steps": 2440, "loss": 0.5119, "lr": 3.996726317608652e-05, "epoch": 0.5813715455475946, "percentage": 11.64, "elapsed_time": "1:27:37", "remaining_time": "11:05:13"}
+{"current_steps": 285, "total_steps": 2440, "loss": 0.5504, "lr": 3.996560634954088e-05, "epoch": 0.5834186284544524, "percentage": 11.68, "elapsed_time": "1:27:57", "remaining_time": "11:05:06"}
+{"current_steps": 286, "total_steps": 2440, "loss": 0.5323, "lr": 3.9963908661194285e-05, "epoch": 0.5854657113613101, "percentage": 11.72, "elapsed_time": "1:28:16", "remaining_time": "11:04:50"}
+{"current_steps": 287, "total_steps": 2440, "loss": 0.5086, "lr": 3.9962170114521246e-05, "epoch": 0.5875127942681678, "percentage": 11.76, "elapsed_time": "1:28:35", "remaining_time": "11:04:33"}
+{"current_steps": 288, "total_steps": 2440, "loss": 0.5451, "lr": 3.996039071307989e-05, "epoch": 0.5895598771750256, "percentage": 11.8, "elapsed_time": "1:28:53", "remaining_time": "11:04:13"}
+{"current_steps": 289, "total_steps": 2440, "loss": 0.5375, "lr": 3.995857046051196e-05, "epoch": 0.5916069600818833, "percentage": 11.84, "elapsed_time": "1:29:10", "remaining_time": "11:03:40"}
+{"current_steps": 290, "total_steps": 2440, "loss": 0.5241, "lr": 3.995670936054279e-05, "epoch": 0.593654042988741, "percentage": 11.89, "elapsed_time": "1:29:33", "remaining_time": "11:03:58"}
+{"current_steps": 291, "total_steps": 2440, "loss": 0.5049, "lr": 3.9954807416981335e-05, "epoch": 0.5957011258955988, "percentage": 11.93, "elapsed_time": "1:29:52", "remaining_time": "11:03:40"}
+{"current_steps": 292, "total_steps": 2440, "loss": 0.5117, "lr": 3.995286463372013e-05, "epoch": 0.5977482088024565, "percentage": 11.97, "elapsed_time": "1:30:11", "remaining_time": "11:03:28"}
+{"current_steps": 293, "total_steps": 2440, "loss": 0.5053, "lr": 3.9950881014735295e-05, "epoch": 0.5997952917093142, "percentage": 12.01, "elapsed_time": "1:30:27", "remaining_time": "11:02:48"}
+{"current_steps": 294, "total_steps": 2440, "loss": 0.4948, "lr": 3.994885656408651e-05, "epoch": 0.601842374616172, "percentage": 12.05, "elapsed_time": "1:30:43", "remaining_time": "11:02:12"}
+{"current_steps": 295, "total_steps": 2440, "loss": 0.5566, "lr": 3.994679128591706e-05, "epoch": 0.6038894575230297, "percentage": 12.09, "elapsed_time": "1:31:04", "remaining_time": "11:02:12"}
+{"current_steps": 296, "total_steps": 2440, "loss": 0.53, "lr": 3.9944685184453746e-05, "epoch": 0.6059365404298874, "percentage": 12.13, "elapsed_time": "1:31:22", "remaining_time": "11:01:54"}
+{"current_steps": 297, "total_steps": 2440, "loss": 0.5314, "lr": 3.994253826400693e-05, "epoch": 0.6079836233367452, "percentage": 12.17, "elapsed_time": "1:31:41", "remaining_time": "11:01:35"}
+{"current_steps": 298, "total_steps": 2440, "loss": 0.5116, "lr": 3.9940350528970535e-05, "epoch": 0.6100307062436029, "percentage": 12.21, "elapsed_time": "1:31:58", "remaining_time": "11:01:07"}
+{"current_steps": 299, "total_steps": 2440, "loss": 0.5028, "lr": 3.993812198382199e-05, "epoch": 0.6120777891504606, "percentage": 12.25, "elapsed_time": "1:32:12", "remaining_time": "11:00:14"}
+{"current_steps": 300, "total_steps": 2440, "loss": 0.5202, "lr": 3.993585263312227e-05, "epoch": 0.6141248720573184, "percentage": 12.3, "elapsed_time": "1:32:33", "remaining_time": "11:00:16"}
+{"current_steps": 301, "total_steps": 2440, "loss": 0.4912, "lr": 3.993354248151583e-05, "epoch": 0.6161719549641761, "percentage": 12.34, "elapsed_time": "1:32:50", "remaining_time": "10:59:45"}
+{"current_steps": 302, "total_steps": 2440, "loss": 0.5438, "lr": 3.993119153373067e-05, "epoch": 0.6182190378710338, "percentage": 12.38, "elapsed_time": "1:33:07", "remaining_time": "10:59:14"}
+{"current_steps": 303, "total_steps": 2440, "loss": 0.5123, "lr": 3.992879979457824e-05, "epoch": 0.6202661207778914, "percentage": 12.42, "elapsed_time": "1:33:25", "remaining_time": "10:58:54"}
+{"current_steps": 304, "total_steps": 2440, "loss": 0.5047, "lr": 3.9926367268953514e-05, "epoch": 0.6223132036847492, "percentage": 12.46, "elapsed_time": "1:33:43", "remaining_time": "10:58:34"}
+{"current_steps": 305, "total_steps": 2440, "loss": 0.5126, "lr": 3.9923893961834914e-05, "epoch": 0.6243602865916069, "percentage": 12.5, "elapsed_time": "1:34:03", "remaining_time": "10:58:24"}
+{"current_steps": 306, "total_steps": 2440, "loss": 0.512, "lr": 3.992137987828434e-05, "epoch": 0.6264073694984647, "percentage": 12.54, "elapsed_time": "1:34:22", "remaining_time": "10:58:06"}
+{"current_steps": 307, "total_steps": 2440, "loss": 0.5086, "lr": 3.991882502344712e-05, "epoch": 0.6284544524053224, "percentage": 12.58, "elapsed_time": "1:34:38", "remaining_time": "10:57:36"}
+{"current_steps": 308, "total_steps": 2440, "loss": 0.4928, "lr": 3.991622940255208e-05, "epoch": 0.6305015353121801, "percentage": 12.62, "elapsed_time": "1:34:54", "remaining_time": "10:56:55"}
+{"current_steps": 309, "total_steps": 2440, "loss": 0.4823, "lr": 3.991359302091141e-05, "epoch": 0.6325486182190379, "percentage": 12.66, "elapsed_time": "1:35:08", "remaining_time": "10:56:07"}
+{"current_steps": 310, "total_steps": 2440, "loss": 0.553, "lr": 3.991091588392077e-05, "epoch": 0.6345957011258956, "percentage": 12.7, "elapsed_time": "1:35:25", "remaining_time": "10:55:36"}
+{"current_steps": 311, "total_steps": 2440, "loss": 0.5191, "lr": 3.99081979970592e-05, "epoch": 0.6366427840327533, "percentage": 12.75, "elapsed_time": "1:35:45", "remaining_time": "10:55:30"}
+{"current_steps": 312, "total_steps": 2440, "loss": 0.4833, "lr": 3.9905439365889176e-05, "epoch": 0.638689866939611, "percentage": 12.79, "elapsed_time": "1:36:03", "remaining_time": "10:55:09"}
+{"current_steps": 313, "total_steps": 2440, "loss": 0.4932, "lr": 3.990263999605652e-05, "epoch": 0.6407369498464688, "percentage": 12.83, "elapsed_time": "1:36:20", "remaining_time": "10:54:43"}
+{"current_steps": 314, "total_steps": 2440, "loss": 0.5475, "lr": 3.989979989329046e-05, "epoch": 0.6427840327533265, "percentage": 12.87, "elapsed_time": "1:36:42", "remaining_time": "10:54:49"}
+{"current_steps": 315, "total_steps": 2440, "loss": 0.5656, "lr": 3.9896919063403567e-05, "epoch": 0.6448311156601843, "percentage": 12.91, "elapsed_time": "1:37:00", "remaining_time": "10:54:27"}
+{"current_steps": 316, "total_steps": 2440, "loss": 0.4812, "lr": 3.989399751229179e-05, "epoch": 0.646878198567042, "percentage": 12.95, "elapsed_time": "1:37:20", "remaining_time": "10:54:17"}
+{"current_steps": 317, "total_steps": 2440, "loss": 0.5448, "lr": 3.989103524593439e-05, "epoch": 0.6489252814738997, "percentage": 12.99, "elapsed_time": "1:37:37", "remaining_time": "10:53:48"}
+{"current_steps": 318, "total_steps": 2440, "loss": 0.5391, "lr": 3.9888032270393966e-05, "epoch": 0.6509723643807575, "percentage": 13.03, "elapsed_time": "1:37:57", "remaining_time": "10:53:38"}
+{"current_steps": 319, "total_steps": 2440, "loss": 0.5857, "lr": 3.988498859181645e-05, "epoch": 0.6530194472876152, "percentage": 13.07, "elapsed_time": "1:38:17", "remaining_time": "10:53:29"}
+{"current_steps": 320, "total_steps": 2440, "loss": 0.4775, "lr": 3.988190421643105e-05, "epoch": 0.6550665301944729, "percentage": 13.11, "elapsed_time": "1:38:36", "remaining_time": "10:53:18"}
+{"current_steps": 321, "total_steps": 2440, "loss": 0.4953, "lr": 3.9878779150550306e-05, "epoch": 0.6571136131013307, "percentage": 13.16, "elapsed_time": "1:38:52", "remaining_time": "10:52:44"}
+{"current_steps": 322, "total_steps": 2440, "loss": 0.4593, "lr": 3.9875613400569975e-05, "epoch": 0.6591606960081884, "percentage": 13.2, "elapsed_time": "1:39:10", "remaining_time": "10:52:19"}
+{"current_steps": 323, "total_steps": 2440, "loss": 0.4943, "lr": 3.987240697296912e-05, "epoch": 0.661207778915046, "percentage": 13.24, "elapsed_time": "1:39:25", "remaining_time": "10:51:40"}
+{"current_steps": 324, "total_steps": 2440, "loss": 0.5312, "lr": 3.986915987431006e-05, "epoch": 0.6632548618219037, "percentage": 13.28, "elapsed_time": "1:39:43", "remaining_time": "10:51:15"}
+{"current_steps": 325, "total_steps": 2440, "loss": 0.5066, "lr": 3.986587211123833e-05, "epoch": 0.6653019447287615, "percentage": 13.32, "elapsed_time": "1:39:57", "remaining_time": "10:50:30"}
+{"current_steps": 326, "total_steps": 2440, "loss": 0.519, "lr": 3.986254369048268e-05, "epoch": 0.6673490276356192, "percentage": 13.36, "elapsed_time": "1:40:16", "remaining_time": "10:50:17"}
+{"current_steps": 327, "total_steps": 2440, "loss": 0.526, "lr": 3.985917461885512e-05, "epoch": 0.6693961105424769, "percentage": 13.4, "elapsed_time": "1:40:36", "remaining_time": "10:50:05"}
+{"current_steps": 328, "total_steps": 2440, "loss": 0.4941, "lr": 3.98557649032508e-05, "epoch": 0.6714431934493347, "percentage": 13.44, "elapsed_time": "1:40:53", "remaining_time": "10:49:37"}
+{"current_steps": 329, "total_steps": 2440, "loss": 0.5161, "lr": 3.985231455064809e-05, "epoch": 0.6734902763561924, "percentage": 13.48, "elapsed_time": "1:41:13", "remaining_time": "10:49:32"}
+{"current_steps": 330, "total_steps": 2440, "loss": 0.5252, "lr": 3.9848823568108515e-05, "epoch": 0.6755373592630501, "percentage": 13.52, "elapsed_time": "1:41:33", "remaining_time": "10:49:19"}
+{"current_steps": 331, "total_steps": 2440, "loss": 0.5153, "lr": 3.984529196277674e-05, "epoch": 0.6775844421699079, "percentage": 13.57, "elapsed_time": "1:41:56", "remaining_time": "10:49:32"}
+{"current_steps": 332, "total_steps": 2440, "loss": 0.5136, "lr": 3.9841719741880583e-05, "epoch": 0.6796315250767656, "percentage": 13.61, "elapsed_time": "1:42:19", "remaining_time": "10:49:39"}
+{"current_steps": 333, "total_steps": 2440, "loss": 0.5056, "lr": 3.9838106912731e-05, "epoch": 0.6816786079836233, "percentage": 13.65, "elapsed_time": "1:42:30", "remaining_time": "10:48:36"}
+{"current_steps": 334, "total_steps": 2440, "loss": 0.5022, "lr": 3.983445348272203e-05, "epoch": 0.6837256908904811, "percentage": 13.69, "elapsed_time": "1:42:47", "remaining_time": "10:48:11"}
+{"current_steps": 335, "total_steps": 2440, "loss": 0.5063, "lr": 3.983075945933083e-05, "epoch": 0.6857727737973388, "percentage": 13.73, "elapsed_time": "1:43:06", "remaining_time": "10:47:51"}
+{"current_steps": 336, "total_steps": 2440, "loss": 0.5067, "lr": 3.9827024850117606e-05, "epoch": 0.6878198567041965, "percentage": 13.77, "elapsed_time": "1:43:28", "remaining_time": "10:47:55"}
+{"current_steps": 337, "total_steps": 2440, "loss": 0.5112, "lr": 3.982324966272566e-05, "epoch": 0.6898669396110543, "percentage": 13.81, "elapsed_time": "1:43:47", "remaining_time": "10:47:43"}
+{"current_steps": 338, "total_steps": 2440, "loss": 0.4987, "lr": 3.9819433904881324e-05, "epoch": 0.691914022517912, "percentage": 13.85, "elapsed_time": "1:44:08", "remaining_time": "10:47:40"}
+{"current_steps": 339, "total_steps": 2440, "loss": 0.5174, "lr": 3.981557758439396e-05, "epoch": 0.6939611054247697, "percentage": 13.89, "elapsed_time": "1:44:23", "remaining_time": "10:47:01"}
+{"current_steps": 340, "total_steps": 2440, "loss": 0.5205, "lr": 3.981168070915594e-05, "epoch": 0.6960081883316275, "percentage": 13.93, "elapsed_time": "1:44:41", "remaining_time": "10:46:39"}
+{"current_steps": 341, "total_steps": 2440, "loss": 0.5406, "lr": 3.980774328714267e-05, "epoch": 0.6980552712384852, "percentage": 13.98, "elapsed_time": "1:44:55", "remaining_time": "10:45:53"}
+{"current_steps": 342, "total_steps": 2440, "loss": 0.5491, "lr": 3.9803765326412506e-05, "epoch": 0.7001023541453428, "percentage": 14.02, "elapsed_time": "1:45:17", "remaining_time": "10:45:52"}
+{"current_steps": 343, "total_steps": 2440, "loss": 0.5507, "lr": 3.979974683510677e-05, "epoch": 0.7021494370522006, "percentage": 14.06, "elapsed_time": "1:45:38", "remaining_time": "10:45:53"}
+{"current_steps": 344, "total_steps": 2440, "loss": 0.5136, "lr": 3.9795687821449754e-05, "epoch": 0.7041965199590583, "percentage": 14.1, "elapsed_time": "1:45:57", "remaining_time": "10:45:37"}
+{"current_steps": 345, "total_steps": 2440, "loss": 0.4908, "lr": 3.9791588293748676e-05, "epoch": 0.706243602865916, "percentage": 14.14, "elapsed_time": "1:46:17", "remaining_time": "10:45:26"}
+{"current_steps": 346, "total_steps": 2440, "loss": 0.5046, "lr": 3.978744826039366e-05, "epoch": 0.7082906857727738, "percentage": 14.18, "elapsed_time": "1:46:36", "remaining_time": "10:45:12"}
+{"current_steps": 347, "total_steps": 2440, "loss": 0.4994, "lr": 3.9783267729857756e-05, "epoch": 0.7103377686796315, "percentage": 14.22, "elapsed_time": "1:46:57", "remaining_time": "10:45:09"}
+{"current_steps": 348, "total_steps": 2440, "loss": 0.4813, "lr": 3.9779046710696854e-05, "epoch": 0.7123848515864892, "percentage": 14.26, "elapsed_time": "1:47:14", "remaining_time": "10:44:40"}
+{"current_steps": 349, "total_steps": 2440, "loss": 0.487, "lr": 3.977478521154974e-05, "epoch": 0.714431934493347, "percentage": 14.3, "elapsed_time": "1:47:33", "remaining_time": "10:44:23"}
+{"current_steps": 350, "total_steps": 2440, "loss": 0.5418, "lr": 3.977048324113805e-05, "epoch": 0.7164790174002047, "percentage": 14.34, "elapsed_time": "1:47:51", "remaining_time": "10:44:03"}
+{"current_steps": 351, "total_steps": 2440, "loss": 0.5313, "lr": 3.976614080826623e-05, "epoch": 0.7185261003070624, "percentage": 14.39, "elapsed_time": "1:48:07", "remaining_time": "10:43:28"}
+{"current_steps": 352, "total_steps": 2440, "loss": 0.5136, "lr": 3.9761757921821544e-05, "epoch": 0.7205731832139202, "percentage": 14.43, "elapsed_time": "1:48:23", "remaining_time": "10:42:57"}
+{"current_steps": 353, "total_steps": 2440, "loss": 0.5396, "lr": 3.975733459077405e-05, "epoch": 0.7226202661207779, "percentage": 14.47, "elapsed_time": "1:48:44", "remaining_time": "10:42:54"}
+{"current_steps": 354, "total_steps": 2440, "loss": 0.4912, "lr": 3.9752870824176585e-05, "epoch": 0.7246673490276356, "percentage": 14.51, "elapsed_time": "1:49:01", "remaining_time": "10:42:26"}
+{"current_steps": 355, "total_steps": 2440, "loss": 0.5246, "lr": 3.974836663116472e-05, "epoch": 0.7267144319344934, "percentage": 14.55, "elapsed_time": "1:49:25", "remaining_time": "10:42:40"}
+{"current_steps": 356, "total_steps": 2440, "loss": 0.4697, "lr": 3.97438220209568e-05, "epoch": 0.7287615148413511, "percentage": 14.59, "elapsed_time": "1:49:47", "remaining_time": "10:42:42"}
+{"current_steps": 357, "total_steps": 2440, "loss": 0.5033, "lr": 3.973923700285386e-05, "epoch": 0.7308085977482088, "percentage": 14.63, "elapsed_time": "1:50:09", "remaining_time": "10:42:43"}
+{"current_steps": 358, "total_steps": 2440, "loss": 0.51, "lr": 3.973461158623963e-05, "epoch": 0.7328556806550666, "percentage": 14.67, "elapsed_time": "1:50:30", "remaining_time": "10:42:40"}
+{"current_steps": 359, "total_steps": 2440, "loss": 0.5183, "lr": 3.972994578058055e-05, "epoch": 0.7349027635619243, "percentage": 14.71, "elapsed_time": "1:50:50", "remaining_time": "10:42:29"}
+{"current_steps": 360, "total_steps": 2440, "loss": 0.4791, "lr": 3.972523959542569e-05, "epoch": 0.736949846468782, "percentage": 14.75, "elapsed_time": "1:51:08", "remaining_time": "10:42:08"}
+{"current_steps": 361, "total_steps": 2440, "loss": 0.5053, "lr": 3.9720493040406786e-05, "epoch": 0.7389969293756398, "percentage": 14.8, "elapsed_time": "1:51:28", "remaining_time": "10:41:59"}
+{"current_steps": 362, "total_steps": 2440, "loss": 0.4902, "lr": 3.9715706125238164e-05, "epoch": 0.7410440122824974, "percentage": 14.84, "elapsed_time": "1:51:49", "remaining_time": "10:41:53"}
+{"current_steps": 363, "total_steps": 2440, "loss": 0.5002, "lr": 3.971087885971679e-05, "epoch": 0.7430910951893551, "percentage": 14.88, "elapsed_time": "1:52:05", "remaining_time": "10:41:22"}
+{"current_steps": 364, "total_steps": 2440, "loss": 0.5077, "lr": 3.970601125372218e-05, "epoch": 0.7451381780962129, "percentage": 14.92, "elapsed_time": "1:52:22", "remaining_time": "10:40:54"}
+{"current_steps": 365, "total_steps": 2440, "loss": 0.5152, "lr": 3.970110331721643e-05, "epoch": 0.7471852610030706, "percentage": 14.96, "elapsed_time": "1:52:41", "remaining_time": "10:40:39"}
+{"current_steps": 366, "total_steps": 2440, "loss": 0.5016, "lr": 3.9696155060244166e-05, "epoch": 0.7492323439099283, "percentage": 15.0, "elapsed_time": "1:52:58", "remaining_time": "10:40:10"}
+{"current_steps": 367, "total_steps": 2440, "loss": 0.5484, "lr": 3.9691166492932535e-05, "epoch": 0.7512794268167861, "percentage": 15.04, "elapsed_time": "1:53:14", "remaining_time": "10:39:36"}
+{"current_steps": 368, "total_steps": 2440, "loss": 0.4839, "lr": 3.968613762549119e-05, "epoch": 0.7533265097236438, "percentage": 15.08, "elapsed_time": "1:53:30", "remaining_time": "10:39:05"}
+{"current_steps": 369, "total_steps": 2440, "loss": 0.4961, "lr": 3.968106846821226e-05, "epoch": 0.7553735926305015, "percentage": 15.12, "elapsed_time": "1:53:52", "remaining_time": "10:39:08"}
+{"current_steps": 370, "total_steps": 2440, "loss": 0.5347, "lr": 3.9675959031470336e-05, "epoch": 0.7574206755373593, "percentage": 15.16, "elapsed_time": "1:54:11", "remaining_time": "10:38:50"}
+{"current_steps": 371, "total_steps": 2440, "loss": 0.5025, "lr": 3.9670809325722425e-05, "epoch": 0.759467758444217, "percentage": 15.2, "elapsed_time": "1:54:30", "remaining_time": "10:38:38"}
+{"current_steps": 372, "total_steps": 2440, "loss": 0.527, "lr": 3.966561936150797e-05, "epoch": 0.7615148413510747, "percentage": 15.25, "elapsed_time": "1:54:48", "remaining_time": "10:38:15"}
+{"current_steps": 373, "total_steps": 2440, "loss": 0.5024, "lr": 3.966038914944881e-05, "epoch": 0.7635619242579325, "percentage": 15.29, "elapsed_time": "1:55:08", "remaining_time": "10:38:01"}
+{"current_steps": 374, "total_steps": 2440, "loss": 0.4543, "lr": 3.9655118700249146e-05, "epoch": 0.7656090071647902, "percentage": 15.33, "elapsed_time": "1:55:26", "remaining_time": "10:37:41"}
+{"current_steps": 375, "total_steps": 2440, "loss": 0.4918, "lr": 3.964980802469552e-05, "epoch": 0.7676560900716479, "percentage": 15.37, "elapsed_time": "1:55:43", "remaining_time": "10:37:17"}
+{"current_steps": 376, "total_steps": 2440, "loss": 0.5219, "lr": 3.964445713365682e-05, "epoch": 0.7697031729785057, "percentage": 15.41, "elapsed_time": "1:56:06", "remaining_time": "10:37:21"}
+{"current_steps": 377, "total_steps": 2440, "loss": 0.5491, "lr": 3.963906603808422e-05, "epoch": 0.7717502558853634, "percentage": 15.45, "elapsed_time": "1:56:23", "remaining_time": "10:36:55"}
+{"current_steps": 378, "total_steps": 2440, "loss": 0.5193, "lr": 3.96336347490112e-05, "epoch": 0.7737973387922211, "percentage": 15.49, "elapsed_time": "1:56:41", "remaining_time": "10:36:35"}
+{"current_steps": 379, "total_steps": 2440, "loss": 0.5059, "lr": 3.9628163277553486e-05, "epoch": 0.7758444216990789, "percentage": 15.53, "elapsed_time": "1:57:02", "remaining_time": "10:36:29"}
+{"current_steps": 380, "total_steps": 2440, "loss": 0.5242, "lr": 3.962265163490903e-05, "epoch": 0.7778915046059366, "percentage": 15.57, "elapsed_time": "1:57:23", "remaining_time": "10:36:22"}
+{"current_steps": 381, "total_steps": 2440, "loss": 0.5131, "lr": 3.9617099832358035e-05, "epoch": 0.7799385875127943, "percentage": 15.61, "elapsed_time": "1:57:41", "remaining_time": "10:36:01"}
+{"current_steps": 382, "total_steps": 2440, "loss": 0.5247, "lr": 3.961150788126286e-05, "epoch": 0.781985670419652, "percentage": 15.66, "elapsed_time": "1:58:00", "remaining_time": "10:35:43"}
+{"current_steps": 383, "total_steps": 2440, "loss": 0.5082, "lr": 3.960587579306805e-05, "epoch": 0.7840327533265097, "percentage": 15.7, "elapsed_time": "1:58:19", "remaining_time": "10:35:29"}
+{"current_steps": 384, "total_steps": 2440, "loss": 0.4886, "lr": 3.960020357930028e-05, "epoch": 0.7860798362333674, "percentage": 15.74, "elapsed_time": "1:58:36", "remaining_time": "10:35:04"}
+{"current_steps": 385, "total_steps": 2440, "loss": 0.5323, "lr": 3.9594491251568376e-05, "epoch": 0.7881269191402251, "percentage": 15.78, "elapsed_time": "1:58:55", "remaining_time": "10:34:47"}
+{"current_steps": 386, "total_steps": 2440, "loss": 0.4913, "lr": 3.958873882156322e-05, "epoch": 0.7901740020470829, "percentage": 15.82, "elapsed_time": "1:59:13", "remaining_time": "10:34:26"}
+{"current_steps": 387, "total_steps": 2440, "loss": 0.5083, "lr": 3.9582946301057806e-05, "epoch": 0.7922210849539406, "percentage": 15.86, "elapsed_time": "1:59:32", "remaining_time": "10:34:11"}
+{"current_steps": 388, "total_steps": 2440, "loss": 0.503, "lr": 3.957711370190716e-05, "epoch": 0.7942681678607983, "percentage": 15.9, "elapsed_time": "1:59:53", "remaining_time": "10:34:05"}
+{"current_steps": 389, "total_steps": 2440, "loss": 0.5082, "lr": 3.957124103604833e-05, "epoch": 0.7963152507676561, "percentage": 15.94, "elapsed_time": "2:00:11", "remaining_time": "10:33:43"}
+{"current_steps": 390, "total_steps": 2440, "loss": 0.5166, "lr": 3.9565328315500375e-05, "epoch": 0.7983623336745138, "percentage": 15.98, "elapsed_time": "2:00:27", "remaining_time": "10:33:12"}
+{"current_steps": 391, "total_steps": 2440, "loss": 0.5143, "lr": 3.9559375552364325e-05, "epoch": 0.8004094165813715, "percentage": 16.02, "elapsed_time": "2:00:44", "remaining_time": "10:32:45"}
+{"current_steps": 392, "total_steps": 2440, "loss": 0.4713, "lr": 3.955338275882316e-05, "epoch": 0.8024564994882293, "percentage": 16.07, "elapsed_time": "2:01:02", "remaining_time": "10:32:24"}
+{"current_steps": 393, "total_steps": 2440, "loss": 0.4701, "lr": 3.9547349947141787e-05, "epoch": 0.804503582395087, "percentage": 16.11, "elapsed_time": "2:01:21", "remaining_time": "10:32:07"}
+{"current_steps": 394, "total_steps": 2440, "loss": 0.4916, "lr": 3.954127712966702e-05, "epoch": 0.8065506653019447, "percentage": 16.15, "elapsed_time": "2:01:43", "remaining_time": "10:32:05"}
+{"current_steps": 395, "total_steps": 2440, "loss": 0.535, "lr": 3.953516431882754e-05, "epoch": 0.8085977482088025, "percentage": 16.19, "elapsed_time": "2:02:04", "remaining_time": "10:32:00"}
+{"current_steps": 396, "total_steps": 2440, "loss": 0.4756, "lr": 3.952901152713389e-05, "epoch": 0.8106448311156602, "percentage": 16.23, "elapsed_time": "2:02:20", "remaining_time": "10:31:27"}
+{"current_steps": 397, "total_steps": 2440, "loss": 0.5013, "lr": 3.952281876717843e-05, "epoch": 0.812691914022518, "percentage": 16.27, "elapsed_time": "2:02:41", "remaining_time": "10:31:22"}
+{"current_steps": 398, "total_steps": 2440, "loss": 0.5159, "lr": 3.951658605163533e-05, "epoch": 0.8147389969293757, "percentage": 16.31, "elapsed_time": "2:03:01", "remaining_time": "10:31:12"}
+{"current_steps": 399, "total_steps": 2440, "loss": 0.4876, "lr": 3.9510313393260507e-05, "epoch": 0.8167860798362334, "percentage": 16.35, "elapsed_time": "2:03:15", "remaining_time": "10:30:28"}
+{"current_steps": 400, "total_steps": 2440, "loss": 0.486, "lr": 3.950400080489165e-05, "epoch": 0.8188331627430911, "percentage": 16.39, "elapsed_time": "2:03:32", "remaining_time": "10:30:03"}
+{"current_steps": 401, "total_steps": 2440, "loss": 0.6514, "lr": 3.9497648299448174e-05, "epoch": 0.8208802456499488, "percentage": 16.43, "elapsed_time": "2:03:54", "remaining_time": "10:30:03"}
+{"current_steps": 402, "total_steps": 2440, "loss": 0.5247, "lr": 3.949125588993117e-05, "epoch": 0.8229273285568065, "percentage": 16.48, "elapsed_time": "2:04:15", "remaining_time": "10:29:56"}
+{"current_steps": 403, "total_steps": 2440, "loss": 0.5023, "lr": 3.94848235894234e-05, "epoch": 0.8249744114636642, "percentage": 16.52, "elapsed_time": "2:04:34", "remaining_time": "10:29:38"}
+{"current_steps": 404, "total_steps": 2440, "loss": 0.5187, "lr": 3.947835141108928e-05, "epoch": 0.827021494370522, "percentage": 16.56, "elapsed_time": "2:04:49", "remaining_time": "10:29:06"}
+{"current_steps": 405, "total_steps": 2440, "loss": 0.5089, "lr": 3.947183936817483e-05, "epoch": 0.8290685772773797, "percentage": 16.6, "elapsed_time": "2:05:08", "remaining_time": "10:28:49"}
+{"current_steps": 406, "total_steps": 2440, "loss": 0.4946, "lr": 3.9465287474007654e-05, "epoch": 0.8311156601842374, "percentage": 16.64, "elapsed_time": "2:05:25", "remaining_time": "10:28:20"}
+{"current_steps": 407, "total_steps": 2440, "loss": 0.4905, "lr": 3.945869574199693e-05, "epoch": 0.8331627430910952, "percentage": 16.68, "elapsed_time": "2:05:42", "remaining_time": "10:27:57"}
+{"current_steps": 408, "total_steps": 2440, "loss": 0.4873, "lr": 3.9452064185633345e-05, "epoch": 0.8352098259979529, "percentage": 16.72, "elapsed_time": "2:06:02", "remaining_time": "10:27:45"}
+{"current_steps": 409, "total_steps": 2440, "loss": 0.4995, "lr": 3.944539281848912e-05, "epoch": 0.8372569089048106, "percentage": 16.76, "elapsed_time": "2:06:21", "remaining_time": "10:27:26"}
+{"current_steps": 410, "total_steps": 2440, "loss": 0.4597, "lr": 3.943868165421793e-05, "epoch": 0.8393039918116684, "percentage": 16.8, "elapsed_time": "2:06:38", "remaining_time": "10:26:59"}
+{"current_steps": 411, "total_steps": 2440, "loss": 0.4768, "lr": 3.943193070655492e-05, "epoch": 0.8413510747185261, "percentage": 16.84, "elapsed_time": "2:06:58", "remaining_time": "10:26:51"}
+{"current_steps": 412, "total_steps": 2440, "loss": 0.4936, "lr": 3.942513998931663e-05, "epoch": 0.8433981576253838, "percentage": 16.89, "elapsed_time": "2:07:16", "remaining_time": "10:26:27"}
+{"current_steps": 413, "total_steps": 2440, "loss": 0.4998, "lr": 3.9418309516401015e-05, "epoch": 0.8454452405322416, "percentage": 16.93, "elapsed_time": "2:07:37", "remaining_time": "10:26:21"}
+{"current_steps": 414, "total_steps": 2440, "loss": 0.4922, "lr": 3.9411439301787383e-05, "epoch": 0.8474923234390993, "percentage": 16.97, "elapsed_time": "2:07:56", "remaining_time": "10:26:05"}
+{"current_steps": 415, "total_steps": 2440, "loss": 0.4932, "lr": 3.940452935953639e-05, "epoch": 0.849539406345957, "percentage": 17.01, "elapsed_time": "2:08:12", "remaining_time": "10:25:34"}
+{"current_steps": 416, "total_steps": 2440, "loss": 0.5325, "lr": 3.939757970378997e-05, "epoch": 0.8515864892528148, "percentage": 17.05, "elapsed_time": "2:08:31", "remaining_time": "10:25:19"}
+{"current_steps": 417, "total_steps": 2440, "loss": 0.5161, "lr": 3.9390590348771374e-05, "epoch": 0.8536335721596725, "percentage": 17.09, "elapsed_time": "2:08:50", "remaining_time": "10:25:04"}
+{"current_steps": 418, "total_steps": 2440, "loss": 0.4872, "lr": 3.9383561308785075e-05, "epoch": 0.8556806550665302, "percentage": 17.13, "elapsed_time": "2:09:07", "remaining_time": "10:24:36"}
+{"current_steps": 419, "total_steps": 2440, "loss": 0.5033, "lr": 3.937649259821677e-05, "epoch": 0.857727737973388, "percentage": 17.17, "elapsed_time": "2:09:26", "remaining_time": "10:24:20"}
+{"current_steps": 420, "total_steps": 2440, "loss": 0.5022, "lr": 3.9369384231533365e-05, "epoch": 0.8597748208802457, "percentage": 17.21, "elapsed_time": "2:09:46", "remaining_time": "10:24:09"}
+{"current_steps": 421, "total_steps": 2440, "loss": 0.488, "lr": 3.9362236223282885e-05, "epoch": 0.8618219037871033, "percentage": 17.25, "elapsed_time": "2:10:03", "remaining_time": "10:23:41"}
+{"current_steps": 422, "total_steps": 2440, "loss": 0.5364, "lr": 3.935504858809454e-05, "epoch": 0.8638689866939611, "percentage": 17.3, "elapsed_time": "2:10:25", "remaining_time": "10:23:40"}
+{"current_steps": 423, "total_steps": 2440, "loss": 0.519, "lr": 3.9347821340678597e-05, "epoch": 0.8659160696008188, "percentage": 17.34, "elapsed_time": "2:10:46", "remaining_time": "10:23:36"}
+{"current_steps": 424, "total_steps": 2440, "loss": 0.5568, "lr": 3.934055449582641e-05, "epoch": 0.8679631525076765, "percentage": 17.38, "elapsed_time": "2:11:07", "remaining_time": "10:23:25"}
+{"current_steps": 425, "total_steps": 2440, "loss": 0.4743, "lr": 3.9333248068410375e-05, "epoch": 0.8700102354145343, "percentage": 17.42, "elapsed_time": "2:11:26", "remaining_time": "10:23:11"}
+{"current_steps": 426, "total_steps": 2440, "loss": 0.4982, "lr": 3.932590207338391e-05, "epoch": 0.872057318321392, "percentage": 17.46, "elapsed_time": "2:11:41", "remaining_time": "10:22:37"}
+{"current_steps": 427, "total_steps": 2440, "loss": 0.4963, "lr": 3.931851652578137e-05, "epoch": 0.8741044012282497, "percentage": 17.5, "elapsed_time": "2:12:01", "remaining_time": "10:22:25"}
+{"current_steps": 428, "total_steps": 2440, "loss": 0.4719, "lr": 3.931109144071811e-05, "epoch": 0.8761514841351075, "percentage": 17.54, "elapsed_time": "2:12:16", "remaining_time": "10:21:50"}
+{"current_steps": 429, "total_steps": 2440, "loss": 0.4863, "lr": 3.930362683339037e-05, "epoch": 0.8781985670419652, "percentage": 17.58, "elapsed_time": "2:12:33", "remaining_time": "10:21:23"}
+{"current_steps": 430, "total_steps": 2440, "loss": 0.5167, "lr": 3.92961227190753e-05, "epoch": 0.8802456499488229, "percentage": 17.62, "elapsed_time": "2:12:51", "remaining_time": "10:21:02"}
+{"current_steps": 431, "total_steps": 2440, "loss": 0.5031, "lr": 3.928857911313088e-05, "epoch": 0.8822927328556807, "percentage": 17.66, "elapsed_time": "2:13:07", "remaining_time": "10:20:30"}
+{"current_steps": 432, "total_steps": 2440, "loss": 0.5013, "lr": 3.928099603099591e-05, "epoch": 0.8843398157625384, "percentage": 17.7, "elapsed_time": "2:13:26", "remaining_time": "10:20:13"}
+{"current_steps": 433, "total_steps": 2440, "loss": 0.5574, "lr": 3.9273373488190036e-05, "epoch": 0.8863868986693961, "percentage": 17.75, "elapsed_time": "2:13:48", "remaining_time": "10:20:12"}
+{"current_steps": 434, "total_steps": 2440, "loss": 0.5253, "lr": 3.92657115003136e-05, "epoch": 0.8884339815762539, "percentage": 17.79, "elapsed_time": "2:14:11", "remaining_time": "10:20:16"}
+{"current_steps": 435, "total_steps": 2440, "loss": 0.5485, "lr": 3.9258010083047715e-05, "epoch": 0.8904810644831116, "percentage": 17.83, "elapsed_time": "2:14:33", "remaining_time": "10:20:10"}
+{"current_steps": 436, "total_steps": 2440, "loss": 0.4636, "lr": 3.925026925215417e-05, "epoch": 0.8925281473899693, "percentage": 17.87, "elapsed_time": "2:14:49", "remaining_time": "10:19:43"}
+{"current_steps": 437, "total_steps": 2440, "loss": 0.5464, "lr": 3.924248902347541e-05, "epoch": 0.8945752302968271, "percentage": 17.91, "elapsed_time": "2:15:12", "remaining_time": "10:19:41"}
+{"current_steps": 438, "total_steps": 2440, "loss": 0.5461, "lr": 3.9234669412934546e-05, "epoch": 0.8966223132036848, "percentage": 17.95, "elapsed_time": "2:15:32", "remaining_time": "10:19:31"}
+{"current_steps": 439, "total_steps": 2440, "loss": 0.5146, "lr": 3.922681043653526e-05, "epoch": 0.8986693961105425, "percentage": 17.99, "elapsed_time": "2:15:52", "remaining_time": "10:19:17"}
+{"current_steps": 440, "total_steps": 2440, "loss": 0.4898, "lr": 3.92189121103618e-05, "epoch": 0.9007164790174002, "percentage": 18.03, "elapsed_time": "2:16:12", "remaining_time": "10:19:09"}
+{"current_steps": 441, "total_steps": 2440, "loss": 0.4976, "lr": 3.921097445057896e-05, "epoch": 0.9027635619242579, "percentage": 18.07, "elapsed_time": "2:16:32", "remaining_time": "10:18:57"}
+{"current_steps": 442, "total_steps": 2440, "loss": 0.4894, "lr": 3.920299747343204e-05, "epoch": 0.9048106448311156, "percentage": 18.11, "elapsed_time": "2:16:53", "remaining_time": "10:18:49"}
+{"current_steps": 443, "total_steps": 2440, "loss": 0.5029, "lr": 3.919498119524679e-05, "epoch": 0.9068577277379734, "percentage": 18.16, "elapsed_time": "2:17:13", "remaining_time": "10:18:35"}
+{"current_steps": 444, "total_steps": 2440, "loss": 0.5477, "lr": 3.9186925632429396e-05, "epoch": 0.9089048106448311, "percentage": 18.2, "elapsed_time": "2:17:30", "remaining_time": "10:18:08"}
+{"current_steps": 445, "total_steps": 2440, "loss": 0.5147, "lr": 3.9178830801466465e-05, "epoch": 0.9109518935516888, "percentage": 18.24, "elapsed_time": "2:17:50", "remaining_time": "10:17:56"}
+{"current_steps": 446, "total_steps": 2440, "loss": 0.5052, "lr": 3.917069671892494e-05, "epoch": 0.9129989764585466, "percentage": 18.28, "elapsed_time": "2:18:09", "remaining_time": "10:17:40"}
+{"current_steps": 447, "total_steps": 2440, "loss": 0.476, "lr": 3.9162523401452125e-05, "epoch": 0.9150460593654043, "percentage": 18.32, "elapsed_time": "2:18:26", "remaining_time": "10:17:15"}
+{"current_steps": 448, "total_steps": 2440, "loss": 0.4811, "lr": 3.915431086577561e-05, "epoch": 0.917093142272262, "percentage": 18.36, "elapsed_time": "2:18:38", "remaining_time": "10:16:29"}
+{"current_steps": 449, "total_steps": 2440, "loss": 0.5082, "lr": 3.914605912870324e-05, "epoch": 0.9191402251791198, "percentage": 18.4, "elapsed_time": "2:18:54", "remaining_time": "10:15:58"}
+{"current_steps": 450, "total_steps": 2440, "loss": 0.4982, "lr": 3.913776820712309e-05, "epoch": 0.9211873080859775, "percentage": 18.44, "elapsed_time": "2:19:16", "remaining_time": "10:15:53"}
+{"current_steps": 451, "total_steps": 2440, "loss": 0.477, "lr": 3.912943811800347e-05, "epoch": 0.9232343909928352, "percentage": 18.48, "elapsed_time": "2:19:32", "remaining_time": "10:15:24"}
+{"current_steps": 452, "total_steps": 2440, "loss": 0.4953, "lr": 3.912106887839278e-05, "epoch": 0.925281473899693, "percentage": 18.52, "elapsed_time": "2:19:57", "remaining_time": "10:15:32"}
+{"current_steps": 453, "total_steps": 2440, "loss": 0.5248, "lr": 3.9112660505419626e-05, "epoch": 0.9273285568065507, "percentage": 18.57, "elapsed_time": "2:20:19", "remaining_time": "10:15:31"}
+{"current_steps": 454, "total_steps": 2440, "loss": 0.4887, "lr": 3.910421301629264e-05, "epoch": 0.9293756397134084, "percentage": 18.61, "elapsed_time": "2:20:38", "remaining_time": "10:15:13"}
+{"current_steps": 455, "total_steps": 2440, "loss": 0.4987, "lr": 3.909572642830053e-05, "epoch": 0.9314227226202662, "percentage": 18.65, "elapsed_time": "2:21:00", "remaining_time": "10:15:09"}
+{"current_steps": 456, "total_steps": 2440, "loss": 0.5429, "lr": 3.9087200758812054e-05, "epoch": 0.9334698055271239, "percentage": 18.69, "elapsed_time": "2:21:20", "remaining_time": "10:14:55"}
+{"current_steps": 457, "total_steps": 2440, "loss": 0.4868, "lr": 3.9078636025275904e-05, "epoch": 0.9355168884339816, "percentage": 18.73, "elapsed_time": "2:21:40", "remaining_time": "10:14:44"}
+{"current_steps": 458, "total_steps": 2440, "loss": 0.5151, "lr": 3.907003224522075e-05, "epoch": 0.9375639713408394, "percentage": 18.77, "elapsed_time": "2:21:58", "remaining_time": "10:14:24"}
+{"current_steps": 459, "total_steps": 2440, "loss": 0.4812, "lr": 3.906138943625519e-05, "epoch": 0.9396110542476971, "percentage": 18.81, "elapsed_time": "2:22:18", "remaining_time": "10:14:10"}
+{"current_steps": 460, "total_steps": 2440, "loss": 0.5405, "lr": 3.9052707616067654e-05, "epoch": 0.9416581371545547, "percentage": 18.85, "elapsed_time": "2:22:41", "remaining_time": "10:14:11"}
+{"current_steps": 461, "total_steps": 2440, "loss": 0.4816, "lr": 3.9043986802426453e-05, "epoch": 0.9437052200614124, "percentage": 18.89, "elapsed_time": "2:23:03", "remaining_time": "10:14:09"}
+{"current_steps": 462, "total_steps": 2440, "loss": 0.4864, "lr": 3.903522701317968e-05, "epoch": 0.9457523029682702, "percentage": 18.93, "elapsed_time": "2:23:26", "remaining_time": "10:14:09"}
+{"current_steps": 463, "total_steps": 2440, "loss": 0.4979, "lr": 3.9026428266255205e-05, "epoch": 0.9477993858751279, "percentage": 18.98, "elapsed_time": "2:23:42", "remaining_time": "10:13:37"}
+{"current_steps": 464, "total_steps": 2440, "loss": 0.4741, "lr": 3.901759057966064e-05, "epoch": 0.9498464687819856, "percentage": 19.02, "elapsed_time": "2:24:03", "remaining_time": "10:13:30"}
+{"current_steps": 465, "total_steps": 2440, "loss": 0.4729, "lr": 3.9008713971483257e-05, "epoch": 0.9518935516888434, "percentage": 19.06, "elapsed_time": "2:24:22", "remaining_time": "10:13:13"}
+{"current_steps": 466, "total_steps": 2440, "loss": 0.5072, "lr": 3.899979845989003e-05, "epoch": 0.9539406345957011, "percentage": 19.1, "elapsed_time": "2:24:41", "remaining_time": "10:12:54"}
+{"current_steps": 467, "total_steps": 2440, "loss": 0.4844, "lr": 3.899084406312751e-05, "epoch": 0.9559877175025588, "percentage": 19.14, "elapsed_time": "2:24:55", "remaining_time": "10:12:17"}
+{"current_steps": 468, "total_steps": 2440, "loss": 0.5021, "lr": 3.8981850799521856e-05, "epoch": 0.9580348004094166, "percentage": 19.18, "elapsed_time": "2:25:14", "remaining_time": "10:12:02"}
+{"current_steps": 469, "total_steps": 2440, "loss": 0.5003, "lr": 3.897281868747878e-05, "epoch": 0.9600818833162743, "percentage": 19.22, "elapsed_time": "2:25:36", "remaining_time": "10:11:55"}
+{"current_steps": 470, "total_steps": 2440, "loss": 0.5054, "lr": 3.896374774548348e-05, "epoch": 0.962128966223132, "percentage": 19.26, "elapsed_time": "2:25:54", "remaining_time": "10:11:33"}
+{"current_steps": 471, "total_steps": 2440, "loss": 0.5186, "lr": 3.895463799210063e-05, "epoch": 0.9641760491299898, "percentage": 19.3, "elapsed_time": "2:26:10", "remaining_time": "10:11:03"}
+{"current_steps": 472, "total_steps": 2440, "loss": 0.5227, "lr": 3.894548944597434e-05, "epoch": 0.9662231320368475, "percentage": 19.34, "elapsed_time": "2:26:30", "remaining_time": "10:10:51"}
+{"current_steps": 473, "total_steps": 2440, "loss": 0.4836, "lr": 3.8936302125828114e-05, "epoch": 0.9682702149437052, "percentage": 19.39, "elapsed_time": "2:26:51", "remaining_time": "10:10:44"}
+{"current_steps": 474, "total_steps": 2440, "loss": 0.465, "lr": 3.892707605046482e-05, "epoch": 0.970317297850563, "percentage": 19.43, "elapsed_time": "2:27:11", "remaining_time": "10:10:31"}
+{"current_steps": 475, "total_steps": 2440, "loss": 0.5103, "lr": 3.8917811238766606e-05, "epoch": 0.9723643807574207, "percentage": 19.47, "elapsed_time": "2:27:33", "remaining_time": "10:10:27"}
+{"current_steps": 476, "total_steps": 2440, "loss": 0.4857, "lr": 3.8908507709694945e-05, "epoch": 0.9744114636642784, "percentage": 19.51, "elapsed_time": "2:27:49", "remaining_time": "10:09:56"}
+{"current_steps": 477, "total_steps": 2440, "loss": 0.5135, "lr": 3.8899165482290524e-05, "epoch": 0.9764585465711362, "percentage": 19.55, "elapsed_time": "2:28:07", "remaining_time": "10:09:35"}
+{"current_steps": 478, "total_steps": 2440, "loss": 0.4826, "lr": 3.888978457567323e-05, "epoch": 0.9785056294779939, "percentage": 19.59, "elapsed_time": "2:28:26", "remaining_time": "10:09:16"}
+{"current_steps": 479, "total_steps": 2440, "loss": 0.5154, "lr": 3.888036500904212e-05, "epoch": 0.9805527123848515, "percentage": 19.63, "elapsed_time": "2:28:43", "remaining_time": "10:08:54"}
+{"current_steps": 480, "total_steps": 2440, "loss": 0.5087, "lr": 3.887090680167537e-05, "epoch": 0.9825997952917093, "percentage": 19.67, "elapsed_time": "2:29:02", "remaining_time": "10:08:33"}
+{"current_steps": 481, "total_steps": 2440, "loss": 0.5033, "lr": 3.886140997293024e-05, "epoch": 0.984646878198567, "percentage": 19.71, "elapsed_time": "2:29:24", "remaining_time": "10:08:30"}
+{"current_steps": 482, "total_steps": 2440, "loss": 0.4737, "lr": 3.8851874542243024e-05, "epoch": 0.9866939611054247, "percentage": 19.75, "elapsed_time": "2:29:41", "remaining_time": "10:08:06"}
+{"current_steps": 483, "total_steps": 2440, "loss": 0.4922, "lr": 3.8842300529129026e-05, "epoch": 0.9887410440122825, "percentage": 19.8, "elapsed_time": "2:30:05", "remaining_time": "10:08:06"}
+{"current_steps": 484, "total_steps": 2440, "loss": 0.5093, "lr": 3.883268795318252e-05, "epoch": 0.9907881269191402, "percentage": 19.84, "elapsed_time": "2:30:24", "remaining_time": "10:07:51"}
+{"current_steps": 485, "total_steps": 2440, "loss": 0.4898, "lr": 3.882303683407669e-05, "epoch": 0.9928352098259979, "percentage": 19.88, "elapsed_time": "2:30:46", "remaining_time": "10:07:46"}
+{"current_steps": 486, "total_steps": 2440, "loss": 0.4809, "lr": 3.8813347191563615e-05, "epoch": 0.9948822927328557, "percentage": 19.92, "elapsed_time": "2:31:02", "remaining_time": "10:07:15"}
+{"current_steps": 487, "total_steps": 2440, "loss": 0.4792, "lr": 3.88036190454742e-05, "epoch": 0.9969293756397134, "percentage": 19.96, "elapsed_time": "2:31:23", "remaining_time": "10:07:06"}
+{"current_steps": 488, "total_steps": 2440, "loss": 0.4945, "lr": 3.879385241571817e-05, "epoch": 0.9989764585465711, "percentage": 20.0, "elapsed_time": "2:31:41", "remaining_time": "10:06:47"}
+{"current_steps": 489, "total_steps": 2440, "loss": 0.7219, "lr": 3.8784047322284e-05, "epoch": 1.0010235414534288, "percentage": 20.04, "elapsed_time": "2:32:58", "remaining_time": "10:10:20"}
+{"current_steps": 490, "total_steps": 2440, "loss": 0.5356, "lr": 3.8774203785238886e-05, "epoch": 1.0030706243602865, "percentage": 20.08, "elapsed_time": "2:33:15", "remaining_time": "10:09:53"}
+{"current_steps": 491, "total_steps": 2440, "loss": 0.3433, "lr": 3.8764321824728715e-05, "epoch": 1.0051177072671442, "percentage": 20.12, "elapsed_time": "2:33:29", "remaining_time": "10:09:15"}
+{"current_steps": 492, "total_steps": 2440, "loss": 0.4658, "lr": 3.875440146097798e-05, "epoch": 1.007164790174002, "percentage": 20.16, "elapsed_time": "2:33:49", "remaining_time": "10:09:02"}
+{"current_steps": 493, "total_steps": 2440, "loss": 0.423, "lr": 3.8744442714289816e-05, "epoch": 1.0092118730808597, "percentage": 20.2, "elapsed_time": "2:34:08", "remaining_time": "10:08:46"}
+{"current_steps": 494, "total_steps": 2440, "loss": 0.4378, "lr": 3.873444560504588e-05, "epoch": 1.0112589559877174, "percentage": 20.25, "elapsed_time": "2:34:28", "remaining_time": "10:08:30"}
+{"current_steps": 495, "total_steps": 2440, "loss": 0.4592, "lr": 3.872441015370635e-05, "epoch": 1.0133060388945752, "percentage": 20.29, "elapsed_time": "2:34:48", "remaining_time": "10:08:18"}
+{"current_steps": 496, "total_steps": 2440, "loss": 0.4401, "lr": 3.8714336380809874e-05, "epoch": 1.015353121801433, "percentage": 20.33, "elapsed_time": "2:35:11", "remaining_time": "10:08:16"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98d959a169a1b55a25b0a2e2804fa070799fb17cf3ae9774eb707c03a29ef143
+size 7288