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 sha256:66445f3c99a00512a39e4d3cbdb1120fea11d7c8fcf55e193cf7f0ee0a7f02c2
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44d09f50adb6bd7c567967ba437ad5e37e923c81a821ad9ca1283f027004e80f
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b204db68134458d628cffe84a743776d1a7dea08c90ddc453123c4f9608ef1c
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44d6ea7a6d1cb4558b59e3f1f41c5eaaa415c7a9e0571fa8b7217e3552102f94
+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 +1,476 @@
+{"current_steps": 1, "total_steps": 2365, "loss": 1.8303, "lr": 3.3755274261603377e-07, "epoch": 0.0021119324181626186, "percentage": 0.04, "elapsed_time": "0:01:02", "remaining_time": "1 day, 17:13:42"}
+{"current_steps": 2, "total_steps": 2365, "loss": 1.8342, "lr": 6.751054852320675e-07, "epoch": 0.004223864836325237, "percentage": 0.08, "elapsed_time": "0:01:51", "remaining_time": "1 day, 12:41:56"}
+{"current_steps": 3, "total_steps": 2365, "loss": 1.8528, "lr": 1.0126582278481013e-06, "epoch": 0.006335797254487857, "percentage": 0.13, "elapsed_time": "0:02:39", "remaining_time": "1 day, 10:53:44"}
+{"current_steps": 4, "total_steps": 2365, "loss": 1.8386, "lr": 1.350210970464135e-06, "epoch": 0.008447729672650475, "percentage": 0.17, "elapsed_time": "0:03:28", "remaining_time": "1 day, 10:06:54"}
+{"current_steps": 5, "total_steps": 2365, "loss": 1.8274, "lr": 1.6877637130801689e-06, "epoch": 0.010559662090813094, "percentage": 0.21, "elapsed_time": "0:04:16", "remaining_time": "1 day, 9:40:08"}
+{"current_steps": 6, "total_steps": 2365, "loss": 1.8286, "lr": 2.0253164556962026e-06, "epoch": 0.012671594508975714, "percentage": 0.25, "elapsed_time": "0:05:05", "remaining_time": "1 day, 9:20:33"}
+{"current_steps": 7, "total_steps": 2365, "loss": 1.7907, "lr": 2.3628691983122364e-06, "epoch": 0.014783526927138331, "percentage": 0.3, "elapsed_time": "0:05:53", "remaining_time": "1 day, 9:01:54"}
+{"current_steps": 8, "total_steps": 2365, "loss": 1.7865, "lr": 2.70042194092827e-06, "epoch": 0.01689545934530095, "percentage": 0.34, "elapsed_time": "0:06:40", "remaining_time": "1 day, 8:48:43"}
+{"current_steps": 9, "total_steps": 2365, "loss": 1.7288, "lr": 3.037974683544304e-06, "epoch": 0.01900739176346357, "percentage": 0.38, "elapsed_time": "0:07:28", "remaining_time": "1 day, 8:38:17"}
+{"current_steps": 10, "total_steps": 2365, "loss": 1.7315, "lr": 3.3755274261603377e-06, "epoch": 0.021119324181626188, "percentage": 0.42, "elapsed_time": "0:08:18", "remaining_time": "1 day, 8:35:19"}
+{"current_steps": 11, "total_steps": 2365, "loss": 1.705, "lr": 3.713080168776372e-06, "epoch": 0.023231256599788808, "percentage": 0.47, "elapsed_time": "0:09:06", "remaining_time": "1 day, 8:28:21"}
+{"current_steps": 12, "total_steps": 2365, "loss": 1.7221, "lr": 4.050632911392405e-06, "epoch": 0.025343189017951427, "percentage": 0.51, "elapsed_time": "0:09:55", "remaining_time": "1 day, 8:24:32"}
+{"current_steps": 13, "total_steps": 2365, "loss": 1.7034, "lr": 4.3881856540084394e-06, "epoch": 0.027455121436114043, "percentage": 0.55, "elapsed_time": "0:10:44", "remaining_time": "1 day, 8:22:18"}
+{"current_steps": 14, "total_steps": 2365, "loss": 1.7033, "lr": 4.725738396624473e-06, "epoch": 0.029567053854276663, "percentage": 0.59, "elapsed_time": "0:11:32", "remaining_time": "1 day, 8:17:01"}
+{"current_steps": 15, "total_steps": 2365, "loss": 1.6944, "lr": 5.063291139240507e-06, "epoch": 0.03167898627243928, "percentage": 0.63, "elapsed_time": "0:12:20", "remaining_time": "1 day, 8:13:05"}
+{"current_steps": 16, "total_steps": 2365, "loss": 1.6269, "lr": 5.40084388185654e-06, "epoch": 0.0337909186906019, "percentage": 0.68, "elapsed_time": "0:13:08", "remaining_time": "1 day, 8:10:14"}
+{"current_steps": 17, "total_steps": 2365, "loss": 1.6425, "lr": 5.7383966244725745e-06, "epoch": 0.03590285110876452, "percentage": 0.72, "elapsed_time": "0:13:56", "remaining_time": "1 day, 8:05:48"}
+{"current_steps": 18, "total_steps": 2365, "loss": 1.5935, "lr": 6.075949367088608e-06, "epoch": 0.03801478352692714, "percentage": 0.76, "elapsed_time": "0:14:44", "remaining_time": "1 day, 8:02:00"}
+{"current_steps": 19, "total_steps": 2365, "loss": 1.5977, "lr": 6.413502109704642e-06, "epoch": 0.04012671594508976, "percentage": 0.8, "elapsed_time": "0:15:32", "remaining_time": "1 day, 7:59:13"}
+{"current_steps": 20, "total_steps": 2365, "loss": 1.5748, "lr": 6.751054852320675e-06, "epoch": 0.042238648363252376, "percentage": 0.85, "elapsed_time": "0:16:20", "remaining_time": "1 day, 7:56:52"}
+{"current_steps": 21, "total_steps": 2365, "loss": 1.5797, "lr": 7.08860759493671e-06, "epoch": 0.044350580781414996, "percentage": 0.89, "elapsed_time": "0:17:08", "remaining_time": "1 day, 7:53:58"}
+{"current_steps": 22, "total_steps": 2365, "loss": 1.5488, "lr": 7.426160337552744e-06, "epoch": 0.046462513199577615, "percentage": 0.93, "elapsed_time": "0:17:57", "remaining_time": "1 day, 7:51:58"}
+{"current_steps": 23, "total_steps": 2365, "loss": 1.5495, "lr": 7.763713080168777e-06, "epoch": 0.048574445617740235, "percentage": 0.97, "elapsed_time": "0:18:45", "remaining_time": "1 day, 7:50:49"}
+{"current_steps": 24, "total_steps": 2365, "loss": 1.5263, "lr": 8.10126582278481e-06, "epoch": 0.050686378035902854, "percentage": 1.01, "elapsed_time": "0:19:34", "remaining_time": "1 day, 7:50:00"}
+{"current_steps": 25, "total_steps": 2365, "loss": 1.5186, "lr": 8.438818565400846e-06, "epoch": 0.05279831045406547, "percentage": 1.06, "elapsed_time": "0:20:23", "remaining_time": "1 day, 7:48:55"}
+{"current_steps": 26, "total_steps": 2365, "loss": 1.5195, "lr": 8.776371308016879e-06, "epoch": 0.054910242872228086, "percentage": 1.1, "elapsed_time": "0:21:11", "remaining_time": "1 day, 7:46:47"}
+{"current_steps": 27, "total_steps": 2365, "loss": 1.5197, "lr": 9.113924050632912e-06, "epoch": 0.057022175290390706, "percentage": 1.14, "elapsed_time": "0:22:00", "remaining_time": "1 day, 7:45:42"}
+{"current_steps": 28, "total_steps": 2365, "loss": 1.4907, "lr": 9.451476793248946e-06, "epoch": 0.059134107708553325, "percentage": 1.18, "elapsed_time": "0:22:48", "remaining_time": "1 day, 7:43:52"}
+{"current_steps": 29, "total_steps": 2365, "loss": 1.4846, "lr": 9.78902953586498e-06, "epoch": 0.061246040126715945, "percentage": 1.23, "elapsed_time": "0:23:36", "remaining_time": "1 day, 7:42:02"}
+{"current_steps": 30, "total_steps": 2365, "loss": 1.4761, "lr": 1.0126582278481014e-05, "epoch": 0.06335797254487856, "percentage": 1.27, "elapsed_time": "0:24:24", "remaining_time": "1 day, 7:40:19"}
+{"current_steps": 31, "total_steps": 2365, "loss": 1.4939, "lr": 1.0464135021097049e-05, "epoch": 0.06546990496304118, "percentage": 1.31, "elapsed_time": "0:25:13", "remaining_time": "1 day, 7:38:40"}
+{"current_steps": 32, "total_steps": 2365, "loss": 1.4677, "lr": 1.080168776371308e-05, "epoch": 0.0675818373812038, "percentage": 1.35, "elapsed_time": "0:26:01", "remaining_time": "1 day, 7:37:22"}
+{"current_steps": 33, "total_steps": 2365, "loss": 1.4733, "lr": 1.1139240506329114e-05, "epoch": 0.06969376979936642, "percentage": 1.4, "elapsed_time": "0:26:49", "remaining_time": "1 day, 7:36:06"}
+{"current_steps": 34, "total_steps": 2365, "loss": 1.4797, "lr": 1.1476793248945149e-05, "epoch": 0.07180570221752904, "percentage": 1.44, "elapsed_time": "0:27:39", "remaining_time": "1 day, 7:35:40"}
+{"current_steps": 35, "total_steps": 2365, "loss": 1.4615, "lr": 1.1814345991561182e-05, "epoch": 0.07391763463569166, "percentage": 1.48, "elapsed_time": "0:28:27", "remaining_time": "1 day, 7:34:53"}
+{"current_steps": 36, "total_steps": 2365, "loss": 1.448, "lr": 1.2151898734177216e-05, "epoch": 0.07602956705385427, "percentage": 1.52, "elapsed_time": "0:29:16", "remaining_time": "1 day, 7:33:50"}
+{"current_steps": 37, "total_steps": 2365, "loss": 1.4474, "lr": 1.248945147679325e-05, "epoch": 0.0781414994720169, "percentage": 1.56, "elapsed_time": "0:30:04", "remaining_time": "1 day, 7:32:36"}
+{"current_steps": 38, "total_steps": 2365, "loss": 1.4574, "lr": 1.2827004219409284e-05, "epoch": 0.08025343189017951, "percentage": 1.61, "elapsed_time": "0:30:53", "remaining_time": "1 day, 7:31:22"}
+{"current_steps": 39, "total_steps": 2365, "loss": 1.4304, "lr": 1.3164556962025317e-05, "epoch": 0.08236536430834214, "percentage": 1.65, "elapsed_time": "0:31:42", "remaining_time": "1 day, 7:31:00"}
+{"current_steps": 40, "total_steps": 2365, "loss": 1.4401, "lr": 1.350210970464135e-05, "epoch": 0.08447729672650475, "percentage": 1.69, "elapsed_time": "0:32:31", "remaining_time": "1 day, 7:30:15"}
+{"current_steps": 41, "total_steps": 2365, "loss": 1.4314, "lr": 1.3839662447257384e-05, "epoch": 0.08658922914466737, "percentage": 1.73, "elapsed_time": "0:33:19", "remaining_time": "1 day, 7:28:38"}
+{"current_steps": 42, "total_steps": 2365, "loss": 1.4624, "lr": 1.417721518987342e-05, "epoch": 0.08870116156282999, "percentage": 1.78, "elapsed_time": "0:34:07", "remaining_time": "1 day, 7:27:22"}
+{"current_steps": 43, "total_steps": 2365, "loss": 1.4481, "lr": 1.4514767932489453e-05, "epoch": 0.0908130939809926, "percentage": 1.82, "elapsed_time": "0:34:56", "remaining_time": "1 day, 7:26:28"}
+{"current_steps": 44, "total_steps": 2365, "loss": 1.4363, "lr": 1.4852320675105488e-05, "epoch": 0.09292502639915523, "percentage": 1.86, "elapsed_time": "0:35:45", "remaining_time": "1 day, 7:25:52"}
+{"current_steps": 45, "total_steps": 2365, "loss": 1.4314, "lr": 1.5189873417721521e-05, "epoch": 0.09503695881731784, "percentage": 1.9, "elapsed_time": "0:36:32", "remaining_time": "1 day, 7:24:18"}
+{"current_steps": 46, "total_steps": 2365, "loss": 1.4301, "lr": 1.5527426160337554e-05, "epoch": 0.09714889123548047, "percentage": 1.95, "elapsed_time": "0:37:21", "remaining_time": "1 day, 7:23:05"}
+{"current_steps": 47, "total_steps": 2365, "loss": 1.4075, "lr": 1.5864978902953586e-05, "epoch": 0.09926082365364308, "percentage": 1.99, "elapsed_time": "0:38:09", "remaining_time": "1 day, 7:21:39"}
+{"current_steps": 48, "total_steps": 2365, "loss": 1.4314, "lr": 1.620253164556962e-05, "epoch": 0.10137275607180571, "percentage": 2.03, "elapsed_time": "0:38:57", "remaining_time": "1 day, 7:20:51"}
+{"current_steps": 49, "total_steps": 2365, "loss": 1.4345, "lr": 1.6540084388185656e-05, "epoch": 0.10348468848996832, "percentage": 2.07, "elapsed_time": "0:39:46", "remaining_time": "1 day, 7:19:37"}
+{"current_steps": 50, "total_steps": 2365, "loss": 1.4036, "lr": 1.687763713080169e-05, "epoch": 0.10559662090813093, "percentage": 2.11, "elapsed_time": "0:40:34", "remaining_time": "1 day, 7:18:27"}
+{"current_steps": 51, "total_steps": 2365, "loss": 1.4078, "lr": 1.7215189873417723e-05, "epoch": 0.10770855332629356, "percentage": 2.16, "elapsed_time": "0:41:22", "remaining_time": "1 day, 7:17:12"}
+{"current_steps": 52, "total_steps": 2365, "loss": 1.4258, "lr": 1.7552742616033758e-05, "epoch": 0.10982048574445617, "percentage": 2.2, "elapsed_time": "0:42:11", "remaining_time": "1 day, 7:16:25"}
+{"current_steps": 53, "total_steps": 2365, "loss": 1.4145, "lr": 1.789029535864979e-05, "epoch": 0.1119324181626188, "percentage": 2.24, "elapsed_time": "0:42:59", "remaining_time": "1 day, 7:15:19"}
+{"current_steps": 54, "total_steps": 2365, "loss": 1.4226, "lr": 1.8227848101265824e-05, "epoch": 0.11404435058078141, "percentage": 2.28, "elapsed_time": "0:43:47", "remaining_time": "1 day, 7:14:01"}
+{"current_steps": 55, "total_steps": 2365, "loss": 1.4075, "lr": 1.856540084388186e-05, "epoch": 0.11615628299894404, "percentage": 2.33, "elapsed_time": "0:44:35", "remaining_time": "1 day, 7:12:47"}
+{"current_steps": 56, "total_steps": 2365, "loss": 1.4092, "lr": 1.890295358649789e-05, "epoch": 0.11826821541710665, "percentage": 2.37, "elapsed_time": "0:45:23", "remaining_time": "1 day, 7:11:33"}
+{"current_steps": 57, "total_steps": 2365, "loss": 1.3971, "lr": 1.9240506329113926e-05, "epoch": 0.12038014783526928, "percentage": 2.41, "elapsed_time": "0:46:11", "remaining_time": "1 day, 7:10:33"}
+{"current_steps": 58, "total_steps": 2365, "loss": 1.3927, "lr": 1.957805907172996e-05, "epoch": 0.12249208025343189, "percentage": 2.45, "elapsed_time": "0:46:59", "remaining_time": "1 day, 7:09:16"}
+{"current_steps": 59, "total_steps": 2365, "loss": 1.3886, "lr": 1.9915611814345993e-05, "epoch": 0.1246040126715945, "percentage": 2.49, "elapsed_time": "0:47:47", "remaining_time": "1 day, 7:08:14"}
+{"current_steps": 60, "total_steps": 2365, "loss": 1.4144, "lr": 2.0253164556962028e-05, "epoch": 0.12671594508975711, "percentage": 2.54, "elapsed_time": "0:48:36", "remaining_time": "1 day, 7:07:12"}
+{"current_steps": 61, "total_steps": 2365, "loss": 1.4068, "lr": 2.0590717299578063e-05, "epoch": 0.12882787750791974, "percentage": 2.58, "elapsed_time": "0:49:25", "remaining_time": "1 day, 7:06:32"}
+{"current_steps": 62, "total_steps": 2365, "loss": 1.39, "lr": 2.0928270042194098e-05, "epoch": 0.13093980992608237, "percentage": 2.62, "elapsed_time": "0:50:12", "remaining_time": "1 day, 7:05:14"}
+{"current_steps": 63, "total_steps": 2365, "loss": 1.4075, "lr": 2.1265822784810126e-05, "epoch": 0.133051742344245, "percentage": 2.66, "elapsed_time": "0:51:01", "remaining_time": "1 day, 7:04:10"}
+{"current_steps": 64, "total_steps": 2365, "loss": 1.3832, "lr": 2.160337552742616e-05, "epoch": 0.1351636747624076, "percentage": 2.71, "elapsed_time": "0:51:49", "remaining_time": "1 day, 7:03:04"}
+{"current_steps": 65, "total_steps": 2365, "loss": 1.4017, "lr": 2.1940928270042193e-05, "epoch": 0.13727560718057022, "percentage": 2.75, "elapsed_time": "0:52:37", "remaining_time": "1 day, 7:01:55"}
+{"current_steps": 66, "total_steps": 2365, "loss": 1.4023, "lr": 2.2278481012658228e-05, "epoch": 0.13938753959873285, "percentage": 2.79, "elapsed_time": "0:53:25", "remaining_time": "1 day, 7:00:44"}
+{"current_steps": 67, "total_steps": 2365, "loss": 1.3853, "lr": 2.2616033755274263e-05, "epoch": 0.14149947201689547, "percentage": 2.83, "elapsed_time": "0:54:13", "remaining_time": "1 day, 6:59:44"}
+{"current_steps": 68, "total_steps": 2365, "loss": 1.4249, "lr": 2.2953586497890298e-05, "epoch": 0.14361140443505807, "percentage": 2.88, "elapsed_time": "0:55:01", "remaining_time": "1 day, 6:58:36"}
+{"current_steps": 69, "total_steps": 2365, "loss": 1.395, "lr": 2.329113924050633e-05, "epoch": 0.1457233368532207, "percentage": 2.92, "elapsed_time": "0:55:49", "remaining_time": "1 day, 6:57:29"}
+{"current_steps": 70, "total_steps": 2365, "loss": 1.3965, "lr": 2.3628691983122365e-05, "epoch": 0.14783526927138332, "percentage": 2.96, "elapsed_time": "0:56:37", "remaining_time": "1 day, 6:56:20"}
+{"current_steps": 71, "total_steps": 2365, "loss": 1.4209, "lr": 2.39662447257384e-05, "epoch": 0.14994720168954592, "percentage": 3.0, "elapsed_time": "0:57:25", "remaining_time": "1 day, 6:55:23"}
+{"current_steps": 72, "total_steps": 2365, "loss": 1.402, "lr": 2.430379746835443e-05, "epoch": 0.15205913410770855, "percentage": 3.04, "elapsed_time": "0:58:13", "remaining_time": "1 day, 6:54:13"}
+{"current_steps": 73, "total_steps": 2365, "loss": 1.4029, "lr": 2.4641350210970467e-05, "epoch": 0.15417106652587118, "percentage": 3.09, "elapsed_time": "0:59:02", "remaining_time": "1 day, 6:53:34"}
+{"current_steps": 74, "total_steps": 2365, "loss": 1.4202, "lr": 2.49789029535865e-05, "epoch": 0.1562829989440338, "percentage": 3.13, "elapsed_time": "0:59:50", "remaining_time": "1 day, 6:52:24"}
+{"current_steps": 75, "total_steps": 2365, "loss": 1.3869, "lr": 2.5316455696202537e-05, "epoch": 0.1583949313621964, "percentage": 3.17, "elapsed_time": "1:00:38", "remaining_time": "1 day, 6:51:38"}
+{"current_steps": 76, "total_steps": 2365, "loss": 1.4078, "lr": 2.5654008438818568e-05, "epoch": 0.16050686378035903, "percentage": 3.21, "elapsed_time": "1:01:26", "remaining_time": "1 day, 6:50:40"}
+{"current_steps": 77, "total_steps": 2365, "loss": 1.4175, "lr": 2.59915611814346e-05, "epoch": 0.16261879619852165, "percentage": 3.26, "elapsed_time": "1:02:15", "remaining_time": "1 day, 6:49:43"}
+{"current_steps": 78, "total_steps": 2365, "loss": 1.3896, "lr": 2.6329113924050635e-05, "epoch": 0.16473072861668428, "percentage": 3.3, "elapsed_time": "1:03:03", "remaining_time": "1 day, 6:48:50"}
+{"current_steps": 79, "total_steps": 2365, "loss": 1.3933, "lr": 2.6666666666666667e-05, "epoch": 0.16684266103484688, "percentage": 3.34, "elapsed_time": "1:03:51", "remaining_time": "1 day, 6:47:54"}
+{"current_steps": 80, "total_steps": 2365, "loss": 1.3782, "lr": 2.70042194092827e-05, "epoch": 0.1689545934530095, "percentage": 3.38, "elapsed_time": "1:04:39", "remaining_time": "1 day, 6:46:49"}
+{"current_steps": 81, "total_steps": 2365, "loss": 1.3772, "lr": 2.7341772151898737e-05, "epoch": 0.17106652587117213, "percentage": 3.42, "elapsed_time": "1:05:28", "remaining_time": "1 day, 6:46:00"}
+{"current_steps": 82, "total_steps": 2365, "loss": 1.3746, "lr": 2.767932489451477e-05, "epoch": 0.17317845828933473, "percentage": 3.47, "elapsed_time": "1:06:16", "remaining_time": "1 day, 6:45:01"}
+{"current_steps": 83, "total_steps": 2365, "loss": 1.3816, "lr": 2.8016877637130803e-05, "epoch": 0.17529039070749736, "percentage": 3.51, "elapsed_time": "1:07:04", "remaining_time": "1 day, 6:44:13"}
+{"current_steps": 84, "total_steps": 2365, "loss": 1.3731, "lr": 2.835443037974684e-05, "epoch": 0.17740232312565998, "percentage": 3.55, "elapsed_time": "1:07:52", "remaining_time": "1 day, 6:43:16"}
+{"current_steps": 85, "total_steps": 2365, "loss": 1.3911, "lr": 2.8691983122362873e-05, "epoch": 0.1795142555438226, "percentage": 3.59, "elapsed_time": "1:08:40", "remaining_time": "1 day, 6:42:09"}
+{"current_steps": 86, "total_steps": 2365, "loss": 1.3893, "lr": 2.9029535864978905e-05, "epoch": 0.1816261879619852, "percentage": 3.64, "elapsed_time": "1:09:28", "remaining_time": "1 day, 6:41:08"}
+{"current_steps": 87, "total_steps": 2365, "loss": 1.3732, "lr": 2.936708860759494e-05, "epoch": 0.18373812038014783, "percentage": 3.68, "elapsed_time": "1:10:16", "remaining_time": "1 day, 6:40:14"}
+{"current_steps": 88, "total_steps": 2365, "loss": 1.3718, "lr": 2.9704641350210975e-05, "epoch": 0.18585005279831046, "percentage": 3.72, "elapsed_time": "1:11:04", "remaining_time": "1 day, 6:39:09"}
+{"current_steps": 89, "total_steps": 2365, "loss": 1.3919, "lr": 3.0042194092827007e-05, "epoch": 0.18796198521647306, "percentage": 3.76, "elapsed_time": "1:11:52", "remaining_time": "1 day, 6:38:13"}
+{"current_steps": 90, "total_steps": 2365, "loss": 1.3815, "lr": 3.0379746835443042e-05, "epoch": 0.1900739176346357, "percentage": 3.81, "elapsed_time": "1:12:40", "remaining_time": "1 day, 6:37:16"}
+{"current_steps": 91, "total_steps": 2365, "loss": 1.3795, "lr": 3.0717299578059074e-05, "epoch": 0.1921858500527983, "percentage": 3.85, "elapsed_time": "1:13:29", "remaining_time": "1 day, 6:36:20"}
+{"current_steps": 92, "total_steps": 2365, "loss": 1.3792, "lr": 3.105485232067511e-05, "epoch": 0.19429778247096094, "percentage": 3.89, "elapsed_time": "1:14:17", "remaining_time": "1 day, 6:35:23"}
+{"current_steps": 93, "total_steps": 2365, "loss": 1.3747, "lr": 3.1392405063291144e-05, "epoch": 0.19640971488912354, "percentage": 3.93, "elapsed_time": "1:15:05", "remaining_time": "1 day, 6:34:27"}
+{"current_steps": 94, "total_steps": 2365, "loss": 1.3592, "lr": 3.172995780590717e-05, "epoch": 0.19852164730728616, "percentage": 3.97, "elapsed_time": "1:15:53", "remaining_time": "1 day, 6:33:33"}
+{"current_steps": 95, "total_steps": 2365, "loss": 1.3786, "lr": 3.206751054852321e-05, "epoch": 0.2006335797254488, "percentage": 4.02, "elapsed_time": "1:16:42", "remaining_time": "1 day, 6:33:01"}
+{"current_steps": 96, "total_steps": 2365, "loss": 1.3591, "lr": 3.240506329113924e-05, "epoch": 0.20274551214361142, "percentage": 4.06, "elapsed_time": "1:17:31", "remaining_time": "1 day, 6:32:15"}
+{"current_steps": 97, "total_steps": 2365, "loss": 1.388, "lr": 3.274261603375528e-05, "epoch": 0.20485744456177402, "percentage": 4.1, "elapsed_time": "1:18:20", "remaining_time": "1 day, 6:31:42"}
+{"current_steps": 98, "total_steps": 2365, "loss": 1.3676, "lr": 3.308016877637131e-05, "epoch": 0.20696937697993664, "percentage": 4.14, "elapsed_time": "1:19:08", "remaining_time": "1 day, 6:30:46"}
+{"current_steps": 99, "total_steps": 2365, "loss": 1.3675, "lr": 3.341772151898735e-05, "epoch": 0.20908130939809927, "percentage": 4.19, "elapsed_time": "1:19:57", "remaining_time": "1 day, 6:30:18"}
+{"current_steps": 100, "total_steps": 2365, "loss": 1.3882, "lr": 3.375527426160338e-05, "epoch": 0.21119324181626187, "percentage": 4.23, "elapsed_time": "1:20:46", "remaining_time": "1 day, 6:29:23"}
+{"current_steps": 101, "total_steps": 2365, "loss": 1.3676, "lr": 3.409282700421941e-05, "epoch": 0.2133051742344245, "percentage": 4.27, "elapsed_time": "1:21:34", "remaining_time": "1 day, 6:28:23"}
+{"current_steps": 102, "total_steps": 2365, "loss": 1.3759, "lr": 3.4430379746835445e-05, "epoch": 0.21541710665258712, "percentage": 4.31, "elapsed_time": "1:22:21", "remaining_time": "1 day, 6:27:21"}
+{"current_steps": 103, "total_steps": 2365, "loss": 1.3539, "lr": 3.476793248945148e-05, "epoch": 0.21752903907074975, "percentage": 4.36, "elapsed_time": "1:23:10", "remaining_time": "1 day, 6:26:32"}
+{"current_steps": 104, "total_steps": 2365, "loss": 1.3805, "lr": 3.5105485232067516e-05, "epoch": 0.21964097148891235, "percentage": 4.4, "elapsed_time": "1:23:58", "remaining_time": "1 day, 6:25:30"}
+{"current_steps": 105, "total_steps": 2365, "loss": 1.3887, "lr": 3.5443037974683544e-05, "epoch": 0.22175290390707497, "percentage": 4.44, "elapsed_time": "1:24:46", "remaining_time": "1 day, 6:24:41"}
+{"current_steps": 106, "total_steps": 2365, "loss": 1.3654, "lr": 3.578059071729958e-05, "epoch": 0.2238648363252376, "percentage": 4.48, "elapsed_time": "1:25:34", "remaining_time": "1 day, 6:23:50"}
+{"current_steps": 107, "total_steps": 2365, "loss": 1.3671, "lr": 3.6118143459915614e-05, "epoch": 0.22597676874340022, "percentage": 4.52, "elapsed_time": "1:26:23", "remaining_time": "1 day, 6:23:01"}
+{"current_steps": 108, "total_steps": 2365, "loss": 1.3751, "lr": 3.645569620253165e-05, "epoch": 0.22808870116156282, "percentage": 4.57, "elapsed_time": "1:27:11", "remaining_time": "1 day, 6:22:05"}
+{"current_steps": 109, "total_steps": 2365, "loss": 1.3798, "lr": 3.6793248945147684e-05, "epoch": 0.23020063357972545, "percentage": 4.61, "elapsed_time": "1:27:59", "remaining_time": "1 day, 6:21:09"}
+{"current_steps": 110, "total_steps": 2365, "loss": 1.3668, "lr": 3.713080168776372e-05, "epoch": 0.23231256599788808, "percentage": 4.65, "elapsed_time": "1:28:47", "remaining_time": "1 day, 6:20:17"}
+{"current_steps": 111, "total_steps": 2365, "loss": 1.3794, "lr": 3.746835443037975e-05, "epoch": 0.23442449841605068, "percentage": 4.69, "elapsed_time": "1:29:36", "remaining_time": "1 day, 6:19:31"}
+{"current_steps": 112, "total_steps": 2365, "loss": 1.3877, "lr": 3.780590717299578e-05, "epoch": 0.2365364308342133, "percentage": 4.74, "elapsed_time": "1:30:24", "remaining_time": "1 day, 6:18:43"}
+{"current_steps": 113, "total_steps": 2365, "loss": 1.3844, "lr": 3.814345991561182e-05, "epoch": 0.23864836325237593, "percentage": 4.78, "elapsed_time": "1:31:12", "remaining_time": "1 day, 6:17:45"}
+{"current_steps": 114, "total_steps": 2365, "loss": 1.3742, "lr": 3.848101265822785e-05, "epoch": 0.24076029567053855, "percentage": 4.82, "elapsed_time": "1:32:00", "remaining_time": "1 day, 6:16:54"}
+{"current_steps": 115, "total_steps": 2365, "loss": 1.3486, "lr": 3.881856540084389e-05, "epoch": 0.24287222808870115, "percentage": 4.86, "elapsed_time": "1:32:48", "remaining_time": "1 day, 6:15:54"}
+{"current_steps": 116, "total_steps": 2365, "loss": 1.3573, "lr": 3.915611814345992e-05, "epoch": 0.24498416050686378, "percentage": 4.9, "elapsed_time": "1:33:36", "remaining_time": "1 day, 6:14:56"}
+{"current_steps": 117, "total_steps": 2365, "loss": 1.371, "lr": 3.949367088607596e-05, "epoch": 0.2470960929250264, "percentage": 4.95, "elapsed_time": "1:34:24", "remaining_time": "1 day, 6:14:04"}
+{"current_steps": 118, "total_steps": 2365, "loss": 1.3732, "lr": 3.9831223628691986e-05, "epoch": 0.249208025343189, "percentage": 4.99, "elapsed_time": "1:35:13", "remaining_time": "1 day, 6:13:11"}
+{"current_steps": 119, "total_steps": 2365, "loss": 1.3824, "lr": 4.016877637130802e-05, "epoch": 0.25131995776135163, "percentage": 5.03, "elapsed_time": "1:36:01", "remaining_time": "1 day, 6:12:22"}
+{"current_steps": 120, "total_steps": 2365, "loss": 1.3813, "lr": 4.0506329113924056e-05, "epoch": 0.25343189017951423, "percentage": 5.07, "elapsed_time": "1:36:49", "remaining_time": "1 day, 6:11:26"}
+{"current_steps": 121, "total_steps": 2365, "loss": 1.3651, "lr": 4.0843881856540084e-05, "epoch": 0.2555438225976769, "percentage": 5.12, "elapsed_time": "1:37:37", "remaining_time": "1 day, 6:10:29"}
+{"current_steps": 122, "total_steps": 2365, "loss": 1.3617, "lr": 4.1181434599156126e-05, "epoch": 0.2576557550158395, "percentage": 5.16, "elapsed_time": "1:38:25", "remaining_time": "1 day, 6:09:32"}
+{"current_steps": 123, "total_steps": 2365, "loss": 1.3793, "lr": 4.1518987341772154e-05, "epoch": 0.25976768743400214, "percentage": 5.2, "elapsed_time": "1:39:13", "remaining_time": "1 day, 6:08:31"}
+{"current_steps": 124, "total_steps": 2365, "loss": 1.3731, "lr": 4.1856540084388196e-05, "epoch": 0.26187961985216474, "percentage": 5.24, "elapsed_time": "1:40:00", "remaining_time": "1 day, 6:07:33"}
+{"current_steps": 125, "total_steps": 2365, "loss": 1.3483, "lr": 4.2194092827004224e-05, "epoch": 0.26399155227032733, "percentage": 5.29, "elapsed_time": "1:40:48", "remaining_time": "1 day, 6:06:35"}
+{"current_steps": 126, "total_steps": 2365, "loss": 1.3632, "lr": 4.253164556962025e-05, "epoch": 0.26610348468849, "percentage": 5.33, "elapsed_time": "1:41:36", "remaining_time": "1 day, 6:05:36"}
+{"current_steps": 127, "total_steps": 2365, "loss": 1.3635, "lr": 4.286919831223629e-05, "epoch": 0.2682154171066526, "percentage": 5.37, "elapsed_time": "1:42:24", "remaining_time": "1 day, 6:04:43"}
+{"current_steps": 128, "total_steps": 2365, "loss": 1.3637, "lr": 4.320675105485232e-05, "epoch": 0.2703273495248152, "percentage": 5.41, "elapsed_time": "1:43:13", "remaining_time": "1 day, 6:03:55"}
+{"current_steps": 129, "total_steps": 2365, "loss": 1.3571, "lr": 4.354430379746836e-05, "epoch": 0.27243928194297784, "percentage": 5.45, "elapsed_time": "1:44:01", "remaining_time": "1 day, 6:02:57"}
+{"current_steps": 130, "total_steps": 2365, "loss": 1.3785, "lr": 4.3881856540084386e-05, "epoch": 0.27455121436114044, "percentage": 5.5, "elapsed_time": "1:44:49", "remaining_time": "1 day, 6:02:05"}
+{"current_steps": 131, "total_steps": 2365, "loss": 1.3558, "lr": 4.421940928270043e-05, "epoch": 0.27666314677930304, "percentage": 5.54, "elapsed_time": "1:45:37", "remaining_time": "1 day, 6:01:12"}
+{"current_steps": 132, "total_steps": 2365, "loss": 1.3642, "lr": 4.4556962025316456e-05, "epoch": 0.2787750791974657, "percentage": 5.58, "elapsed_time": "1:46:25", "remaining_time": "1 day, 6:00:21"}
+{"current_steps": 133, "total_steps": 2365, "loss": 1.3414, "lr": 4.48945147679325e-05, "epoch": 0.2808870116156283, "percentage": 5.62, "elapsed_time": "1:47:13", "remaining_time": "1 day, 5:59:23"}
+{"current_steps": 134, "total_steps": 2365, "loss": 1.3559, "lr": 4.5232067510548526e-05, "epoch": 0.28299894403379094, "percentage": 5.67, "elapsed_time": "1:48:01", "remaining_time": "1 day, 5:58:25"}
+{"current_steps": 135, "total_steps": 2365, "loss": 1.3741, "lr": 4.556962025316456e-05, "epoch": 0.28511087645195354, "percentage": 5.71, "elapsed_time": "1:48:49", "remaining_time": "1 day, 5:57:32"}
+{"current_steps": 136, "total_steps": 2365, "loss": 1.3665, "lr": 4.5907172995780596e-05, "epoch": 0.28722280887011614, "percentage": 5.75, "elapsed_time": "1:49:37", "remaining_time": "1 day, 5:56:35"}
+{"current_steps": 137, "total_steps": 2365, "loss": 1.3875, "lr": 4.624472573839663e-05, "epoch": 0.2893347412882788, "percentage": 5.79, "elapsed_time": "1:50:25", "remaining_time": "1 day, 5:55:54"}
+{"current_steps": 138, "total_steps": 2365, "loss": 1.3711, "lr": 4.658227848101266e-05, "epoch": 0.2914466737064414, "percentage": 5.84, "elapsed_time": "1:51:14", "remaining_time": "1 day, 5:55:04"}
+{"current_steps": 139, "total_steps": 2365, "loss": 1.3579, "lr": 4.69198312236287e-05, "epoch": 0.293558606124604, "percentage": 5.88, "elapsed_time": "1:52:02", "remaining_time": "1 day, 5:54:15"}
+{"current_steps": 140, "total_steps": 2365, "loss": 1.3613, "lr": 4.725738396624473e-05, "epoch": 0.29567053854276665, "percentage": 5.92, "elapsed_time": "1:52:50", "remaining_time": "1 day, 5:53:24"}
+{"current_steps": 141, "total_steps": 2365, "loss": 1.369, "lr": 4.759493670886076e-05, "epoch": 0.29778247096092925, "percentage": 5.96, "elapsed_time": "1:53:39", "remaining_time": "1 day, 5:52:37"}
+{"current_steps": 142, "total_steps": 2365, "loss": 1.3671, "lr": 4.79324894514768e-05, "epoch": 0.29989440337909185, "percentage": 6.0, "elapsed_time": "1:54:27", "remaining_time": "1 day, 5:51:44"}
+{"current_steps": 143, "total_steps": 2365, "loss": 1.3656, "lr": 4.827004219409283e-05, "epoch": 0.3020063357972545, "percentage": 6.05, "elapsed_time": "1:55:15", "remaining_time": "1 day, 5:50:59"}
+{"current_steps": 144, "total_steps": 2365, "loss": 1.3491, "lr": 4.860759493670886e-05, "epoch": 0.3041182682154171, "percentage": 6.09, "elapsed_time": "1:56:04", "remaining_time": "1 day, 5:50:12"}
+{"current_steps": 145, "total_steps": 2365, "loss": 1.3564, "lr": 4.89451476793249e-05, "epoch": 0.30623020063357975, "percentage": 6.13, "elapsed_time": "1:56:52", "remaining_time": "1 day, 5:49:23"}
+{"current_steps": 146, "total_steps": 2365, "loss": 1.3687, "lr": 4.928270042194093e-05, "epoch": 0.30834213305174235, "percentage": 6.17, "elapsed_time": "1:57:40", "remaining_time": "1 day, 5:48:31"}
+{"current_steps": 147, "total_steps": 2365, "loss": 1.3547, "lr": 4.962025316455696e-05, "epoch": 0.31045406546990495, "percentage": 6.22, "elapsed_time": "1:58:28", "remaining_time": "1 day, 5:47:37"}
+{"current_steps": 148, "total_steps": 2365, "loss": 1.3532, "lr": 4.9957805907173e-05, "epoch": 0.3125659978880676, "percentage": 6.26, "elapsed_time": "1:59:16", "remaining_time": "1 day, 5:46:41"}
+{"current_steps": 149, "total_steps": 2365, "loss": 1.3535, "lr": 5.029535864978903e-05, "epoch": 0.3146779303062302, "percentage": 6.3, "elapsed_time": "2:00:04", "remaining_time": "1 day, 5:45:55"}
+{"current_steps": 150, "total_steps": 2365, "loss": 1.3506, "lr": 5.063291139240507e-05, "epoch": 0.3167898627243928, "percentage": 6.34, "elapsed_time": "2:00:52", "remaining_time": "1 day, 5:45:02"}
+{"current_steps": 151, "total_steps": 2365, "loss": 1.3335, "lr": 5.09704641350211e-05, "epoch": 0.31890179514255546, "percentage": 6.38, "elapsed_time": "2:01:41", "remaining_time": "1 day, 5:44:13"}
+{"current_steps": 152, "total_steps": 2365, "loss": 1.3556, "lr": 5.1308016877637136e-05, "epoch": 0.32101372756071805, "percentage": 6.43, "elapsed_time": "2:02:29", "remaining_time": "1 day, 5:43:25"}
+{"current_steps": 153, "total_steps": 2365, "loss": 1.3556, "lr": 5.164556962025317e-05, "epoch": 0.32312565997888065, "percentage": 6.47, "elapsed_time": "2:03:18", "remaining_time": "1 day, 5:42:38"}
+{"current_steps": 154, "total_steps": 2365, "loss": 1.3604, "lr": 5.19831223628692e-05, "epoch": 0.3252375923970433, "percentage": 6.51, "elapsed_time": "2:04:06", "remaining_time": "1 day, 5:41:54"}
+{"current_steps": 155, "total_steps": 2365, "loss": 1.368, "lr": 5.2320675105485235e-05, "epoch": 0.3273495248152059, "percentage": 6.55, "elapsed_time": "2:04:55", "remaining_time": "1 day, 5:41:05"}
+{"current_steps": 156, "total_steps": 2365, "loss": 1.3454, "lr": 5.265822784810127e-05, "epoch": 0.32946145723336856, "percentage": 6.6, "elapsed_time": "2:05:43", "remaining_time": "1 day, 5:40:14"}
+{"current_steps": 157, "total_steps": 2365, "loss": 1.3518, "lr": 5.2995780590717305e-05, "epoch": 0.33157338965153116, "percentage": 6.64, "elapsed_time": "2:06:31", "remaining_time": "1 day, 5:39:23"}
+{"current_steps": 158, "total_steps": 2365, "loss": 1.3602, "lr": 5.333333333333333e-05, "epoch": 0.33368532206969376, "percentage": 6.68, "elapsed_time": "2:07:19", "remaining_time": "1 day, 5:38:36"}
+{"current_steps": 159, "total_steps": 2365, "loss": 1.3608, "lr": 5.3670886075949375e-05, "epoch": 0.3357972544878564, "percentage": 6.72, "elapsed_time": "2:08:08", "remaining_time": "1 day, 5:37:46"}
+{"current_steps": 160, "total_steps": 2365, "loss": 1.3594, "lr": 5.40084388185654e-05, "epoch": 0.337909186906019, "percentage": 6.77, "elapsed_time": "2:08:56", "remaining_time": "1 day, 5:36:54"}
+{"current_steps": 161, "total_steps": 2365, "loss": 1.3421, "lr": 5.434599156118144e-05, "epoch": 0.3400211193241816, "percentage": 6.81, "elapsed_time": "2:09:44", "remaining_time": "1 day, 5:36:07"}
+{"current_steps": 162, "total_steps": 2365, "loss": 1.3557, "lr": 5.468354430379747e-05, "epoch": 0.34213305174234426, "percentage": 6.85, "elapsed_time": "2:10:32", "remaining_time": "1 day, 5:35:15"}
+{"current_steps": 163, "total_steps": 2365, "loss": 1.3716, "lr": 5.502109704641351e-05, "epoch": 0.34424498416050686, "percentage": 6.89, "elapsed_time": "2:11:20", "remaining_time": "1 day, 5:34:21"}
+{"current_steps": 164, "total_steps": 2365, "loss": 1.3614, "lr": 5.535864978902954e-05, "epoch": 0.34635691657866946, "percentage": 6.93, "elapsed_time": "2:12:08", "remaining_time": "1 day, 5:33:27"}
+{"current_steps": 165, "total_steps": 2365, "loss": 1.3459, "lr": 5.569620253164558e-05, "epoch": 0.3484688489968321, "percentage": 6.98, "elapsed_time": "2:12:56", "remaining_time": "1 day, 5:32:34"}
+{"current_steps": 166, "total_steps": 2365, "loss": 1.3607, "lr": 5.603375527426161e-05, "epoch": 0.3505807814149947, "percentage": 7.02, "elapsed_time": "2:13:45", "remaining_time": "1 day, 5:31:47"}
+{"current_steps": 167, "total_steps": 2365, "loss": 1.3596, "lr": 5.637130801687765e-05, "epoch": 0.3526927138331573, "percentage": 7.06, "elapsed_time": "2:14:33", "remaining_time": "1 day, 5:30:56"}
+{"current_steps": 168, "total_steps": 2365, "loss": 1.3472, "lr": 5.670886075949368e-05, "epoch": 0.35480464625131997, "percentage": 7.1, "elapsed_time": "2:15:21", "remaining_time": "1 day, 5:30:06"}
+{"current_steps": 169, "total_steps": 2365, "loss": 1.349, "lr": 5.7046413502109705e-05, "epoch": 0.35691657866948256, "percentage": 7.15, "elapsed_time": "2:16:09", "remaining_time": "1 day, 5:29:15"}
+{"current_steps": 170, "total_steps": 2365, "loss": 1.3573, "lr": 5.738396624472575e-05, "epoch": 0.3590285110876452, "percentage": 7.19, "elapsed_time": "2:16:57", "remaining_time": "1 day, 5:28:25"}
+{"current_steps": 171, "total_steps": 2365, "loss": 1.3547, "lr": 5.7721518987341775e-05, "epoch": 0.3611404435058078, "percentage": 7.23, "elapsed_time": "2:17:45", "remaining_time": "1 day, 5:27:35"}
+{"current_steps": 172, "total_steps": 2365, "loss": 1.3559, "lr": 5.805907172995781e-05, "epoch": 0.3632523759239704, "percentage": 7.27, "elapsed_time": "2:18:34", "remaining_time": "1 day, 5:26:46"}
+{"current_steps": 173, "total_steps": 2365, "loss": 1.3378, "lr": 5.839662447257384e-05, "epoch": 0.36536430834213307, "percentage": 7.32, "elapsed_time": "2:19:22", "remaining_time": "1 day, 5:25:58"}
+{"current_steps": 174, "total_steps": 2365, "loss": 1.3467, "lr": 5.873417721518988e-05, "epoch": 0.36747624076029567, "percentage": 7.36, "elapsed_time": "2:20:10", "remaining_time": "1 day, 5:25:09"}
+{"current_steps": 175, "total_steps": 2365, "loss": 1.3436, "lr": 5.907172995780591e-05, "epoch": 0.36958817317845827, "percentage": 7.4, "elapsed_time": "2:20:59", "remaining_time": "1 day, 5:24:21"}
+{"current_steps": 176, "total_steps": 2365, "loss": 1.3493, "lr": 5.940928270042195e-05, "epoch": 0.3717001055966209, "percentage": 7.44, "elapsed_time": "2:21:47", "remaining_time": "1 day, 5:23:32"}
+{"current_steps": 177, "total_steps": 2365, "loss": 1.3614, "lr": 5.974683544303798e-05, "epoch": 0.3738120380147835, "percentage": 7.48, "elapsed_time": "2:22:36", "remaining_time": "1 day, 5:22:45"}
+{"current_steps": 178, "total_steps": 2365, "loss": 1.3568, "lr": 6.0084388185654014e-05, "epoch": 0.3759239704329461, "percentage": 7.53, "elapsed_time": "2:23:24", "remaining_time": "1 day, 5:21:57"}
+{"current_steps": 179, "total_steps": 2365, "loss": 1.3667, "lr": 6.042194092827005e-05, "epoch": 0.3780359028511088, "percentage": 7.57, "elapsed_time": "2:24:12", "remaining_time": "1 day, 5:21:05"}
+{"current_steps": 180, "total_steps": 2365, "loss": 1.3686, "lr": 6.0759493670886084e-05, "epoch": 0.3801478352692714, "percentage": 7.61, "elapsed_time": "2:25:00", "remaining_time": "1 day, 5:20:14"}
+{"current_steps": 181, "total_steps": 2365, "loss": 1.3611, "lr": 6.109704641350211e-05, "epoch": 0.382259767687434, "percentage": 7.65, "elapsed_time": "2:25:48", "remaining_time": "1 day, 5:19:25"}
+{"current_steps": 182, "total_steps": 2365, "loss": 1.3382, "lr": 6.143459915611815e-05, "epoch": 0.3843717001055966, "percentage": 7.7, "elapsed_time": "2:26:37", "remaining_time": "1 day, 5:18:36"}
+{"current_steps": 183, "total_steps": 2365, "loss": 1.3543, "lr": 6.177215189873418e-05, "epoch": 0.3864836325237592, "percentage": 7.74, "elapsed_time": "2:27:24", "remaining_time": "1 day, 5:17:43"}
+{"current_steps": 184, "total_steps": 2365, "loss": 1.3754, "lr": 6.210970464135022e-05, "epoch": 0.3885955649419219, "percentage": 7.78, "elapsed_time": "2:28:12", "remaining_time": "1 day, 5:16:49"}
+{"current_steps": 185, "total_steps": 2365, "loss": 1.3745, "lr": 6.244725738396625e-05, "epoch": 0.3907074973600845, "percentage": 7.82, "elapsed_time": "2:29:01", "remaining_time": "1 day, 5:16:00"}
+{"current_steps": 186, "total_steps": 2365, "loss": 1.3587, "lr": 6.278481012658229e-05, "epoch": 0.3928194297782471, "percentage": 7.86, "elapsed_time": "2:29:49", "remaining_time": "1 day, 5:15:16"}
+{"current_steps": 187, "total_steps": 2365, "loss": 1.3525, "lr": 6.312236286919832e-05, "epoch": 0.39493136219640973, "percentage": 7.91, "elapsed_time": "2:30:37", "remaining_time": "1 day, 5:14:22"}
+{"current_steps": 188, "total_steps": 2365, "loss": 1.353, "lr": 6.345991561181434e-05, "epoch": 0.39704329461457233, "percentage": 7.95, "elapsed_time": "2:31:26", "remaining_time": "1 day, 5:13:41"}
+{"current_steps": 189, "total_steps": 2365, "loss": 1.3452, "lr": 6.379746835443039e-05, "epoch": 0.3991552270327349, "percentage": 7.99, "elapsed_time": "2:32:15", "remaining_time": "1 day, 5:12:54"}
+{"current_steps": 190, "total_steps": 2365, "loss": 1.3494, "lr": 6.413502109704641e-05, "epoch": 0.4012671594508976, "percentage": 8.03, "elapsed_time": "2:33:03", "remaining_time": "1 day, 5:12:07"}
+{"current_steps": 191, "total_steps": 2365, "loss": 1.3487, "lr": 6.447257383966245e-05, "epoch": 0.4033790918690602, "percentage": 8.08, "elapsed_time": "2:33:52", "remaining_time": "1 day, 5:11:24"}
+{"current_steps": 192, "total_steps": 2365, "loss": 1.3452, "lr": 6.481012658227848e-05, "epoch": 0.40549102428722283, "percentage": 8.12, "elapsed_time": "2:34:41", "remaining_time": "1 day, 5:10:40"}
+{"current_steps": 193, "total_steps": 2365, "loss": 1.3623, "lr": 6.514767932489452e-05, "epoch": 0.40760295670538543, "percentage": 8.16, "elapsed_time": "2:35:29", "remaining_time": "1 day, 5:09:50"}
+{"current_steps": 194, "total_steps": 2365, "loss": 1.3558, "lr": 6.548523206751055e-05, "epoch": 0.40971488912354803, "percentage": 8.2, "elapsed_time": "2:36:17", "remaining_time": "1 day, 5:08:59"}
+{"current_steps": 195, "total_steps": 2365, "loss": 1.3638, "lr": 6.582278481012659e-05, "epoch": 0.4118268215417107, "percentage": 8.25, "elapsed_time": "2:37:06", "remaining_time": "1 day, 5:08:15"}
+{"current_steps": 196, "total_steps": 2365, "loss": 1.3605, "lr": 6.616033755274262e-05, "epoch": 0.4139387539598733, "percentage": 8.29, "elapsed_time": "2:37:54", "remaining_time": "1 day, 5:07:25"}
+{"current_steps": 197, "total_steps": 2365, "loss": 1.3447, "lr": 6.649789029535865e-05, "epoch": 0.4160506863780359, "percentage": 8.33, "elapsed_time": "2:38:42", "remaining_time": "1 day, 5:06:34"}
+{"current_steps": 198, "total_steps": 2365, "loss": 1.3519, "lr": 6.68354430379747e-05, "epoch": 0.41816261879619854, "percentage": 8.37, "elapsed_time": "2:39:30", "remaining_time": "1 day, 5:05:41"}
+{"current_steps": 199, "total_steps": 2365, "loss": 1.3379, "lr": 6.717299578059072e-05, "epoch": 0.42027455121436114, "percentage": 8.41, "elapsed_time": "2:40:18", "remaining_time": "1 day, 5:04:50"}
+{"current_steps": 200, "total_steps": 2365, "loss": 1.3409, "lr": 6.751054852320676e-05, "epoch": 0.42238648363252373, "percentage": 8.46, "elapsed_time": "2:41:06", "remaining_time": "1 day, 5:03:57"}
+{"current_steps": 201, "total_steps": 2365, "loss": 1.3404, "lr": 6.784810126582279e-05, "epoch": 0.4244984160506864, "percentage": 8.5, "elapsed_time": "2:41:54", "remaining_time": "1 day, 5:03:07"}
+{"current_steps": 202, "total_steps": 2365, "loss": 1.3403, "lr": 6.818565400843882e-05, "epoch": 0.426610348468849, "percentage": 8.54, "elapsed_time": "2:42:42", "remaining_time": "1 day, 5:02:16"}
+{"current_steps": 203, "total_steps": 2365, "loss": 1.3445, "lr": 6.852320675105486e-05, "epoch": 0.42872228088701164, "percentage": 8.58, "elapsed_time": "2:43:30", "remaining_time": "1 day, 5:01:21"}
+{"current_steps": 204, "total_steps": 2365, "loss": 1.3371, "lr": 6.886075949367089e-05, "epoch": 0.43083421330517424, "percentage": 8.63, "elapsed_time": "2:44:18", "remaining_time": "1 day, 5:00:35"}
+{"current_steps": 205, "total_steps": 2365, "loss": 1.3548, "lr": 6.919831223628693e-05, "epoch": 0.43294614572333684, "percentage": 8.67, "elapsed_time": "2:45:06", "remaining_time": "1 day, 4:59:43"}
+{"current_steps": 206, "total_steps": 2365, "loss": 1.3257, "lr": 6.953586497890296e-05, "epoch": 0.4350580781414995, "percentage": 8.71, "elapsed_time": "2:45:55", "remaining_time": "1 day, 4:58:56"}
+{"current_steps": 207, "total_steps": 2365, "loss": 1.345, "lr": 6.9873417721519e-05, "epoch": 0.4371700105596621, "percentage": 8.75, "elapsed_time": "2:46:43", "remaining_time": "1 day, 4:58:06"}
+{"current_steps": 208, "total_steps": 2365, "loss": 1.3513, "lr": 7.021097046413503e-05, "epoch": 0.4392819429778247, "percentage": 8.79, "elapsed_time": "2:47:31", "remaining_time": "1 day, 4:57:12"}
+{"current_steps": 209, "total_steps": 2365, "loss": 1.346, "lr": 7.054852320675107e-05, "epoch": 0.44139387539598735, "percentage": 8.84, "elapsed_time": "2:48:19", "remaining_time": "1 day, 4:56:19"}
+{"current_steps": 210, "total_steps": 2365, "loss": 1.3543, "lr": 7.088607594936709e-05, "epoch": 0.44350580781414994, "percentage": 8.88, "elapsed_time": "2:49:07", "remaining_time": "1 day, 4:55:30"}
+{"current_steps": 211, "total_steps": 2365, "loss": 1.3534, "lr": 7.122362869198312e-05, "epoch": 0.44561774023231254, "percentage": 8.92, "elapsed_time": "2:49:55", "remaining_time": "1 day, 4:54:42"}
+{"current_steps": 212, "total_steps": 2365, "loss": 1.3416, "lr": 7.156118143459916e-05, "epoch": 0.4477296726504752, "percentage": 8.96, "elapsed_time": "2:50:43", "remaining_time": "1 day, 4:53:50"}
+{"current_steps": 213, "total_steps": 2365, "loss": 1.3449, "lr": 7.189873417721519e-05, "epoch": 0.4498416050686378, "percentage": 9.01, "elapsed_time": "2:51:31", "remaining_time": "1 day, 4:53:00"}
+{"current_steps": 214, "total_steps": 2365, "loss": 1.3448, "lr": 7.223628691983123e-05, "epoch": 0.45195353748680045, "percentage": 9.05, "elapsed_time": "2:52:20", "remaining_time": "1 day, 4:52:15"}
+{"current_steps": 215, "total_steps": 2365, "loss": 1.3428, "lr": 7.257383966244726e-05, "epoch": 0.45406546990496305, "percentage": 9.09, "elapsed_time": "2:53:08", "remaining_time": "1 day, 4:51:27"}
+{"current_steps": 216, "total_steps": 2365, "loss": 1.3437, "lr": 7.29113924050633e-05, "epoch": 0.45617740232312565, "percentage": 9.13, "elapsed_time": "2:53:56", "remaining_time": "1 day, 4:50:36"}
+{"current_steps": 217, "total_steps": 2365, "loss": 1.3527, "lr": 7.324894514767933e-05, "epoch": 0.4582893347412883, "percentage": 9.18, "elapsed_time": "2:54:45", "remaining_time": "1 day, 4:49:51"}
+{"current_steps": 218, "total_steps": 2365, "loss": 1.3476, "lr": 7.358649789029537e-05, "epoch": 0.4604012671594509, "percentage": 9.22, "elapsed_time": "2:55:34", "remaining_time": "1 day, 4:49:07"}
+{"current_steps": 219, "total_steps": 2365, "loss": 1.3476, "lr": 7.39240506329114e-05, "epoch": 0.4625131995776135, "percentage": 9.26, "elapsed_time": "2:56:22", "remaining_time": "1 day, 4:48:22"}
+{"current_steps": 220, "total_steps": 2365, "loss": 1.3616, "lr": 7.426160337552744e-05, "epoch": 0.46462513199577615, "percentage": 9.3, "elapsed_time": "2:57:11", "remaining_time": "1 day, 4:47:32"}
+{"current_steps": 221, "total_steps": 2365, "loss": 1.3436, "lr": 7.459915611814347e-05, "epoch": 0.46673706441393875, "percentage": 9.34, "elapsed_time": "2:57:59", "remaining_time": "1 day, 4:46:44"}
+{"current_steps": 222, "total_steps": 2365, "loss": 1.346, "lr": 7.49367088607595e-05, "epoch": 0.46884899683210135, "percentage": 9.39, "elapsed_time": "2:58:47", "remaining_time": "1 day, 4:45:52"}
+{"current_steps": 223, "total_steps": 2365, "loss": 1.3474, "lr": 7.527426160337554e-05, "epoch": 0.470960929250264, "percentage": 9.43, "elapsed_time": "2:59:35", "remaining_time": "1 day, 4:45:03"}
+{"current_steps": 224, "total_steps": 2365, "loss": 1.3529, "lr": 7.561181434599156e-05, "epoch": 0.4730728616684266, "percentage": 9.47, "elapsed_time": "3:00:23", "remaining_time": "1 day, 4:44:13"}
+{"current_steps": 225, "total_steps": 2365, "loss": 1.3331, "lr": 7.59493670886076e-05, "epoch": 0.4751847940865892, "percentage": 9.51, "elapsed_time": "3:01:11", "remaining_time": "1 day, 4:43:21"}
+{"current_steps": 226, "total_steps": 2365, "loss": 1.3355, "lr": 7.628691983122363e-05, "epoch": 0.47729672650475186, "percentage": 9.56, "elapsed_time": "3:01:59", "remaining_time": "1 day, 4:42:32"}
+{"current_steps": 227, "total_steps": 2365, "loss": 1.3588, "lr": 7.662447257383967e-05, "epoch": 0.47940865892291445, "percentage": 9.6, "elapsed_time": "3:02:48", "remaining_time": "1 day, 4:41:43"}
+{"current_steps": 228, "total_steps": 2365, "loss": 1.352, "lr": 7.69620253164557e-05, "epoch": 0.4815205913410771, "percentage": 9.64, "elapsed_time": "3:03:36", "remaining_time": "1 day, 4:40:57"}
+{"current_steps": 229, "total_steps": 2365, "loss": 1.3434, "lr": 7.729957805907174e-05, "epoch": 0.4836325237592397, "percentage": 9.68, "elapsed_time": "3:04:24", "remaining_time": "1 day, 4:40:06"}
+{"current_steps": 230, "total_steps": 2365, "loss": 1.3494, "lr": 7.763713080168777e-05, "epoch": 0.4857444561774023, "percentage": 9.73, "elapsed_time": "3:05:12", "remaining_time": "1 day, 4:39:14"}
+{"current_steps": 231, "total_steps": 2365, "loss": 1.3627, "lr": 7.79746835443038e-05, "epoch": 0.48785638859556496, "percentage": 9.77, "elapsed_time": "3:06:00", "remaining_time": "1 day, 4:38:23"}
+{"current_steps": 232, "total_steps": 2365, "loss": 1.3665, "lr": 7.831223628691984e-05, "epoch": 0.48996832101372756, "percentage": 9.81, "elapsed_time": "3:06:48", "remaining_time": "1 day, 4:37:34"}
+{"current_steps": 233, "total_steps": 2365, "loss": 1.362, "lr": 7.864978902953587e-05, "epoch": 0.49208025343189016, "percentage": 9.85, "elapsed_time": "3:07:37", "remaining_time": "1 day, 4:36:46"}
+{"current_steps": 234, "total_steps": 2365, "loss": 1.3613, "lr": 7.898734177215191e-05, "epoch": 0.4941921858500528, "percentage": 9.89, "elapsed_time": "3:08:25", "remaining_time": "1 day, 4:35:58"}
+{"current_steps": 235, "total_steps": 2365, "loss": 1.3515, "lr": 7.932489451476794e-05, "epoch": 0.4963041182682154, "percentage": 9.94, "elapsed_time": "3:09:13", "remaining_time": "1 day, 4:35:10"}
+{"current_steps": 236, "total_steps": 2365, "loss": 1.3573, "lr": 7.966244725738397e-05, "epoch": 0.498416050686378, "percentage": 9.98, "elapsed_time": "3:10:02", "remaining_time": "1 day, 4:34:22"}
+{"current_steps": 237, "total_steps": 2365, "loss": 1.3404, "lr": 8e-05, "epoch": 0.5005279831045406, "percentage": 10.02, "elapsed_time": "3:10:50", "remaining_time": "1 day, 4:33:34"}
+{"current_steps": 238, "total_steps": 2365, "loss": 1.3203, "lr": 7.999995641004559e-05, "epoch": 0.5026399155227033, "percentage": 10.06, "elapsed_time": "3:11:39", "remaining_time": "1 day, 4:32:47"}
+{"current_steps": 239, "total_steps": 2365, "loss": 1.3595, "lr": 7.999982564027732e-05, "epoch": 0.5047518479408659, "percentage": 10.11, "elapsed_time": "3:12:27", "remaining_time": "1 day, 4:31:58"}
+{"current_steps": 240, "total_steps": 2365, "loss": 1.3443, "lr": 7.999960769098022e-05, "epoch": 0.5068637803590285, "percentage": 10.15, "elapsed_time": "3:13:15", "remaining_time": "1 day, 4:31:09"}
+{"current_steps": 241, "total_steps": 2365, "loss": 1.3323, "lr": 7.999930256262932e-05, "epoch": 0.5089757127771911, "percentage": 10.19, "elapsed_time": "3:14:04", "remaining_time": "1 day, 4:30:23"}
+{"current_steps": 242, "total_steps": 2365, "loss": 1.3494, "lr": 7.999891025588963e-05, "epoch": 0.5110876451953538, "percentage": 10.23, "elapsed_time": "3:14:52", "remaining_time": "1 day, 4:29:35"}
+{"current_steps": 243, "total_steps": 2365, "loss": 1.3379, "lr": 7.999843077161619e-05, "epoch": 0.5131995776135164, "percentage": 10.27, "elapsed_time": "3:15:40", "remaining_time": "1 day, 4:28:43"}
+{"current_steps": 244, "total_steps": 2365, "loss": 1.3573, "lr": 7.999786411085404e-05, "epoch": 0.515311510031679, "percentage": 10.32, "elapsed_time": "3:16:28", "remaining_time": "1 day, 4:27:53"}
+{"current_steps": 245, "total_steps": 2365, "loss": 1.3424, "lr": 7.999721027483819e-05, "epoch": 0.5174234424498416, "percentage": 10.36, "elapsed_time": "3:17:16", "remaining_time": "1 day, 4:27:01"}
+{"current_steps": 246, "total_steps": 2365, "loss": 1.3403, "lr": 7.999646926499371e-05, "epoch": 0.5195353748680043, "percentage": 10.4, "elapsed_time": "3:18:04", "remaining_time": "1 day, 4:26:12"}
+{"current_steps": 247, "total_steps": 2365, "loss": 1.3414, "lr": 7.999564108293562e-05, "epoch": 0.5216473072861668, "percentage": 10.44, "elapsed_time": "3:18:52", "remaining_time": "1 day, 4:25:22"}
+{"current_steps": 248, "total_steps": 2365, "loss": 1.337, "lr": 7.999472573046892e-05, "epoch": 0.5237592397043295, "percentage": 10.49, "elapsed_time": "3:19:41", "remaining_time": "1 day, 4:24:36"}
+{"current_steps": 249, "total_steps": 2365, "loss": 1.3692, "lr": 7.999372320958863e-05, "epoch": 0.5258711721224921, "percentage": 10.53, "elapsed_time": "3:20:29", "remaining_time": "1 day, 4:23:47"}
+{"current_steps": 250, "total_steps": 2365, "loss": 1.3433, "lr": 7.999263352247976e-05, "epoch": 0.5279831045406547, "percentage": 10.57, "elapsed_time": "3:21:18", "remaining_time": "1 day, 4:23:02"}
+{"current_steps": 251, "total_steps": 2365, "loss": 1.3496, "lr": 7.999145667151725e-05, "epoch": 0.5300950369588173, "percentage": 10.61, "elapsed_time": "3:22:06", "remaining_time": "1 day, 4:22:12"}
+{"current_steps": 252, "total_steps": 2365, "loss": 1.3494, "lr": 7.999019265926606e-05, "epoch": 0.53220696937698, "percentage": 10.66, "elapsed_time": "3:22:54", "remaining_time": "1 day, 4:21:19"}
+{"current_steps": 253, "total_steps": 2365, "loss": 1.3477, "lr": 7.99888414884811e-05, "epoch": 0.5343189017951425, "percentage": 10.7, "elapsed_time": "3:23:42", "remaining_time": "1 day, 4:20:29"}
+{"current_steps": 254, "total_steps": 2365, "loss": 1.3525, "lr": 7.998740316210726e-05, "epoch": 0.5364308342133052, "percentage": 10.74, "elapsed_time": "3:24:30", "remaining_time": "1 day, 4:19:42"}
+{"current_steps": 255, "total_steps": 2365, "loss": 1.3439, "lr": 7.998587768327933e-05, "epoch": 0.5385427666314678, "percentage": 10.78, "elapsed_time": "3:25:18", "remaining_time": "1 day, 4:18:53"}
+{"current_steps": 256, "total_steps": 2365, "loss": 1.3478, "lr": 7.998426505532213e-05, "epoch": 0.5406546990496304, "percentage": 10.82, "elapsed_time": "3:26:07", "remaining_time": "1 day, 4:18:06"}
+{"current_steps": 257, "total_steps": 2365, "loss": 1.3649, "lr": 7.998256528175034e-05, "epoch": 0.542766631467793, "percentage": 10.87, "elapsed_time": "3:26:55", "remaining_time": "1 day, 4:17:14"}
+{"current_steps": 258, "total_steps": 2365, "loss": 1.3627, "lr": 7.998077836626865e-05, "epoch": 0.5448785638859557, "percentage": 10.91, "elapsed_time": "3:27:43", "remaining_time": "1 day, 4:16:23"}
+{"current_steps": 259, "total_steps": 2365, "loss": 1.3525, "lr": 7.997890431277161e-05, "epoch": 0.5469904963041182, "percentage": 10.95, "elapsed_time": "3:28:31", "remaining_time": "1 day, 4:15:35"}
+{"current_steps": 260, "total_steps": 2365, "loss": 1.3254, "lr": 7.997694312534374e-05, "epoch": 0.5491024287222809, "percentage": 10.99, "elapsed_time": "3:29:19", "remaining_time": "1 day, 4:14:43"}
+{"current_steps": 261, "total_steps": 2365, "loss": 1.3291, "lr": 7.997489480825941e-05, "epoch": 0.5512143611404435, "percentage": 11.04, "elapsed_time": "3:30:07", "remaining_time": "1 day, 4:13:52"}
+{"current_steps": 262, "total_steps": 2365, "loss": 1.3357, "lr": 7.997275936598295e-05, "epoch": 0.5533262935586061, "percentage": 11.08, "elapsed_time": "3:30:55", "remaining_time": "1 day, 4:13:00"}
+{"current_steps": 263, "total_steps": 2365, "loss": 1.3313, "lr": 7.997053680316855e-05, "epoch": 0.5554382259767687, "percentage": 11.12, "elapsed_time": "3:31:44", "remaining_time": "1 day, 4:12:15"}
+{"current_steps": 264, "total_steps": 2365, "loss": 1.3433, "lr": 7.996822712466026e-05, "epoch": 0.5575501583949314, "percentage": 11.16, "elapsed_time": "3:32:32", "remaining_time": "1 day, 4:11:25"}
+{"current_steps": 265, "total_steps": 2365, "loss": 1.3246, "lr": 7.996583033549204e-05, "epoch": 0.5596620908130939, "percentage": 11.21, "elapsed_time": "3:33:20", "remaining_time": "1 day, 4:10:37"}
+{"current_steps": 266, "total_steps": 2365, "loss": 1.3533, "lr": 7.996334644088769e-05, "epoch": 0.5617740232312566, "percentage": 11.25, "elapsed_time": "3:34:08", "remaining_time": "1 day, 4:09:49"}
+{"current_steps": 267, "total_steps": 2365, "loss": 1.3249, "lr": 7.996077544626083e-05, "epoch": 0.5638859556494192, "percentage": 11.29, "elapsed_time": "3:34:56", "remaining_time": "1 day, 4:08:59"}
+{"current_steps": 268, "total_steps": 2365, "loss": 1.3124, "lr": 7.995811735721495e-05, "epoch": 0.5659978880675819, "percentage": 11.33, "elapsed_time": "3:35:44", "remaining_time": "1 day, 4:08:08"}
+{"current_steps": 269, "total_steps": 2365, "loss": 1.3293, "lr": 7.995537217954336e-05, "epoch": 0.5681098204857444, "percentage": 11.37, "elapsed_time": "3:36:32", "remaining_time": "1 day, 4:07:17"}
+{"current_steps": 270, "total_steps": 2365, "loss": 1.3378, "lr": 7.995253991922914e-05, "epoch": 0.5702217529039071, "percentage": 11.42, "elapsed_time": "3:37:21", "remaining_time": "1 day, 4:06:30"}
+{"current_steps": 271, "total_steps": 2365, "loss": 1.3405, "lr": 7.994962058244522e-05, "epoch": 0.5723336853220697, "percentage": 11.46, "elapsed_time": "3:38:09", "remaining_time": "1 day, 4:05:40"}
+{"current_steps": 272, "total_steps": 2365, "loss": 1.3321, "lr": 7.994661417555428e-05, "epoch": 0.5744456177402323, "percentage": 11.5, "elapsed_time": "3:38:57", "remaining_time": "1 day, 4:04:52"}
+{"current_steps": 273, "total_steps": 2365, "loss": 1.3307, "lr": 7.994352070510878e-05, "epoch": 0.5765575501583949, "percentage": 11.54, "elapsed_time": "3:39:45", "remaining_time": "1 day, 4:04:03"}
+{"current_steps": 274, "total_steps": 2365, "loss": 1.3307, "lr": 7.994034017785092e-05, "epoch": 0.5786694825765576, "percentage": 11.59, "elapsed_time": "3:40:34", "remaining_time": "1 day, 4:03:15"}
+{"current_steps": 275, "total_steps": 2365, "loss": 1.3162, "lr": 7.993707260071268e-05, "epoch": 0.5807814149947201, "percentage": 11.63, "elapsed_time": "3:41:22", "remaining_time": "1 day, 4:02:25"}
+{"current_steps": 276, "total_steps": 2365, "loss": 1.351, "lr": 7.993371798081569e-05, "epoch": 0.5828933474128828, "percentage": 11.67, "elapsed_time": "3:42:10", "remaining_time": "1 day, 4:01:33"}
+{"current_steps": 277, "total_steps": 2365, "loss": 1.3182, "lr": 7.993027632547138e-05, "epoch": 0.5850052798310454, "percentage": 11.71, "elapsed_time": "3:42:58", "remaining_time": "1 day, 4:00:43"}
+{"current_steps": 278, "total_steps": 2365, "loss": 1.3202, "lr": 7.992674764218081e-05, "epoch": 0.587117212249208, "percentage": 11.75, "elapsed_time": "3:43:46", "remaining_time": "1 day, 3:59:58"}
+{"current_steps": 279, "total_steps": 2365, "loss": 1.3216, "lr": 7.992313193863473e-05, "epoch": 0.5892291446673706, "percentage": 11.8, "elapsed_time": "3:44:35", "remaining_time": "1 day, 3:59:09"}
+{"current_steps": 280, "total_steps": 2365, "loss": 1.332, "lr": 7.991942922271358e-05, "epoch": 0.5913410770855333, "percentage": 11.84, "elapsed_time": "3:45:23", "remaining_time": "1 day, 3:58:19"}
+{"current_steps": 281, "total_steps": 2365, "loss": 1.3274, "lr": 7.99156395024874e-05, "epoch": 0.5934530095036958, "percentage": 11.88, "elapsed_time": "3:46:11", "remaining_time": "1 day, 3:57:32"}
+{"current_steps": 282, "total_steps": 2365, "loss": 1.3332, "lr": 7.991176278621589e-05, "epoch": 0.5955649419218585, "percentage": 11.92, "elapsed_time": "3:46:59", "remaining_time": "1 day, 3:56:41"}
+{"current_steps": 283, "total_steps": 2365, "loss": 1.3217, "lr": 7.990779908234835e-05, "epoch": 0.5976768743400211, "percentage": 11.97, "elapsed_time": "3:47:47", "remaining_time": "1 day, 3:55:49"}
+{"current_steps": 284, "total_steps": 2365, "loss": 1.3423, "lr": 7.990374839952364e-05, "epoch": 0.5997888067581837, "percentage": 12.01, "elapsed_time": "3:48:35", "remaining_time": "1 day, 3:54:57"}
+{"current_steps": 285, "total_steps": 2365, "loss": 1.3391, "lr": 7.989961074657024e-05, "epoch": 0.6019007391763463, "percentage": 12.05, "elapsed_time": "3:49:23", "remaining_time": "1 day, 3:54:06"}
+{"current_steps": 286, "total_steps": 2365, "loss": 1.3138, "lr": 7.989538613250613e-05, "epoch": 0.604012671594509, "percentage": 12.09, "elapsed_time": "3:50:11", "remaining_time": "1 day, 3:53:15"}
+{"current_steps": 287, "total_steps": 2365, "loss": 1.314, "lr": 7.989107456653888e-05, "epoch": 0.6061246040126715, "percentage": 12.14, "elapsed_time": "3:50:59", "remaining_time": "1 day, 3:52:25"}
+{"current_steps": 288, "total_steps": 2365, "loss": 1.3426, "lr": 7.988667605806549e-05, "epoch": 0.6082365364308342, "percentage": 12.18, "elapsed_time": "3:51:47", "remaining_time": "1 day, 3:51:38"}
+{"current_steps": 289, "total_steps": 2365, "loss": 1.3379, "lr": 7.988219061667254e-05, "epoch": 0.6103484688489969, "percentage": 12.22, "elapsed_time": "3:52:36", "remaining_time": "1 day, 3:50:52"}
+{"current_steps": 290, "total_steps": 2365, "loss": 1.3091, "lr": 7.987761825213603e-05, "epoch": 0.6124604012671595, "percentage": 12.26, "elapsed_time": "3:53:24", "remaining_time": "1 day, 3:50:03"}
+{"current_steps": 291, "total_steps": 2365, "loss": 1.3236, "lr": 7.987295897442141e-05, "epoch": 0.614572333685322, "percentage": 12.3, "elapsed_time": "3:54:12", "remaining_time": "1 day, 3:49:13"}
+{"current_steps": 292, "total_steps": 2365, "loss": 1.3098, "lr": 7.986821279368358e-05, "epoch": 0.6166842661034847, "percentage": 12.35, "elapsed_time": "3:55:00", "remaining_time": "1 day, 3:48:23"}
+{"current_steps": 293, "total_steps": 2365, "loss": 1.3426, "lr": 7.98633797202668e-05, "epoch": 0.6187961985216474, "percentage": 12.39, "elapsed_time": "3:55:49", "remaining_time": "1 day, 3:47:38"}
+{"current_steps": 294, "total_steps": 2365, "loss": 1.3214, "lr": 7.985845976470478e-05, "epoch": 0.6209081309398099, "percentage": 12.43, "elapsed_time": "3:56:37", "remaining_time": "1 day, 3:46:51"}
+{"current_steps": 295, "total_steps": 2365, "loss": 1.3179, "lr": 7.985345293772052e-05, "epoch": 0.6230200633579726, "percentage": 12.47, "elapsed_time": "3:57:25", "remaining_time": "1 day, 3:46:03"}
+{"current_steps": 296, "total_steps": 2365, "loss": 1.3394, "lr": 7.984835925022641e-05, "epoch": 0.6251319957761352, "percentage": 12.52, "elapsed_time": "3:58:14", "remaining_time": "1 day, 3:45:16"}
+{"current_steps": 297, "total_steps": 2365, "loss": 1.3384, "lr": 7.984317871332412e-05, "epoch": 0.6272439281942978, "percentage": 12.56, "elapsed_time": "3:59:02", "remaining_time": "1 day, 3:44:28"}
+{"current_steps": 298, "total_steps": 2365, "loss": 1.3166, "lr": 7.983791133830463e-05, "epoch": 0.6293558606124604, "percentage": 12.6, "elapsed_time": "3:59:50", "remaining_time": "1 day, 3:43:38"}
+{"current_steps": 299, "total_steps": 2365, "loss": 1.3518, "lr": 7.983255713664814e-05, "epoch": 0.6314677930306231, "percentage": 12.64, "elapsed_time": "4:00:38", "remaining_time": "1 day, 3:42:47"}
+{"current_steps": 300, "total_steps": 2365, "loss": 1.3381, "lr": 7.982711612002415e-05, "epoch": 0.6335797254487856, "percentage": 12.68, "elapsed_time": "4:01:26", "remaining_time": "1 day, 3:41:56"}
+{"current_steps": 301, "total_steps": 2365, "loss": 1.3292, "lr": 7.982158830029136e-05, "epoch": 0.6356916578669483, "percentage": 12.73, "elapsed_time": "4:02:14", "remaining_time": "1 day, 3:41:08"}
+{"current_steps": 302, "total_steps": 2365, "loss": 1.3314, "lr": 7.98159736894976e-05, "epoch": 0.6378035902851109, "percentage": 12.77, "elapsed_time": "4:03:02", "remaining_time": "1 day, 3:40:18"}
+{"current_steps": 303, "total_steps": 2365, "loss": 1.3097, "lr": 7.98102722998799e-05, "epoch": 0.6399155227032735, "percentage": 12.81, "elapsed_time": "4:03:51", "remaining_time": "1 day, 3:39:29"}
+{"current_steps": 304, "total_steps": 2365, "loss": 1.3373, "lr": 7.980448414386447e-05, "epoch": 0.6420274551214361, "percentage": 12.85, "elapsed_time": "4:04:39", "remaining_time": "1 day, 3:38:42"}
+{"current_steps": 305, "total_steps": 2365, "loss": 1.3389, "lr": 7.979860923406654e-05, "epoch": 0.6441393875395988, "percentage": 12.9, "elapsed_time": "4:05:27", "remaining_time": "1 day, 3:37:52"}
+{"current_steps": 306, "total_steps": 2365, "loss": 1.3531, "lr": 7.979264758329048e-05, "epoch": 0.6462513199577613, "percentage": 12.94, "elapsed_time": "4:06:15", "remaining_time": "1 day, 3:37:03"}
+{"current_steps": 307, "total_steps": 2365, "loss": 1.3381, "lr": 7.978659920452972e-05, "epoch": 0.648363252375924, "percentage": 12.98, "elapsed_time": "4:07:04", "remaining_time": "1 day, 3:36:16"}
+{"current_steps": 308, "total_steps": 2365, "loss": 1.3361, "lr": 7.978046411096663e-05, "epoch": 0.6504751847940866, "percentage": 13.02, "elapsed_time": "4:07:52", "remaining_time": "1 day, 3:35:28"}
+{"current_steps": 309, "total_steps": 2365, "loss": 1.3291, "lr": 7.977424231597268e-05, "epoch": 0.6525871172122492, "percentage": 13.07, "elapsed_time": "4:08:41", "remaining_time": "1 day, 3:34:42"}
+{"current_steps": 310, "total_steps": 2365, "loss": 1.3215, "lr": 7.976793383310822e-05, "epoch": 0.6546990496304118, "percentage": 13.11, "elapsed_time": "4:09:30", "remaining_time": "1 day, 3:33:56"}
+{"current_steps": 311, "total_steps": 2365, "loss": 1.3313, "lr": 7.976153867612261e-05, "epoch": 0.6568109820485745, "percentage": 13.15, "elapsed_time": "4:10:18", "remaining_time": "1 day, 3:33:08"}
+{"current_steps": 312, "total_steps": 2365, "loss": 1.3353, "lr": 7.97550568589541e-05, "epoch": 0.6589229144667371, "percentage": 13.19, "elapsed_time": "4:11:06", "remaining_time": "1 day, 3:32:19"}
+{"current_steps": 313, "total_steps": 2365, "loss": 1.3416, "lr": 7.974848839572971e-05, "epoch": 0.6610348468848997, "percentage": 13.23, "elapsed_time": "4:11:54", "remaining_time": "1 day, 3:31:28"}
+{"current_steps": 314, "total_steps": 2365, "loss": 1.3464, "lr": 7.974183330076546e-05, "epoch": 0.6631467793030623, "percentage": 13.28, "elapsed_time": "4:12:42", "remaining_time": "1 day, 3:30:39"}
+{"current_steps": 315, "total_steps": 2365, "loss": 1.3338, "lr": 7.97350915885661e-05, "epoch": 0.665258711721225, "percentage": 13.32, "elapsed_time": "4:13:30", "remaining_time": "1 day, 3:29:51"}
+{"current_steps": 316, "total_steps": 2365, "loss": 1.306, "lr": 7.972826327382516e-05, "epoch": 0.6673706441393875, "percentage": 13.36, "elapsed_time": "4:14:19", "remaining_time": "1 day, 3:29:05"}
+{"current_steps": 317, "total_steps": 2365, "loss": 1.3257, "lr": 7.972134837142497e-05, "epoch": 0.6694825765575502, "percentage": 13.4, "elapsed_time": "4:15:07", "remaining_time": "1 day, 3:28:17"}
+{"current_steps": 318, "total_steps": 2365, "loss": 1.3294, "lr": 7.971434689643652e-05, "epoch": 0.6715945089757128, "percentage": 13.45, "elapsed_time": "4:15:56", "remaining_time": "1 day, 3:27:31"}
+{"current_steps": 319, "total_steps": 2365, "loss": 1.3337, "lr": 7.97072588641195e-05, "epoch": 0.6737064413938754, "percentage": 13.49, "elapsed_time": "4:16:44", "remaining_time": "1 day, 3:26:42"}
+{"current_steps": 320, "total_steps": 2365, "loss": 1.3382, "lr": 7.970008428992228e-05, "epoch": 0.675818373812038, "percentage": 13.53, "elapsed_time": "4:17:32", "remaining_time": "1 day, 3:25:51"}
+{"current_steps": 321, "total_steps": 2365, "loss": 1.3189, "lr": 7.96928231894818e-05, "epoch": 0.6779303062302007, "percentage": 13.57, "elapsed_time": "4:18:20", "remaining_time": "1 day, 3:24:59"}
+{"current_steps": 322, "total_steps": 2365, "loss": 1.337, "lr": 7.968547557862366e-05, "epoch": 0.6800422386483632, "percentage": 13.62, "elapsed_time": "4:19:08", "remaining_time": "1 day, 3:24:09"}
+{"current_steps": 323, "total_steps": 2365, "loss": 1.3251, "lr": 7.967804147336191e-05, "epoch": 0.6821541710665259, "percentage": 13.66, "elapsed_time": "4:19:56", "remaining_time": "1 day, 3:23:20"}
+{"current_steps": 324, "total_steps": 2365, "loss": 1.3157, "lr": 7.967052088989921e-05, "epoch": 0.6842661034846885, "percentage": 13.7, "elapsed_time": "4:20:44", "remaining_time": "1 day, 3:22:32"}
+{"current_steps": 325, "total_steps": 2365, "loss": 1.315, "lr": 7.966291384462664e-05, "epoch": 0.6863780359028511, "percentage": 13.74, "elapsed_time": "4:21:33", "remaining_time": "1 day, 3:21:46"}
+{"current_steps": 326, "total_steps": 2365, "loss": 1.3234, "lr": 7.965522035412371e-05, "epoch": 0.6884899683210137, "percentage": 13.78, "elapsed_time": "4:22:21", "remaining_time": "1 day, 3:20:59"}
+{"current_steps": 327, "total_steps": 2365, "loss": 1.3292, "lr": 7.964744043515839e-05, "epoch": 0.6906019007391764, "percentage": 13.83, "elapsed_time": "4:23:10", "remaining_time": "1 day, 3:20:10"}
+{"current_steps": 328, "total_steps": 2365, "loss": 1.3169, "lr": 7.963957410468698e-05, "epoch": 0.6927138331573389, "percentage": 13.87, "elapsed_time": "4:23:58", "remaining_time": "1 day, 3:19:20"}
+{"current_steps": 329, "total_steps": 2365, "loss": 1.3194, "lr": 7.963162137985416e-05, "epoch": 0.6948257655755016, "percentage": 13.91, "elapsed_time": "4:24:46", "remaining_time": "1 day, 3:18:30"}
+{"current_steps": 330, "total_steps": 2365, "loss": 1.3069, "lr": 7.962358227799286e-05, "epoch": 0.6969376979936642, "percentage": 13.95, "elapsed_time": "4:25:34", "remaining_time": "1 day, 3:17:42"}
+{"current_steps": 331, "total_steps": 2365, "loss": 1.3195, "lr": 7.961545681662427e-05, "epoch": 0.6990496304118268, "percentage": 14.0, "elapsed_time": "4:26:22", "remaining_time": "1 day, 3:16:55"}
+{"current_steps": 332, "total_steps": 2365, "loss": 1.3289, "lr": 7.960724501345783e-05, "epoch": 0.7011615628299894, "percentage": 14.04, "elapsed_time": "4:27:11", "remaining_time": "1 day, 3:16:08"}
+{"current_steps": 333, "total_steps": 2365, "loss": 1.3184, "lr": 7.959894688639115e-05, "epoch": 0.7032734952481521, "percentage": 14.08, "elapsed_time": "4:27:59", "remaining_time": "1 day, 3:15:21"}
+{"current_steps": 334, "total_steps": 2365, "loss": 1.3273, "lr": 7.959056245350996e-05, "epoch": 0.7053854276663146, "percentage": 14.12, "elapsed_time": "4:28:47", "remaining_time": "1 day, 3:14:30"}
+{"current_steps": 335, "total_steps": 2365, "loss": 1.3203, "lr": 7.958209173308811e-05, "epoch": 0.7074973600844773, "percentage": 14.16, "elapsed_time": "4:29:36", "remaining_time": "1 day, 3:13:43"}
+{"current_steps": 336, "total_steps": 2365, "loss": 1.3155, "lr": 7.957353474358755e-05, "epoch": 0.7096092925026399, "percentage": 14.21, "elapsed_time": "4:30:24", "remaining_time": "1 day, 3:12:54"}
+{"current_steps": 337, "total_steps": 2365, "loss": 1.3259, "lr": 7.95648915036582e-05, "epoch": 0.7117212249208026, "percentage": 14.25, "elapsed_time": "4:31:12", "remaining_time": "1 day, 3:12:05"}
+{"current_steps": 338, "total_steps": 2365, "loss": 1.3024, "lr": 7.955616203213797e-05, "epoch": 0.7138331573389651, "percentage": 14.29, "elapsed_time": "4:32:00", "remaining_time": "1 day, 3:11:17"}
+{"current_steps": 339, "total_steps": 2365, "loss": 1.3048, "lr": 7.954734634805274e-05, "epoch": 0.7159450897571278, "percentage": 14.33, "elapsed_time": "4:32:49", "remaining_time": "1 day, 3:10:30"}
+{"current_steps": 340, "total_steps": 2365, "loss": 1.3098, "lr": 7.953844447061626e-05, "epoch": 0.7180570221752904, "percentage": 14.38, "elapsed_time": "4:33:38", "remaining_time": "1 day, 3:09:44"}
+{"current_steps": 341, "total_steps": 2365, "loss": 1.3117, "lr": 7.952945641923015e-05, "epoch": 0.720168954593453, "percentage": 14.42, "elapsed_time": "4:34:26", "remaining_time": "1 day, 3:08:54"}
+{"current_steps": 342, "total_steps": 2365, "loss": 1.3225, "lr": 7.952038221348387e-05, "epoch": 0.7222808870116156, "percentage": 14.46, "elapsed_time": "4:35:14", "remaining_time": "1 day, 3:08:08"}
+{"current_steps": 343, "total_steps": 2365, "loss": 1.317, "lr": 7.951122187315462e-05, "epoch": 0.7243928194297783, "percentage": 14.5, "elapsed_time": "4:36:02", "remaining_time": "1 day, 3:07:18"}
+{"current_steps": 344, "total_steps": 2365, "loss": 1.3085, "lr": 7.950197541820732e-05, "epoch": 0.7265047518479408, "percentage": 14.55, "elapsed_time": "4:36:51", "remaining_time": "1 day, 3:06:30"}
+{"current_steps": 345, "total_steps": 2365, "loss": 1.2978, "lr": 7.949264286879462e-05, "epoch": 0.7286166842661035, "percentage": 14.59, "elapsed_time": "4:37:39", "remaining_time": "1 day, 3:05:42"}
+{"current_steps": 346, "total_steps": 2365, "loss": 1.3184, "lr": 7.948322424525679e-05, "epoch": 0.7307286166842661, "percentage": 14.63, "elapsed_time": "4:38:27", "remaining_time": "1 day, 3:04:52"}
+{"current_steps": 347, "total_steps": 2365, "loss": 1.3329, "lr": 7.94737195681217e-05, "epoch": 0.7328405491024287, "percentage": 14.67, "elapsed_time": "4:39:15", "remaining_time": "1 day, 3:04:03"}
+{"current_steps": 348, "total_steps": 2365, "loss": 1.3007, "lr": 7.946412885810475e-05, "epoch": 0.7349524815205913, "percentage": 14.71, "elapsed_time": "4:40:03", "remaining_time": "1 day, 3:03:13"}
+{"current_steps": 349, "total_steps": 2365, "loss": 1.3077, "lr": 7.94544521361089e-05, "epoch": 0.737064413938754, "percentage": 14.76, "elapsed_time": "4:40:51", "remaining_time": "1 day, 3:02:22"}
+{"current_steps": 350, "total_steps": 2365, "loss": 1.3114, "lr": 7.944468942322453e-05, "epoch": 0.7391763463569165, "percentage": 14.8, "elapsed_time": "4:41:39", "remaining_time": "1 day, 3:01:31"}
+{"current_steps": 351, "total_steps": 2365, "loss": 1.3095, "lr": 7.943484074072943e-05, "epoch": 0.7412882787750792, "percentage": 14.84, "elapsed_time": "4:42:27", "remaining_time": "1 day, 3:00:41"}
+{"current_steps": 352, "total_steps": 2365, "loss": 1.3282, "lr": 7.942490611008882e-05, "epoch": 0.7434002111932418, "percentage": 14.88, "elapsed_time": "4:43:15", "remaining_time": "1 day, 2:59:51"}
+{"current_steps": 353, "total_steps": 2365, "loss": 1.3184, "lr": 7.94148855529552e-05, "epoch": 0.7455121436114044, "percentage": 14.93, "elapsed_time": "4:44:03", "remaining_time": "1 day, 2:59:02"}
+{"current_steps": 354, "total_steps": 2365, "loss": 1.3194, "lr": 7.94047790911683e-05, "epoch": 0.747624076029567, "percentage": 14.97, "elapsed_time": "4:44:51", "remaining_time": "1 day, 2:58:12"}
+{"current_steps": 355, "total_steps": 2365, "loss": 1.3061, "lr": 7.939458674675519e-05, "epoch": 0.7497360084477297, "percentage": 15.01, "elapsed_time": "4:45:39", "remaining_time": "1 day, 2:57:22"}
+{"current_steps": 356, "total_steps": 2365, "loss": 1.3097, "lr": 7.938430854193005e-05, "epoch": 0.7518479408658922, "percentage": 15.05, "elapsed_time": "4:46:27", "remaining_time": "1 day, 2:56:33"}
+{"current_steps": 357, "total_steps": 2365, "loss": 1.313, "lr": 7.937394449909418e-05, "epoch": 0.7539598732840549, "percentage": 15.1, "elapsed_time": "4:47:15", "remaining_time": "1 day, 2:55:44"}
+{"current_steps": 358, "total_steps": 2365, "loss": 1.3051, "lr": 7.9363494640836e-05, "epoch": 0.7560718057022175, "percentage": 15.14, "elapsed_time": "4:48:03", "remaining_time": "1 day, 2:54:56"}
+{"current_steps": 359, "total_steps": 2365, "loss": 1.3374, "lr": 7.935295898993096e-05, "epoch": 0.7581837381203802, "percentage": 15.18, "elapsed_time": "4:48:52", "remaining_time": "1 day, 2:54:07"}
+{"current_steps": 360, "total_steps": 2365, "loss": 1.3323, "lr": 7.934233756934147e-05, "epoch": 0.7602956705385427, "percentage": 15.22, "elapsed_time": "4:49:39", "remaining_time": "1 day, 2:53:16"}
+{"current_steps": 361, "total_steps": 2365, "loss": 1.3261, "lr": 7.933163040221692e-05, "epoch": 0.7624076029567054, "percentage": 15.26, "elapsed_time": "4:50:28", "remaining_time": "1 day, 2:52:27"}
+{"current_steps": 362, "total_steps": 2365, "loss": 1.3096, "lr": 7.932083751189352e-05, "epoch": 0.764519535374868, "percentage": 15.31, "elapsed_time": "4:51:16", "remaining_time": "1 day, 2:51:40"}
+{"current_steps": 363, "total_steps": 2365, "loss": 1.31, "lr": 7.930995892189439e-05, "epoch": 0.7666314677930306, "percentage": 15.35, "elapsed_time": "4:52:04", "remaining_time": "1 day, 2:50:50"}
+{"current_steps": 364, "total_steps": 2365, "loss": 1.3488, "lr": 7.929899465592934e-05, "epoch": 0.7687434002111933, "percentage": 15.39, "elapsed_time": "4:52:52", "remaining_time": "1 day, 2:50:01"}
+{"current_steps": 365, "total_steps": 2365, "loss": 1.3248, "lr": 7.928794473789503e-05, "epoch": 0.7708553326293559, "percentage": 15.43, "elapsed_time": "4:53:40", "remaining_time": "1 day, 2:49:11"}
+{"current_steps": 366, "total_steps": 2365, "loss": 1.292, "lr": 7.927680919187467e-05, "epoch": 0.7729672650475184, "percentage": 15.48, "elapsed_time": "4:54:28", "remaining_time": "1 day, 2:48:21"}
+{"current_steps": 367, "total_steps": 2365, "loss": 1.3229, "lr": 7.926558804213818e-05, "epoch": 0.7750791974656811, "percentage": 15.52, "elapsed_time": "4:55:16", "remaining_time": "1 day, 2:47:32"}
+{"current_steps": 368, "total_steps": 2365, "loss": 1.3659, "lr": 7.925428131314205e-05, "epoch": 0.7771911298838438, "percentage": 15.56, "elapsed_time": "4:56:04", "remaining_time": "1 day, 2:46:41"}
+{"current_steps": 369, "total_steps": 2365, "loss": 1.3196, "lr": 7.924288902952925e-05, "epoch": 0.7793030623020063, "percentage": 15.6, "elapsed_time": "4:56:52", "remaining_time": "1 day, 2:45:49"}
+{"current_steps": 370, "total_steps": 2365, "loss": 1.3334, "lr": 7.923141121612922e-05, "epoch": 0.781414994720169, "percentage": 15.64, "elapsed_time": "4:57:39", "remaining_time": "1 day, 2:44:59"}
+{"current_steps": 371, "total_steps": 2365, "loss": 1.3504, "lr": 7.921984789795787e-05, "epoch": 0.7835269271383316, "percentage": 15.69, "elapsed_time": "4:58:27", "remaining_time": "1 day, 2:44:07"}
+{"current_steps": 372, "total_steps": 2365, "loss": 1.3354, "lr": 7.920819910021738e-05, "epoch": 0.7856388595564942, "percentage": 15.73, "elapsed_time": "4:59:15", "remaining_time": "1 day, 2:43:16"}
+{"current_steps": 373, "total_steps": 2365, "loss": 1.3293, "lr": 7.919646484829631e-05, "epoch": 0.7877507919746568, "percentage": 15.77, "elapsed_time": "5:00:03", "remaining_time": "1 day, 2:42:27"}
+{"current_steps": 374, "total_steps": 2365, "loss": 1.3217, "lr": 7.918464516776943e-05, "epoch": 0.7898627243928195, "percentage": 15.81, "elapsed_time": "5:00:51", "remaining_time": "1 day, 2:41:39"}
+{"current_steps": 375, "total_steps": 2365, "loss": 1.323, "lr": 7.91727400843977e-05, "epoch": 0.791974656810982, "percentage": 15.86, "elapsed_time": "5:01:40", "remaining_time": "1 day, 2:40:51"}
+{"current_steps": 376, "total_steps": 2365, "loss": 1.3244, "lr": 7.916074962412823e-05, "epoch": 0.7940865892291447, "percentage": 15.9, "elapsed_time": "5:02:28", "remaining_time": "1 day, 2:40:01"}
+{"current_steps": 377, "total_steps": 2365, "loss": 1.3204, "lr": 7.914867381309418e-05, "epoch": 0.7961985216473073, "percentage": 15.94, "elapsed_time": "5:03:16", "remaining_time": "1 day, 2:39:13"}
+{"current_steps": 378, "total_steps": 2365, "loss": 1.3171, "lr": 7.913651267761478e-05, "epoch": 0.7983104540654699, "percentage": 15.98, "elapsed_time": "5:04:04", "remaining_time": "1 day, 2:38:24"}
+{"current_steps": 379, "total_steps": 2365, "loss": 1.3131, "lr": 7.912426624419518e-05, "epoch": 0.8004223864836325, "percentage": 16.03, "elapsed_time": "5:04:52", "remaining_time": "1 day, 2:37:33"}
+{"current_steps": 380, "total_steps": 2365, "loss": 1.3231, "lr": 7.911193453952646e-05, "epoch": 0.8025343189017952, "percentage": 16.07, "elapsed_time": "5:05:40", "remaining_time": "1 day, 2:36:43"}
+{"current_steps": 381, "total_steps": 2365, "loss": 1.3118, "lr": 7.909951759048555e-05, "epoch": 0.8046462513199577, "percentage": 16.11, "elapsed_time": "5:06:28", "remaining_time": "1 day, 2:35:53"}
+{"current_steps": 382, "total_steps": 2365, "loss": 1.3049, "lr": 7.908701542413515e-05, "epoch": 0.8067581837381204, "percentage": 16.15, "elapsed_time": "5:07:16", "remaining_time": "1 day, 2:35:04"}
+{"current_steps": 383, "total_steps": 2365, "loss": 1.3331, "lr": 7.90744280677237e-05, "epoch": 0.808870116156283, "percentage": 16.19, "elapsed_time": "5:08:04", "remaining_time": "1 day, 2:34:15"}
+{"current_steps": 384, "total_steps": 2365, "loss": 1.3168, "lr": 7.906175554868531e-05, "epoch": 0.8109820485744457, "percentage": 16.24, "elapsed_time": "5:08:52", "remaining_time": "1 day, 2:33:26"}
+{"current_steps": 385, "total_steps": 2365, "loss": 1.3024, "lr": 7.904899789463974e-05, "epoch": 0.8130939809926082, "percentage": 16.28, "elapsed_time": "5:09:40", "remaining_time": "1 day, 2:32:37"}
+{"current_steps": 386, "total_steps": 2365, "loss": 1.2967, "lr": 7.903615513339224e-05, "epoch": 0.8152059134107709, "percentage": 16.32, "elapsed_time": "5:10:28", "remaining_time": "1 day, 2:31:47"}
+{"current_steps": 387, "total_steps": 2365, "loss": 1.3132, "lr": 7.902322729293357e-05, "epoch": 0.8173178458289335, "percentage": 16.36, "elapsed_time": "5:11:17", "remaining_time": "1 day, 2:31:00"}
+{"current_steps": 388, "total_steps": 2365, "loss": 1.3053, "lr": 7.901021440143995e-05, "epoch": 0.8194297782470961, "percentage": 16.41, "elapsed_time": "5:12:05", "remaining_time": "1 day, 2:30:12"}
+{"current_steps": 389, "total_steps": 2365, "loss": 1.3412, "lr": 7.899711648727294e-05, "epoch": 0.8215417106652587, "percentage": 16.45, "elapsed_time": "5:12:53", "remaining_time": "1 day, 2:29:22"}
+{"current_steps": 390, "total_steps": 2365, "loss": 1.326, "lr": 7.898393357897942e-05, "epoch": 0.8236536430834214, "percentage": 16.49, "elapsed_time": "5:13:41", "remaining_time": "1 day, 2:28:33"}
+{"current_steps": 391, "total_steps": 2365, "loss": 1.3304, "lr": 7.897066570529151e-05, "epoch": 0.8257655755015839, "percentage": 16.53, "elapsed_time": "5:14:29", "remaining_time": "1 day, 2:27:46"}
+{"current_steps": 392, "total_steps": 2365, "loss": 1.3154, "lr": 7.895731289512649e-05, "epoch": 0.8278775079197466, "percentage": 16.58, "elapsed_time": "5:15:18", "remaining_time": "1 day, 2:26:58"}
+{"current_steps": 393, "total_steps": 2365, "loss": 1.3212, "lr": 7.894387517758679e-05, "epoch": 0.8299894403379092, "percentage": 16.62, "elapsed_time": "5:16:06", "remaining_time": "1 day, 2:26:09"}
+{"current_steps": 394, "total_steps": 2365, "loss": 1.3383, "lr": 7.89303525819599e-05, "epoch": 0.8321013727560718, "percentage": 16.66, "elapsed_time": "5:16:54", "remaining_time": "1 day, 2:25:20"}
+{"current_steps": 395, "total_steps": 2365, "loss": 1.3182, "lr": 7.891674513771827e-05, "epoch": 0.8342133051742344, "percentage": 16.7, "elapsed_time": "5:17:42", "remaining_time": "1 day, 2:24:31"}
+{"current_steps": 396, "total_steps": 2365, "loss": 1.3375, "lr": 7.89030528745193e-05, "epoch": 0.8363252375923971, "percentage": 16.74, "elapsed_time": "5:18:30", "remaining_time": "1 day, 2:23:41"}
+{"current_steps": 397, "total_steps": 2365, "loss": 1.3168, "lr": 7.888927582220523e-05, "epoch": 0.8384371700105596, "percentage": 16.79, "elapsed_time": "5:19:18", "remaining_time": "1 day, 2:22:50"}
+{"current_steps": 398, "total_steps": 2365, "loss": 1.3188, "lr": 7.887541401080313e-05, "epoch": 0.8405491024287223, "percentage": 16.83, "elapsed_time": "5:20:06", "remaining_time": "1 day, 2:22:00"}
+{"current_steps": 399, "total_steps": 2365, "loss": 1.3104, "lr": 7.886146747052479e-05, "epoch": 0.8426610348468849, "percentage": 16.87, "elapsed_time": "5:20:54", "remaining_time": "1 day, 2:21:12"}
+{"current_steps": 400, "total_steps": 2365, "loss": 1.3749, "lr": 7.884743623176666e-05, "epoch": 0.8447729672650475, "percentage": 16.91, "elapsed_time": "5:21:42", "remaining_time": "1 day, 2:20:22"}
+{"current_steps": 401, "total_steps": 2365, "loss": 1.3366, "lr": 7.883332032510979e-05, "epoch": 0.8468848996832101, "percentage": 16.96, "elapsed_time": "5:22:30", "remaining_time": "1 day, 2:19:34"}
+{"current_steps": 402, "total_steps": 2365, "loss": 1.3326, "lr": 7.881911978131977e-05, "epoch": 0.8489968321013728, "percentage": 17.0, "elapsed_time": "5:23:19", "remaining_time": "1 day, 2:18:47"}
+{"current_steps": 403, "total_steps": 2365, "loss": 1.3464, "lr": 7.880483463134663e-05, "epoch": 0.8511087645195353, "percentage": 17.04, "elapsed_time": "5:24:07", "remaining_time": "1 day, 2:17:58"}
+{"current_steps": 404, "total_steps": 2365, "loss": 1.3424, "lr": 7.879046490632487e-05, "epoch": 0.853220696937698, "percentage": 17.08, "elapsed_time": "5:24:55", "remaining_time": "1 day, 2:17:11"}
+{"current_steps": 405, "total_steps": 2365, "loss": 1.3489, "lr": 7.877601063757323e-05, "epoch": 0.8553326293558606, "percentage": 17.12, "elapsed_time": "5:25:43", "remaining_time": "1 day, 2:16:22"}
+{"current_steps": 406, "total_steps": 2365, "loss": 1.3425, "lr": 7.876147185659476e-05, "epoch": 0.8574445617740233, "percentage": 17.17, "elapsed_time": "5:26:31", "remaining_time": "1 day, 2:15:32"}
+{"current_steps": 407, "total_steps": 2365, "loss": 1.3143, "lr": 7.874684859507672e-05, "epoch": 0.8595564941921858, "percentage": 17.21, "elapsed_time": "5:27:20", "remaining_time": "1 day, 2:14:48"}
+{"current_steps": 408, "total_steps": 2365, "loss": 1.3329, "lr": 7.873214088489047e-05, "epoch": 0.8616684266103485, "percentage": 17.25, "elapsed_time": "5:28:09", "remaining_time": "1 day, 2:14:00"}
+{"current_steps": 409, "total_steps": 2365, "loss": 1.3106, "lr": 7.871734875809142e-05, "epoch": 0.8637803590285111, "percentage": 17.29, "elapsed_time": "5:28:57", "remaining_time": "1 day, 2:13:12"}
+{"current_steps": 410, "total_steps": 2365, "loss": 1.3267, "lr": 7.870247224691897e-05, "epoch": 0.8658922914466737, "percentage": 17.34, "elapsed_time": "5:29:45", "remaining_time": "1 day, 2:12:24"}
+{"current_steps": 411, "total_steps": 2365, "loss": 1.318, "lr": 7.868751138379647e-05, "epoch": 0.8680042238648363, "percentage": 17.38, "elapsed_time": "5:30:33", "remaining_time": "1 day, 2:11:35"}
+{"current_steps": 412, "total_steps": 2365, "loss": 1.3035, "lr": 7.867246620133107e-05, "epoch": 0.870116156282999, "percentage": 17.42, "elapsed_time": "5:31:21", "remaining_time": "1 day, 2:10:46"}
+{"current_steps": 413, "total_steps": 2365, "loss": 1.3175, "lr": 7.865733673231372e-05, "epoch": 0.8722280887011615, "percentage": 17.46, "elapsed_time": "5:32:09", "remaining_time": "1 day, 2:09:56"}
+{"current_steps": 414, "total_steps": 2365, "loss": 1.3189, "lr": 7.864212300971904e-05, "epoch": 0.8743400211193242, "percentage": 17.51, "elapsed_time": "5:32:58", "remaining_time": "1 day, 2:09:07"}
+{"current_steps": 415, "total_steps": 2365, "loss": 1.3131, "lr": 7.862682506670532e-05, "epoch": 0.8764519535374868, "percentage": 17.55, "elapsed_time": "5:33:46", "remaining_time": "1 day, 2:08:20"}
+{"current_steps": 416, "total_steps": 2365, "loss": 1.3111, "lr": 7.86114429366144e-05, "epoch": 0.8785638859556494, "percentage": 17.59, "elapsed_time": "5:34:34", "remaining_time": "1 day, 2:07:31"}
+{"current_steps": 417, "total_steps": 2365, "loss": 1.3254, "lr": 7.859597665297159e-05, "epoch": 0.880675818373812, "percentage": 17.63, "elapsed_time": "5:35:22", "remaining_time": "1 day, 2:06:43"}
+{"current_steps": 418, "total_steps": 2365, "loss": 1.3127, "lr": 7.858042624948563e-05, "epoch": 0.8827877507919747, "percentage": 17.67, "elapsed_time": "5:36:10", "remaining_time": "1 day, 2:05:53"}
+{"current_steps": 419, "total_steps": 2365, "loss": 1.3205, "lr": 7.856479176004854e-05, "epoch": 0.8848996832101372, "percentage": 17.72, "elapsed_time": "5:36:58", "remaining_time": "1 day, 2:05:04"}
+{"current_steps": 420, "total_steps": 2365, "loss": 6.688, "lr": 7.854907321873573e-05, "epoch": 0.8870116156282999, "percentage": 17.76, "elapsed_time": "5:37:46", "remaining_time": "1 day, 2:04:13"}
+{"current_steps": 421, "total_steps": 2365, "loss": 1.4205, "lr": 7.853327065980567e-05, "epoch": 0.8891235480464625, "percentage": 17.8, "elapsed_time": "5:38:34", "remaining_time": "1 day, 2:03:22"}
+{"current_steps": 422, "total_steps": 2365, "loss": 1.3607, "lr": 7.851738411770003e-05, "epoch": 0.8912354804646251, "percentage": 17.84, "elapsed_time": "5:39:21", "remaining_time": "1 day, 2:02:31"}
+{"current_steps": 423, "total_steps": 2365, "loss": 1.3449, "lr": 7.850141362704346e-05, "epoch": 0.8933474128827877, "percentage": 17.89, "elapsed_time": "5:40:09", "remaining_time": "1 day, 2:01:41"}
+{"current_steps": 424, "total_steps": 2365, "loss": 1.333, "lr": 7.848535922264365e-05, "epoch": 0.8954593453009504, "percentage": 17.93, "elapsed_time": "5:40:57", "remaining_time": "1 day, 2:00:51"}
+{"current_steps": 425, "total_steps": 2365, "loss": 1.3234, "lr": 7.846922093949109e-05, "epoch": 0.8975712777191129, "percentage": 17.97, "elapsed_time": "5:41:45", "remaining_time": "1 day, 2:00:03"}
+{"current_steps": 426, "total_steps": 2365, "loss": 1.3402, "lr": 7.84529988127592e-05, "epoch": 0.8996832101372756, "percentage": 18.01, "elapsed_time": "5:42:33", "remaining_time": "1 day, 1:59:13"}
+{"current_steps": 427, "total_steps": 2365, "loss": 1.3293, "lr": 7.843669287780399e-05, "epoch": 0.9017951425554382, "percentage": 18.05, "elapsed_time": "5:43:21", "remaining_time": "1 day, 1:58:24"}
+{"current_steps": 428, "total_steps": 2365, "loss": 1.3331, "lr": 7.842030317016425e-05, "epoch": 0.9039070749736009, "percentage": 18.1, "elapsed_time": "5:44:10", "remaining_time": "1 day, 1:57:36"}
+{"current_steps": 429, "total_steps": 2365, "loss": 1.3195, "lr": 7.840382972556132e-05, "epoch": 0.9060190073917634, "percentage": 18.14, "elapsed_time": "5:44:58", "remaining_time": "1 day, 1:56:48"}
+{"current_steps": 430, "total_steps": 2365, "loss": 1.3097, "lr": 7.8387272579899e-05, "epoch": 0.9081309398099261, "percentage": 18.18, "elapsed_time": "5:45:46", "remaining_time": "1 day, 1:56:01"}
+{"current_steps": 431, "total_steps": 2365, "loss": 1.3346, "lr": 7.837063176926357e-05, "epoch": 0.9102428722280888, "percentage": 18.22, "elapsed_time": "5:46:35", "remaining_time": "1 day, 1:55:12"}
+{"current_steps": 432, "total_steps": 2365, "loss": 1.3571, "lr": 7.835390732992366e-05, "epoch": 0.9123548046462513, "percentage": 18.27, "elapsed_time": "5:47:23", "remaining_time": "1 day, 1:54:24"}
+{"current_steps": 433, "total_steps": 2365, "loss": 1.3418, "lr": 7.833709929833012e-05, "epoch": 0.914466737064414, "percentage": 18.31, "elapsed_time": "5:48:11", "remaining_time": "1 day, 1:53:37"}
+{"current_steps": 434, "total_steps": 2365, "loss": 1.3232, "lr": 7.832020771111602e-05, "epoch": 0.9165786694825766, "percentage": 18.35, "elapsed_time": "5:49:00", "remaining_time": "1 day, 1:52:49"}
+{"current_steps": 435, "total_steps": 2365, "loss": 1.3181, "lr": 7.830323260509653e-05, "epoch": 0.9186906019007391, "percentage": 18.39, "elapsed_time": "5:49:48", "remaining_time": "1 day, 1:52:02"}
+{"current_steps": 436, "total_steps": 2365, "loss": 1.3228, "lr": 7.828617401726887e-05, "epoch": 0.9208025343189018, "percentage": 18.44, "elapsed_time": "5:50:37", "remaining_time": "1 day, 1:51:16"}
+{"current_steps": 437, "total_steps": 2365, "loss": 1.3325, "lr": 7.82690319848122e-05, "epoch": 0.9229144667370645, "percentage": 18.48, "elapsed_time": "5:51:25", "remaining_time": "1 day, 1:50:26"}
+{"current_steps": 438, "total_steps": 2365, "loss": 1.3243, "lr": 7.825180654508752e-05, "epoch": 0.925026399155227, "percentage": 18.52, "elapsed_time": "5:52:13", "remaining_time": "1 day, 1:49:36"}
+{"current_steps": 439, "total_steps": 2365, "loss": 1.323, "lr": 7.823449773563763e-05, "epoch": 0.9271383315733897, "percentage": 18.56, "elapsed_time": "5:53:01", "remaining_time": "1 day, 1:48:48"}
+{"current_steps": 440, "total_steps": 2365, "loss": 1.3085, "lr": 7.821710559418707e-05, "epoch": 0.9292502639915523, "percentage": 18.6, "elapsed_time": "5:53:49", "remaining_time": "1 day, 1:47:59"}
+{"current_steps": 441, "total_steps": 2365, "loss": 1.3429, "lr": 7.819963015864196e-05, "epoch": 0.9313621964097148, "percentage": 18.65, "elapsed_time": "5:54:38", "remaining_time": "1 day, 1:47:12"}
+{"current_steps": 442, "total_steps": 2365, "loss": 1.3172, "lr": 7.818207146708997e-05, "epoch": 0.9334741288278775, "percentage": 18.69, "elapsed_time": "5:55:26", "remaining_time": "1 day, 1:46:22"}
+{"current_steps": 443, "total_steps": 2365, "loss": 1.3142, "lr": 7.816442955780024e-05, "epoch": 0.9355860612460402, "percentage": 18.73, "elapsed_time": "5:56:13", "remaining_time": "1 day, 1:45:32"}
+{"current_steps": 444, "total_steps": 2365, "loss": 1.3052, "lr": 7.814670446922324e-05, "epoch": 0.9376979936642027, "percentage": 18.77, "elapsed_time": "5:57:01", "remaining_time": "1 day, 1:44:43"}
+{"current_steps": 445, "total_steps": 2365, "loss": 1.3067, "lr": 7.812889623999077e-05, "epoch": 0.9398099260823654, "percentage": 18.82, "elapsed_time": "5:57:49", "remaining_time": "1 day, 1:43:52"}
+{"current_steps": 446, "total_steps": 2365, "loss": 1.3261, "lr": 7.811100490891586e-05, "epoch": 0.941921858500528, "percentage": 18.86, "elapsed_time": "5:58:37", "remaining_time": "1 day, 1:43:03"}
+{"current_steps": 447, "total_steps": 2365, "loss": 1.3187, "lr": 7.809303051499259e-05, "epoch": 0.9440337909186906, "percentage": 18.9, "elapsed_time": "5:59:25", "remaining_time": "1 day, 1:42:13"}
+{"current_steps": 448, "total_steps": 2365, "loss": 1.3155, "lr": 7.807497309739612e-05, "epoch": 0.9461457233368532, "percentage": 18.94, "elapsed_time": "6:00:13", "remaining_time": "1 day, 1:41:23"}
+{"current_steps": 449, "total_steps": 2365, "loss": 1.3154, "lr": 7.805683269548255e-05, "epoch": 0.9482576557550159, "percentage": 18.99, "elapsed_time": "6:01:01", "remaining_time": "1 day, 1:40:34"}
+{"current_steps": 450, "total_steps": 2365, "loss": 1.3251, "lr": 7.803860934878885e-05, "epoch": 0.9503695881731784, "percentage": 19.03, "elapsed_time": "6:01:49", "remaining_time": "1 day, 1:39:46"}
+{"current_steps": 451, "total_steps": 2365, "loss": 1.3029, "lr": 7.802030309703276e-05, "epoch": 0.9524815205913411, "percentage": 19.07, "elapsed_time": "6:02:38", "remaining_time": "1 day, 1:38:59"}
+{"current_steps": 452, "total_steps": 2365, "loss": 1.3122, "lr": 7.80019139801127e-05, "epoch": 0.9545934530095037, "percentage": 19.11, "elapsed_time": "6:03:26", "remaining_time": "1 day, 1:38:11"}
+{"current_steps": 453, "total_steps": 2365, "loss": 1.3099, "lr": 7.798344203810774e-05, "epoch": 0.9567053854276664, "percentage": 19.15, "elapsed_time": "6:04:15", "remaining_time": "1 day, 1:37:25"}
+{"current_steps": 454, "total_steps": 2365, "loss": 1.336, "lr": 7.796488731127739e-05, "epoch": 0.9588173178458289, "percentage": 19.2, "elapsed_time": "6:05:04", "remaining_time": "1 day, 1:36:39"}
+{"current_steps": 455, "total_steps": 2365, "loss": 1.3102, "lr": 7.794624984006169e-05, "epoch": 0.9609292502639916, "percentage": 19.24, "elapsed_time": "6:05:54", "remaining_time": "1 day, 1:35:59"}
+{"current_steps": 456, "total_steps": 2365, "loss": 1.3118, "lr": 7.792752966508091e-05, "epoch": 0.9630411826821542, "percentage": 19.28, "elapsed_time": "6:06:43", "remaining_time": "1 day, 1:35:14"}
+{"current_steps": 457, "total_steps": 2365, "loss": 1.3226, "lr": 7.790872682713567e-05, "epoch": 0.9651531151003168, "percentage": 19.32, "elapsed_time": "6:07:32", "remaining_time": "1 day, 1:34:29"}
+{"current_steps": 458, "total_steps": 2365, "loss": 1.3122, "lr": 7.78898413672067e-05, "epoch": 0.9672650475184794, "percentage": 19.37, "elapsed_time": "6:08:20", "remaining_time": "1 day, 1:33:42"}
+{"current_steps": 459, "total_steps": 2365, "loss": 1.3179, "lr": 7.787087332645481e-05, "epoch": 0.9693769799366421, "percentage": 19.41, "elapsed_time": "6:09:09", "remaining_time": "1 day, 1:32:56"}
+{"current_steps": 460, "total_steps": 2365, "loss": 1.3005, "lr": 7.785182274622081e-05, "epoch": 0.9714889123548046, "percentage": 19.45, "elapsed_time": "6:09:57", "remaining_time": "1 day, 1:32:08"}
+{"current_steps": 461, "total_steps": 2365, "loss": 1.327, "lr": 7.783268966802539e-05, "epoch": 0.9736008447729673, "percentage": 19.49, "elapsed_time": "6:10:46", "remaining_time": "1 day, 1:31:20"}
+{"current_steps": 462, "total_steps": 2365, "loss": 1.3209, "lr": 7.781347413356906e-05, "epoch": 0.9757127771911299, "percentage": 19.53, "elapsed_time": "6:11:34", "remaining_time": "1 day, 1:30:33"}
+{"current_steps": 463, "total_steps": 2365, "loss": 1.3113, "lr": 7.779417618473203e-05, "epoch": 0.9778247096092925, "percentage": 19.58, "elapsed_time": "6:12:22", "remaining_time": "1 day, 1:29:43"}
+{"current_steps": 464, "total_steps": 2365, "loss": 1.2957, "lr": 7.777479586357412e-05, "epoch": 0.9799366420274551, "percentage": 19.62, "elapsed_time": "6:13:10", "remaining_time": "1 day, 1:28:53"}
+{"current_steps": 465, "total_steps": 2365, "loss": 1.3076, "lr": 7.775533321233471e-05, "epoch": 0.9820485744456178, "percentage": 19.66, "elapsed_time": "6:13:59", "remaining_time": "1 day, 1:28:07"}
+{"current_steps": 466, "total_steps": 2365, "loss": 1.3375, "lr": 7.77357882734326e-05, "epoch": 0.9841605068637803, "percentage": 19.7, "elapsed_time": "6:14:47", "remaining_time": "1 day, 1:27:19"}
+{"current_steps": 467, "total_steps": 2365, "loss": 1.308, "lr": 7.771616108946596e-05, "epoch": 0.986272439281943, "percentage": 19.75, "elapsed_time": "6:15:35", "remaining_time": "1 day, 1:26:30"}
+{"current_steps": 468, "total_steps": 2365, "loss": 1.3313, "lr": 7.769645170321216e-05, "epoch": 0.9883843717001056, "percentage": 19.79, "elapsed_time": "6:16:23", "remaining_time": "1 day, 1:25:42"}
+{"current_steps": 469, "total_steps": 2365, "loss": 1.3147, "lr": 7.767666015762776e-05, "epoch": 0.9904963041182682, "percentage": 19.83, "elapsed_time": "6:17:12", "remaining_time": "1 day, 1:24:53"}
+{"current_steps": 470, "total_steps": 2365, "loss": 1.2932, "lr": 7.765678649584843e-05, "epoch": 0.9926082365364308, "percentage": 19.87, "elapsed_time": "6:18:00", "remaining_time": "1 day, 1:24:05"}
+{"current_steps": 471, "total_steps": 2365, "loss": 1.3195, "lr": 7.763683076118875e-05, "epoch": 0.9947201689545935, "percentage": 19.92, "elapsed_time": "6:18:48", "remaining_time": "1 day, 1:23:16"}
+{"current_steps": 472, "total_steps": 2365, "loss": 1.3188, "lr": 7.761679299714218e-05, "epoch": 0.996832101372756, "percentage": 19.96, "elapsed_time": "6:19:36", "remaining_time": "1 day, 1:22:28"}
+{"current_steps": 473, "total_steps": 2365, "loss": 1.3032, "lr": 7.7596673247381e-05, "epoch": 0.9989440337909187, "percentage": 20.0, "elapsed_time": "6:20:25", "remaining_time": "1 day, 1:21:40"}
+{"current_steps": 474, "total_steps": 2365, "loss": 1.3046, "lr": 7.757647155575616e-05, "epoch": 1.0013199577613516, "percentage": 20.04, "elapsed_time": "6:22:54", "remaining_time": "1 day, 1:27:35"}
+{"current_steps": 475, "total_steps": 2365, "loss": 1.2819, "lr": 7.755618796629719e-05, "epoch": 1.0034318901795143, "percentage": 20.08, "elapsed_time": "6:23:42", "remaining_time": "1 day, 1:26:46"}
+{"current_steps": 476, "total_steps": 2365, "loss": 1.3082, "lr": 7.753582252321213e-05, "epoch": 1.005543822597677, "percentage": 20.13, "elapsed_time": "6:24:30", "remaining_time": "1 day, 1:25:56"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aefca7bf0b5ddb354866b56c6aaf99b0f5440089ab44b7712b0788afddb6423d
+size 7352