Training in progress, epoch 1

.gitattributes

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

added_tokens.json

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

config.json

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

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7515cadec741118460eb508dbcfa33dd55f8ec86fc07cf24474eaf9f64b4d933
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cab42a0e165ebfead16b24cb89fb8378950bf7a7ffd2d29d859385c03217008
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8efb2470a28e777ffd496f2102fcfc6a8f9c23577b12a6c3977ada171d3fc820
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d62fc70128ef52b951bac11e3abcf499de28e16a419f2397dd0fbc514ed887ea
+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 +1,503 @@
+{"current_steps": 1, "total_steps": 1479, "loss": 0.8777, "lr": 6.756756756756757e-08, "epoch": 0.002028397565922921, "percentage": 0.07, "elapsed_time": "0:01:20", "remaining_time": "1 day, 9:08:53"}
+{"current_steps": 2, "total_steps": 1479, "loss": 0.8494, "lr": 1.3513513513513515e-07, "epoch": 0.004056795131845842, "percentage": 0.14, "elapsed_time": "0:01:33", "remaining_time": "19:08:26"}
+{"current_steps": 3, "total_steps": 1479, "loss": 0.8494, "lr": 2.0270270270270273e-07, "epoch": 0.006085192697768763, "percentage": 0.2, "elapsed_time": "0:01:46", "remaining_time": "14:29:28"}
+{"current_steps": 4, "total_steps": 1479, "loss": 0.8642, "lr": 2.702702702702703e-07, "epoch": 0.008113590263691683, "percentage": 0.27, "elapsed_time": "0:01:58", "remaining_time": "12:10:19"}
+{"current_steps": 5, "total_steps": 1479, "loss": 0.8593, "lr": 3.378378378378379e-07, "epoch": 0.010141987829614604, "percentage": 0.34, "elapsed_time": "0:02:11", "remaining_time": "10:46:32"}
+{"current_steps": 6, "total_steps": 1479, "loss": 0.8564, "lr": 4.0540540540540546e-07, "epoch": 0.012170385395537525, "percentage": 0.41, "elapsed_time": "0:02:24", "remaining_time": "9:50:28"}
+{"current_steps": 7, "total_steps": 1479, "loss": 0.8638, "lr": 4.7297297297297305e-07, "epoch": 0.014198782961460446, "percentage": 0.47, "elapsed_time": "0:02:37", "remaining_time": "9:10:23"}
+{"current_steps": 8, "total_steps": 1479, "loss": 0.8358, "lr": 5.405405405405406e-07, "epoch": 0.016227180527383367, "percentage": 0.54, "elapsed_time": "0:02:49", "remaining_time": "8:40:24"}
+{"current_steps": 9, "total_steps": 1479, "loss": 0.8596, "lr": 6.081081081081082e-07, "epoch": 0.018255578093306288, "percentage": 0.61, "elapsed_time": "0:03:02", "remaining_time": "8:17:03"}
+{"current_steps": 10, "total_steps": 1479, "loss": 0.8546, "lr": 6.756756756756758e-07, "epoch": 0.02028397565922921, "percentage": 0.68, "elapsed_time": "0:03:15", "remaining_time": "7:58:17"}
+{"current_steps": 11, "total_steps": 1479, "loss": 0.8614, "lr": 7.432432432432434e-07, "epoch": 0.02231237322515213, "percentage": 0.74, "elapsed_time": "0:03:28", "remaining_time": "7:43:05"}
+{"current_steps": 12, "total_steps": 1479, "loss": 0.8695, "lr": 8.108108108108109e-07, "epoch": 0.02434077079107505, "percentage": 0.81, "elapsed_time": "0:03:40", "remaining_time": "7:30:15"}
+{"current_steps": 13, "total_steps": 1479, "loss": 0.8163, "lr": 8.783783783783785e-07, "epoch": 0.02636916835699797, "percentage": 0.88, "elapsed_time": "0:03:53", "remaining_time": "7:19:14"}
+{"current_steps": 14, "total_steps": 1479, "loss": 0.7996, "lr": 9.459459459459461e-07, "epoch": 0.028397565922920892, "percentage": 0.95, "elapsed_time": "0:04:06", "remaining_time": "7:09:55"}
+{"current_steps": 15, "total_steps": 1479, "loss": 0.806, "lr": 1.0135135135135136e-06, "epoch": 0.030425963488843813, "percentage": 1.01, "elapsed_time": "0:04:19", "remaining_time": "7:01:44"}
+{"current_steps": 16, "total_steps": 1479, "loss": 0.8158, "lr": 1.0810810810810812e-06, "epoch": 0.032454361054766734, "percentage": 1.08, "elapsed_time": "0:04:32", "remaining_time": "6:54:35"}
+{"current_steps": 17, "total_steps": 1479, "loss": 0.7731, "lr": 1.148648648648649e-06, "epoch": 0.034482758620689655, "percentage": 1.15, "elapsed_time": "0:04:44", "remaining_time": "6:48:12"}
+{"current_steps": 18, "total_steps": 1479, "loss": 0.7645, "lr": 1.2162162162162164e-06, "epoch": 0.036511156186612576, "percentage": 1.22, "elapsed_time": "0:04:57", "remaining_time": "6:42:34"}
+{"current_steps": 19, "total_steps": 1479, "loss": 0.7879, "lr": 1.2837837837837838e-06, "epoch": 0.038539553752535496, "percentage": 1.28, "elapsed_time": "0:05:10", "remaining_time": "6:37:26"}
+{"current_steps": 20, "total_steps": 1479, "loss": 0.754, "lr": 1.3513513513513515e-06, "epoch": 0.04056795131845842, "percentage": 1.35, "elapsed_time": "0:05:23", "remaining_time": "6:32:48"}
+{"current_steps": 21, "total_steps": 1479, "loss": 0.7749, "lr": 1.418918918918919e-06, "epoch": 0.04259634888438134, "percentage": 1.42, "elapsed_time": "0:05:35", "remaining_time": "6:28:37"}
+{"current_steps": 22, "total_steps": 1479, "loss": 0.7579, "lr": 1.4864864864864868e-06, "epoch": 0.04462474645030426, "percentage": 1.49, "elapsed_time": "0:05:48", "remaining_time": "6:24:48"}
+{"current_steps": 23, "total_steps": 1479, "loss": 0.7208, "lr": 1.5540540540540541e-06, "epoch": 0.04665314401622718, "percentage": 1.56, "elapsed_time": "0:06:01", "remaining_time": "6:21:20"}
+{"current_steps": 24, "total_steps": 1479, "loss": 0.7213, "lr": 1.6216216216216219e-06, "epoch": 0.0486815415821501, "percentage": 1.62, "elapsed_time": "0:06:14", "remaining_time": "6:18:04"}
+{"current_steps": 25, "total_steps": 1479, "loss": 0.7257, "lr": 1.6891891891891894e-06, "epoch": 0.05070993914807302, "percentage": 1.69, "elapsed_time": "0:06:26", "remaining_time": "6:15:01"}
+{"current_steps": 26, "total_steps": 1479, "loss": 0.7203, "lr": 1.756756756756757e-06, "epoch": 0.05273833671399594, "percentage": 1.76, "elapsed_time": "0:06:39", "remaining_time": "6:12:16"}
+{"current_steps": 27, "total_steps": 1479, "loss": 0.7276, "lr": 1.8243243243243245e-06, "epoch": 0.05476673427991886, "percentage": 1.83, "elapsed_time": "0:06:52", "remaining_time": "6:10:04"}
+{"current_steps": 28, "total_steps": 1479, "loss": 0.6903, "lr": 1.8918918918918922e-06, "epoch": 0.056795131845841784, "percentage": 1.89, "elapsed_time": "0:07:05", "remaining_time": "6:07:39"}
+{"current_steps": 29, "total_steps": 1479, "loss": 0.6985, "lr": 1.9594594594594595e-06, "epoch": 0.058823529411764705, "percentage": 1.96, "elapsed_time": "0:07:18", "remaining_time": "6:05:24"}
+{"current_steps": 30, "total_steps": 1479, "loss": 0.6946, "lr": 2.0270270270270273e-06, "epoch": 0.060851926977687626, "percentage": 2.03, "elapsed_time": "0:07:31", "remaining_time": "6:03:18"}
+{"current_steps": 31, "total_steps": 1479, "loss": 0.6879, "lr": 2.0945945945945946e-06, "epoch": 0.06288032454361055, "percentage": 2.1, "elapsed_time": "0:07:44", "remaining_time": "6:01:21"}
+{"current_steps": 32, "total_steps": 1479, "loss": 0.6734, "lr": 2.1621621621621623e-06, "epoch": 0.06490872210953347, "percentage": 2.16, "elapsed_time": "0:07:56", "remaining_time": "5:59:22"}
+{"current_steps": 33, "total_steps": 1479, "loss": 0.667, "lr": 2.22972972972973e-06, "epoch": 0.06693711967545639, "percentage": 2.23, "elapsed_time": "0:08:10", "remaining_time": "5:57:52"}
+{"current_steps": 34, "total_steps": 1479, "loss": 0.6736, "lr": 2.297297297297298e-06, "epoch": 0.06896551724137931, "percentage": 2.3, "elapsed_time": "0:08:22", "remaining_time": "5:56:08"}
+{"current_steps": 35, "total_steps": 1479, "loss": 0.6692, "lr": 2.364864864864865e-06, "epoch": 0.07099391480730223, "percentage": 2.37, "elapsed_time": "0:08:35", "remaining_time": "5:54:28"}
+{"current_steps": 36, "total_steps": 1479, "loss": 0.6576, "lr": 2.432432432432433e-06, "epoch": 0.07302231237322515, "percentage": 2.43, "elapsed_time": "0:08:48", "remaining_time": "5:52:53"}
+{"current_steps": 37, "total_steps": 1479, "loss": 0.6692, "lr": 2.5e-06, "epoch": 0.07505070993914807, "percentage": 2.5, "elapsed_time": "0:09:00", "remaining_time": "5:51:21"}
+{"current_steps": 38, "total_steps": 1479, "loss": 0.6557, "lr": 2.5675675675675675e-06, "epoch": 0.07707910750507099, "percentage": 2.57, "elapsed_time": "0:09:13", "remaining_time": "5:49:56"}
+{"current_steps": 39, "total_steps": 1479, "loss": 0.6633, "lr": 2.6351351351351353e-06, "epoch": 0.07910750507099391, "percentage": 2.64, "elapsed_time": "0:09:26", "remaining_time": "5:48:35"}
+{"current_steps": 40, "total_steps": 1479, "loss": 0.6717, "lr": 2.702702702702703e-06, "epoch": 0.08113590263691683, "percentage": 2.7, "elapsed_time": "0:09:39", "remaining_time": "5:47:18"}
+{"current_steps": 41, "total_steps": 1479, "loss": 0.6615, "lr": 2.7702702702702703e-06, "epoch": 0.08316430020283976, "percentage": 2.77, "elapsed_time": "0:09:52", "remaining_time": "5:46:03"}
+{"current_steps": 42, "total_steps": 1479, "loss": 0.6589, "lr": 2.837837837837838e-06, "epoch": 0.08519269776876268, "percentage": 2.84, "elapsed_time": "0:10:04", "remaining_time": "5:44:51"}
+{"current_steps": 43, "total_steps": 1479, "loss": 0.6577, "lr": 2.9054054054054054e-06, "epoch": 0.0872210953346856, "percentage": 2.91, "elapsed_time": "0:10:17", "remaining_time": "5:43:45"}
+{"current_steps": 44, "total_steps": 1479, "loss": 0.6064, "lr": 2.9729729729729736e-06, "epoch": 0.08924949290060852, "percentage": 2.97, "elapsed_time": "0:10:30", "remaining_time": "5:42:38"}
+{"current_steps": 45, "total_steps": 1479, "loss": 0.6082, "lr": 3.040540540540541e-06, "epoch": 0.09127789046653144, "percentage": 3.04, "elapsed_time": "0:10:43", "remaining_time": "5:41:34"}
+{"current_steps": 46, "total_steps": 1479, "loss": 0.6027, "lr": 3.1081081081081082e-06, "epoch": 0.09330628803245436, "percentage": 3.11, "elapsed_time": "0:10:55", "remaining_time": "5:40:30"}
+{"current_steps": 47, "total_steps": 1479, "loss": 0.6146, "lr": 3.1756756756756755e-06, "epoch": 0.09533468559837728, "percentage": 3.18, "elapsed_time": "0:11:08", "remaining_time": "5:39:30"}
+{"current_steps": 48, "total_steps": 1479, "loss": 0.6098, "lr": 3.2432432432432437e-06, "epoch": 0.0973630831643002, "percentage": 3.25, "elapsed_time": "0:11:21", "remaining_time": "5:38:33"}
+{"current_steps": 49, "total_steps": 1479, "loss": 0.6215, "lr": 3.310810810810811e-06, "epoch": 0.09939148073022312, "percentage": 3.31, "elapsed_time": "0:11:34", "remaining_time": "5:37:35"}
+{"current_steps": 50, "total_steps": 1479, "loss": 0.6109, "lr": 3.3783783783783788e-06, "epoch": 0.10141987829614604, "percentage": 3.38, "elapsed_time": "0:11:46", "remaining_time": "5:36:43"}
+{"current_steps": 51, "total_steps": 1479, "loss": 0.6035, "lr": 3.445945945945946e-06, "epoch": 0.10344827586206896, "percentage": 3.45, "elapsed_time": "0:11:59", "remaining_time": "5:35:51"}
+{"current_steps": 52, "total_steps": 1479, "loss": 0.5938, "lr": 3.513513513513514e-06, "epoch": 0.10547667342799188, "percentage": 3.52, "elapsed_time": "0:12:12", "remaining_time": "5:34:59"}
+{"current_steps": 53, "total_steps": 1479, "loss": 0.5948, "lr": 3.5810810810810816e-06, "epoch": 0.1075050709939148, "percentage": 3.58, "elapsed_time": "0:12:25", "remaining_time": "5:34:10"}
+{"current_steps": 54, "total_steps": 1479, "loss": 0.6032, "lr": 3.648648648648649e-06, "epoch": 0.10953346855983773, "percentage": 3.65, "elapsed_time": "0:12:37", "remaining_time": "5:33:21"}
+{"current_steps": 55, "total_steps": 1479, "loss": 0.5888, "lr": 3.7162162162162162e-06, "epoch": 0.11156186612576065, "percentage": 3.72, "elapsed_time": "0:12:50", "remaining_time": "5:32:35"}
+{"current_steps": 56, "total_steps": 1479, "loss": 0.58, "lr": 3.7837837837837844e-06, "epoch": 0.11359026369168357, "percentage": 3.79, "elapsed_time": "0:13:03", "remaining_time": "5:31:48"}
+{"current_steps": 57, "total_steps": 1479, "loss": 0.564, "lr": 3.851351351351352e-06, "epoch": 0.11561866125760649, "percentage": 3.85, "elapsed_time": "0:13:16", "remaining_time": "5:31:05"}
+{"current_steps": 58, "total_steps": 1479, "loss": 0.5896, "lr": 3.918918918918919e-06, "epoch": 0.11764705882352941, "percentage": 3.92, "elapsed_time": "0:13:29", "remaining_time": "5:30:22"}
+{"current_steps": 59, "total_steps": 1479, "loss": 0.5712, "lr": 3.986486486486487e-06, "epoch": 0.11967545638945233, "percentage": 3.99, "elapsed_time": "0:13:41", "remaining_time": "5:29:38"}
+{"current_steps": 60, "total_steps": 1479, "loss": 0.583, "lr": 4.0540540540540545e-06, "epoch": 0.12170385395537525, "percentage": 4.06, "elapsed_time": "0:13:54", "remaining_time": "5:28:58"}
+{"current_steps": 61, "total_steps": 1479, "loss": 0.5587, "lr": 4.121621621621622e-06, "epoch": 0.12373225152129817, "percentage": 4.12, "elapsed_time": "0:14:07", "remaining_time": "5:28:17"}
+{"current_steps": 62, "total_steps": 1479, "loss": 0.597, "lr": 4.189189189189189e-06, "epoch": 0.1257606490872211, "percentage": 4.19, "elapsed_time": "0:14:20", "remaining_time": "5:27:38"}
+{"current_steps": 63, "total_steps": 1479, "loss": 0.5765, "lr": 4.256756756756757e-06, "epoch": 0.12778904665314403, "percentage": 4.26, "elapsed_time": "0:14:32", "remaining_time": "5:27:00"}
+{"current_steps": 64, "total_steps": 1479, "loss": 0.5993, "lr": 4.324324324324325e-06, "epoch": 0.12981744421906694, "percentage": 4.33, "elapsed_time": "0:14:45", "remaining_time": "5:26:23"}
+{"current_steps": 65, "total_steps": 1479, "loss": 0.5874, "lr": 4.391891891891892e-06, "epoch": 0.13184584178498987, "percentage": 4.39, "elapsed_time": "0:14:58", "remaining_time": "5:25:45"}
+{"current_steps": 66, "total_steps": 1479, "loss": 0.5537, "lr": 4.45945945945946e-06, "epoch": 0.13387423935091278, "percentage": 4.46, "elapsed_time": "0:15:11", "remaining_time": "5:25:08"}
+{"current_steps": 67, "total_steps": 1479, "loss": 0.5672, "lr": 4.527027027027027e-06, "epoch": 0.1359026369168357, "percentage": 4.53, "elapsed_time": "0:15:24", "remaining_time": "5:24:33"}
+{"current_steps": 68, "total_steps": 1479, "loss": 0.5826, "lr": 4.594594594594596e-06, "epoch": 0.13793103448275862, "percentage": 4.6, "elapsed_time": "0:15:36", "remaining_time": "5:23:59"}
+{"current_steps": 69, "total_steps": 1479, "loss": 0.5587, "lr": 4.6621621621621625e-06, "epoch": 0.13995943204868155, "percentage": 4.67, "elapsed_time": "0:15:49", "remaining_time": "5:23:23"}
+{"current_steps": 70, "total_steps": 1479, "loss": 0.5704, "lr": 4.72972972972973e-06, "epoch": 0.14198782961460446, "percentage": 4.73, "elapsed_time": "0:16:02", "remaining_time": "5:22:48"}
+{"current_steps": 71, "total_steps": 1479, "loss": 0.5946, "lr": 4.797297297297297e-06, "epoch": 0.1440162271805274, "percentage": 4.8, "elapsed_time": "0:16:14", "remaining_time": "5:22:14"}
+{"current_steps": 72, "total_steps": 1479, "loss": 0.5789, "lr": 4.864864864864866e-06, "epoch": 0.1460446247464503, "percentage": 4.87, "elapsed_time": "0:16:27", "remaining_time": "5:21:41"}
+{"current_steps": 73, "total_steps": 1479, "loss": 0.5786, "lr": 4.932432432432433e-06, "epoch": 0.14807302231237324, "percentage": 4.94, "elapsed_time": "0:16:40", "remaining_time": "5:21:08"}
+{"current_steps": 74, "total_steps": 1479, "loss": 0.5598, "lr": 5e-06, "epoch": 0.15010141987829614, "percentage": 5.0, "elapsed_time": "0:16:53", "remaining_time": "5:20:35"}
+{"current_steps": 75, "total_steps": 1479, "loss": 0.5779, "lr": 5.067567567567568e-06, "epoch": 0.15212981744421908, "percentage": 5.07, "elapsed_time": "0:17:06", "remaining_time": "5:20:10"}
+{"current_steps": 76, "total_steps": 1479, "loss": 0.5653, "lr": 5.135135135135135e-06, "epoch": 0.15415821501014199, "percentage": 5.14, "elapsed_time": "0:17:18", "remaining_time": "5:19:39"}
+{"current_steps": 77, "total_steps": 1479, "loss": 0.5698, "lr": 5.202702702702704e-06, "epoch": 0.15618661257606492, "percentage": 5.21, "elapsed_time": "0:17:31", "remaining_time": "5:19:08"}
+{"current_steps": 78, "total_steps": 1479, "loss": 0.5488, "lr": 5.2702702702702705e-06, "epoch": 0.15821501014198783, "percentage": 5.27, "elapsed_time": "0:17:44", "remaining_time": "5:18:46"}
+{"current_steps": 79, "total_steps": 1479, "loss": 0.546, "lr": 5.337837837837838e-06, "epoch": 0.16024340770791076, "percentage": 5.34, "elapsed_time": "0:17:57", "remaining_time": "5:18:17"}
+{"current_steps": 80, "total_steps": 1479, "loss": 0.5623, "lr": 5.405405405405406e-06, "epoch": 0.16227180527383367, "percentage": 5.41, "elapsed_time": "0:18:10", "remaining_time": "5:17:47"}
+{"current_steps": 81, "total_steps": 1479, "loss": 0.5674, "lr": 5.472972972972973e-06, "epoch": 0.1643002028397566, "percentage": 5.48, "elapsed_time": "0:18:23", "remaining_time": "5:17:19"}
+{"current_steps": 82, "total_steps": 1479, "loss": 0.5369, "lr": 5.540540540540541e-06, "epoch": 0.1663286004056795, "percentage": 5.54, "elapsed_time": "0:18:35", "remaining_time": "5:16:50"}
+{"current_steps": 83, "total_steps": 1479, "loss": 0.5464, "lr": 5.608108108108109e-06, "epoch": 0.16835699797160245, "percentage": 5.61, "elapsed_time": "0:18:48", "remaining_time": "5:16:22"}
+{"current_steps": 84, "total_steps": 1479, "loss": 0.532, "lr": 5.675675675675676e-06, "epoch": 0.17038539553752535, "percentage": 5.68, "elapsed_time": "0:19:01", "remaining_time": "5:15:54"}
+{"current_steps": 85, "total_steps": 1479, "loss": 0.5791, "lr": 5.743243243243244e-06, "epoch": 0.1724137931034483, "percentage": 5.75, "elapsed_time": "0:19:14", "remaining_time": "5:15:26"}
+{"current_steps": 86, "total_steps": 1479, "loss": 0.5706, "lr": 5.810810810810811e-06, "epoch": 0.1744421906693712, "percentage": 5.81, "elapsed_time": "0:19:26", "remaining_time": "5:15:00"}
+{"current_steps": 87, "total_steps": 1479, "loss": 0.5683, "lr": 5.8783783783783786e-06, "epoch": 0.17647058823529413, "percentage": 5.88, "elapsed_time": "0:19:39", "remaining_time": "5:14:34"}
+{"current_steps": 88, "total_steps": 1479, "loss": 0.5298, "lr": 5.945945945945947e-06, "epoch": 0.17849898580121704, "percentage": 5.95, "elapsed_time": "0:19:52", "remaining_time": "5:14:07"}
+{"current_steps": 89, "total_steps": 1479, "loss": 0.5571, "lr": 6.013513513513514e-06, "epoch": 0.18052738336713997, "percentage": 6.02, "elapsed_time": "0:20:05", "remaining_time": "5:13:41"}
+{"current_steps": 90, "total_steps": 1479, "loss": 0.5343, "lr": 6.081081081081082e-06, "epoch": 0.18255578093306288, "percentage": 6.09, "elapsed_time": "0:20:17", "remaining_time": "5:13:15"}
+{"current_steps": 91, "total_steps": 1479, "loss": 0.5398, "lr": 6.1486486486486495e-06, "epoch": 0.1845841784989858, "percentage": 6.15, "elapsed_time": "0:20:30", "remaining_time": "5:12:50"}
+{"current_steps": 92, "total_steps": 1479, "loss": 0.5395, "lr": 6.2162162162162164e-06, "epoch": 0.18661257606490872, "percentage": 6.22, "elapsed_time": "0:20:43", "remaining_time": "5:12:25"}
+{"current_steps": 93, "total_steps": 1479, "loss": 0.525, "lr": 6.283783783783784e-06, "epoch": 0.18864097363083165, "percentage": 6.29, "elapsed_time": "0:20:56", "remaining_time": "5:12:00"}
+{"current_steps": 94, "total_steps": 1479, "loss": 0.5284, "lr": 6.351351351351351e-06, "epoch": 0.19066937119675456, "percentage": 6.36, "elapsed_time": "0:21:09", "remaining_time": "5:11:42"}
+{"current_steps": 95, "total_steps": 1479, "loss": 0.5544, "lr": 6.41891891891892e-06, "epoch": 0.1926977687626775, "percentage": 6.42, "elapsed_time": "0:21:22", "remaining_time": "5:11:17"}
+{"current_steps": 96, "total_steps": 1479, "loss": 0.5431, "lr": 6.486486486486487e-06, "epoch": 0.1947261663286004, "percentage": 6.49, "elapsed_time": "0:21:34", "remaining_time": "5:10:54"}
+{"current_steps": 97, "total_steps": 1479, "loss": 0.5107, "lr": 6.554054054054054e-06, "epoch": 0.19675456389452334, "percentage": 6.56, "elapsed_time": "0:21:47", "remaining_time": "5:10:29"}
+{"current_steps": 98, "total_steps": 1479, "loss": 0.5635, "lr": 6.621621621621622e-06, "epoch": 0.19878296146044624, "percentage": 6.63, "elapsed_time": "0:22:00", "remaining_time": "5:10:06"}
+{"current_steps": 99, "total_steps": 1479, "loss": 0.5279, "lr": 6.689189189189191e-06, "epoch": 0.20081135902636918, "percentage": 6.69, "elapsed_time": "0:22:13", "remaining_time": "5:09:43"}
+{"current_steps": 100, "total_steps": 1479, "loss": 0.539, "lr": 6.7567567567567575e-06, "epoch": 0.2028397565922921, "percentage": 6.76, "elapsed_time": "0:22:25", "remaining_time": "5:09:20"}
+{"current_steps": 101, "total_steps": 1479, "loss": 0.5228, "lr": 6.824324324324325e-06, "epoch": 0.20486815415821502, "percentage": 6.83, "elapsed_time": "0:22:38", "remaining_time": "5:08:57"}
+{"current_steps": 102, "total_steps": 1479, "loss": 0.5185, "lr": 6.891891891891892e-06, "epoch": 0.20689655172413793, "percentage": 6.9, "elapsed_time": "0:22:51", "remaining_time": "5:08:34"}
+{"current_steps": 103, "total_steps": 1479, "loss": 0.5231, "lr": 6.95945945945946e-06, "epoch": 0.20892494929006086, "percentage": 6.96, "elapsed_time": "0:23:04", "remaining_time": "5:08:13"}
+{"current_steps": 104, "total_steps": 1479, "loss": 0.5511, "lr": 7.027027027027028e-06, "epoch": 0.21095334685598377, "percentage": 7.03, "elapsed_time": "0:23:17", "remaining_time": "5:07:50"}
+{"current_steps": 105, "total_steps": 1479, "loss": 0.5357, "lr": 7.0945945945945946e-06, "epoch": 0.2129817444219067, "percentage": 7.1, "elapsed_time": "0:23:29", "remaining_time": "5:07:27"}
+{"current_steps": 106, "total_steps": 1479, "loss": 0.5341, "lr": 7.162162162162163e-06, "epoch": 0.2150101419878296, "percentage": 7.17, "elapsed_time": "0:23:42", "remaining_time": "5:07:06"}
+{"current_steps": 107, "total_steps": 1479, "loss": 0.5308, "lr": 7.229729729729731e-06, "epoch": 0.21703853955375255, "percentage": 7.23, "elapsed_time": "0:23:55", "remaining_time": "5:06:44"}
+{"current_steps": 108, "total_steps": 1479, "loss": 0.5406, "lr": 7.297297297297298e-06, "epoch": 0.21906693711967545, "percentage": 7.3, "elapsed_time": "0:24:08", "remaining_time": "5:06:23"}
+{"current_steps": 109, "total_steps": 1479, "loss": 0.5248, "lr": 7.3648648648648655e-06, "epoch": 0.2210953346855984, "percentage": 7.37, "elapsed_time": "0:24:20", "remaining_time": "5:06:01"}
+{"current_steps": 110, "total_steps": 1479, "loss": 0.5448, "lr": 7.4324324324324324e-06, "epoch": 0.2231237322515213, "percentage": 7.44, "elapsed_time": "0:24:33", "remaining_time": "5:05:39"}
+{"current_steps": 111, "total_steps": 1479, "loss": 0.528, "lr": 7.500000000000001e-06, "epoch": 0.22515212981744423, "percentage": 7.51, "elapsed_time": "0:24:46", "remaining_time": "5:05:18"}
+{"current_steps": 112, "total_steps": 1479, "loss": 0.5283, "lr": 7.567567567567569e-06, "epoch": 0.22718052738336714, "percentage": 7.57, "elapsed_time": "0:24:59", "remaining_time": "5:04:57"}
+{"current_steps": 113, "total_steps": 1479, "loss": 0.5197, "lr": 7.635135135135135e-06, "epoch": 0.22920892494929007, "percentage": 7.64, "elapsed_time": "0:25:11", "remaining_time": "5:04:36"}
+{"current_steps": 114, "total_steps": 1479, "loss": 0.5253, "lr": 7.702702702702704e-06, "epoch": 0.23123732251521298, "percentage": 7.71, "elapsed_time": "0:25:24", "remaining_time": "5:04:16"}
+{"current_steps": 115, "total_steps": 1479, "loss": 0.524, "lr": 7.77027027027027e-06, "epoch": 0.2332657200811359, "percentage": 7.78, "elapsed_time": "0:25:37", "remaining_time": "5:03:55"}
+{"current_steps": 116, "total_steps": 1479, "loss": 0.5486, "lr": 7.837837837837838e-06, "epoch": 0.23529411764705882, "percentage": 7.84, "elapsed_time": "0:25:50", "remaining_time": "5:03:35"}
+{"current_steps": 117, "total_steps": 1479, "loss": 0.5421, "lr": 7.905405405405406e-06, "epoch": 0.23732251521298176, "percentage": 7.91, "elapsed_time": "0:26:02", "remaining_time": "5:03:14"}
+{"current_steps": 118, "total_steps": 1479, "loss": 0.5254, "lr": 7.972972972972974e-06, "epoch": 0.23935091277890466, "percentage": 7.98, "elapsed_time": "0:26:15", "remaining_time": "5:02:54"}
+{"current_steps": 119, "total_steps": 1479, "loss": 0.5266, "lr": 8.040540540540541e-06, "epoch": 0.2413793103448276, "percentage": 8.05, "elapsed_time": "0:26:28", "remaining_time": "5:02:34"}
+{"current_steps": 120, "total_steps": 1479, "loss": 0.5372, "lr": 8.108108108108109e-06, "epoch": 0.2434077079107505, "percentage": 8.11, "elapsed_time": "0:26:41", "remaining_time": "5:02:14"}
+{"current_steps": 121, "total_steps": 1479, "loss": 0.5128, "lr": 8.175675675675677e-06, "epoch": 0.24543610547667344, "percentage": 8.18, "elapsed_time": "0:26:53", "remaining_time": "5:01:53"}
+{"current_steps": 122, "total_steps": 1479, "loss": 0.5303, "lr": 8.243243243243245e-06, "epoch": 0.24746450304259635, "percentage": 8.25, "elapsed_time": "0:27:06", "remaining_time": "5:01:34"}
+{"current_steps": 123, "total_steps": 1479, "loss": 0.5223, "lr": 8.31081081081081e-06, "epoch": 0.24949290060851928, "percentage": 8.32, "elapsed_time": "0:27:19", "remaining_time": "5:01:14"}
+{"current_steps": 124, "total_steps": 1479, "loss": 0.5264, "lr": 8.378378378378378e-06, "epoch": 0.2515212981744422, "percentage": 8.38, "elapsed_time": "0:27:32", "remaining_time": "5:00:54"}
+{"current_steps": 125, "total_steps": 1479, "loss": 0.5214, "lr": 8.445945945945948e-06, "epoch": 0.2535496957403651, "percentage": 8.45, "elapsed_time": "0:27:44", "remaining_time": "5:00:34"}
+{"current_steps": 126, "total_steps": 1479, "loss": 0.5261, "lr": 8.513513513513514e-06, "epoch": 0.25557809330628806, "percentage": 8.52, "elapsed_time": "0:27:57", "remaining_time": "5:00:15"}
+{"current_steps": 127, "total_steps": 1479, "loss": 0.5176, "lr": 8.581081081081082e-06, "epoch": 0.25760649087221094, "percentage": 8.59, "elapsed_time": "0:28:10", "remaining_time": "4:59:55"}
+{"current_steps": 128, "total_steps": 1479, "loss": 0.5656, "lr": 8.64864864864865e-06, "epoch": 0.25963488843813387, "percentage": 8.65, "elapsed_time": "0:28:23", "remaining_time": "4:59:35"}
+{"current_steps": 129, "total_steps": 1479, "loss": 0.5224, "lr": 8.716216216216217e-06, "epoch": 0.2616632860040568, "percentage": 8.72, "elapsed_time": "0:28:35", "remaining_time": "4:59:16"}
+{"current_steps": 130, "total_steps": 1479, "loss": 0.5245, "lr": 8.783783783783785e-06, "epoch": 0.26369168356997974, "percentage": 8.79, "elapsed_time": "0:28:48", "remaining_time": "4:58:57"}
+{"current_steps": 131, "total_steps": 1479, "loss": 0.5263, "lr": 8.851351351351351e-06, "epoch": 0.2657200811359026, "percentage": 8.86, "elapsed_time": "0:29:01", "remaining_time": "4:58:39"}
+{"current_steps": 132, "total_steps": 1479, "loss": 0.5376, "lr": 8.91891891891892e-06, "epoch": 0.26774847870182555, "percentage": 8.92, "elapsed_time": "0:29:14", "remaining_time": "4:58:20"}
+{"current_steps": 133, "total_steps": 1479, "loss": 0.5188, "lr": 8.986486486486488e-06, "epoch": 0.2697768762677485, "percentage": 8.99, "elapsed_time": "0:29:27", "remaining_time": "4:58:02"}
+{"current_steps": 134, "total_steps": 1479, "loss": 0.5068, "lr": 9.054054054054054e-06, "epoch": 0.2718052738336714, "percentage": 9.06, "elapsed_time": "0:29:39", "remaining_time": "4:57:44"}
+{"current_steps": 135, "total_steps": 1479, "loss": 0.5016, "lr": 9.121621621621622e-06, "epoch": 0.2738336713995943, "percentage": 9.13, "elapsed_time": "0:29:52", "remaining_time": "4:57:26"}
+{"current_steps": 136, "total_steps": 1479, "loss": 0.4988, "lr": 9.189189189189191e-06, "epoch": 0.27586206896551724, "percentage": 9.2, "elapsed_time": "0:30:05", "remaining_time": "4:57:07"}
+{"current_steps": 137, "total_steps": 1479, "loss": 0.5101, "lr": 9.256756756756757e-06, "epoch": 0.2778904665314402, "percentage": 9.26, "elapsed_time": "0:30:18", "remaining_time": "4:56:49"}
+{"current_steps": 138, "total_steps": 1479, "loss": 0.5054, "lr": 9.324324324324325e-06, "epoch": 0.2799188640973631, "percentage": 9.33, "elapsed_time": "0:30:30", "remaining_time": "4:56:31"}
+{"current_steps": 139, "total_steps": 1479, "loss": 0.5077, "lr": 9.391891891891893e-06, "epoch": 0.281947261663286, "percentage": 9.4, "elapsed_time": "0:30:43", "remaining_time": "4:56:12"}
+{"current_steps": 140, "total_steps": 1479, "loss": 0.539, "lr": 9.45945945945946e-06, "epoch": 0.2839756592292089, "percentage": 9.47, "elapsed_time": "0:30:56", "remaining_time": "4:55:54"}
+{"current_steps": 141, "total_steps": 1479, "loss": 0.5062, "lr": 9.527027027027028e-06, "epoch": 0.28600405679513186, "percentage": 9.53, "elapsed_time": "0:31:09", "remaining_time": "4:55:40"}
+{"current_steps": 142, "total_steps": 1479, "loss": 0.5066, "lr": 9.594594594594594e-06, "epoch": 0.2880324543610548, "percentage": 9.6, "elapsed_time": "0:31:22", "remaining_time": "4:55:26"}
+{"current_steps": 143, "total_steps": 1479, "loss": 0.5178, "lr": 9.662162162162164e-06, "epoch": 0.29006085192697767, "percentage": 9.67, "elapsed_time": "0:31:35", "remaining_time": "4:55:08"}
+{"current_steps": 144, "total_steps": 1479, "loss": 0.496, "lr": 9.729729729729732e-06, "epoch": 0.2920892494929006, "percentage": 9.74, "elapsed_time": "0:31:48", "remaining_time": "4:54:51"}
+{"current_steps": 145, "total_steps": 1479, "loss": 0.5159, "lr": 9.797297297297298e-06, "epoch": 0.29411764705882354, "percentage": 9.8, "elapsed_time": "0:32:01", "remaining_time": "4:54:33"}
+{"current_steps": 146, "total_steps": 1479, "loss": 0.5246, "lr": 9.864864864864865e-06, "epoch": 0.2961460446247465, "percentage": 9.87, "elapsed_time": "0:32:13", "remaining_time": "4:54:15"}
+{"current_steps": 147, "total_steps": 1479, "loss": 0.5226, "lr": 9.932432432432433e-06, "epoch": 0.29817444219066935, "percentage": 9.94, "elapsed_time": "0:32:26", "remaining_time": "4:53:58"}
+{"current_steps": 148, "total_steps": 1479, "loss": 0.5114, "lr": 1e-05, "epoch": 0.3002028397565923, "percentage": 10.01, "elapsed_time": "0:32:39", "remaining_time": "4:53:40"}
+{"current_steps": 149, "total_steps": 1479, "loss": 0.5, "lr": 9.999986072170506e-06, "epoch": 0.3022312373225152, "percentage": 10.07, "elapsed_time": "0:32:52", "remaining_time": "4:53:23"}
+{"current_steps": 150, "total_steps": 1479, "loss": 0.5184, "lr": 9.999944288759616e-06, "epoch": 0.30425963488843816, "percentage": 10.14, "elapsed_time": "0:33:04", "remaining_time": "4:53:06"}
+{"current_steps": 151, "total_steps": 1479, "loss": 0.5302, "lr": 9.99987465000011e-06, "epoch": 0.30628803245436104, "percentage": 10.21, "elapsed_time": "0:33:17", "remaining_time": "4:52:48"}
+{"current_steps": 152, "total_steps": 1479, "loss": 0.519, "lr": 9.999777156279958e-06, "epoch": 0.30831643002028397, "percentage": 10.28, "elapsed_time": "0:33:30", "remaining_time": "4:52:31"}
+{"current_steps": 153, "total_steps": 1479, "loss": 0.5234, "lr": 9.999651808142305e-06, "epoch": 0.3103448275862069, "percentage": 10.34, "elapsed_time": "0:33:43", "remaining_time": "4:52:14"}
+{"current_steps": 154, "total_steps": 1479, "loss": 0.5058, "lr": 9.999498606285486e-06, "epoch": 0.31237322515212984, "percentage": 10.41, "elapsed_time": "0:33:55", "remaining_time": "4:51:57"}
+{"current_steps": 155, "total_steps": 1479, "loss": 0.5082, "lr": 9.999317551563011e-06, "epoch": 0.3144016227180527, "percentage": 10.48, "elapsed_time": "0:34:08", "remaining_time": "4:51:39"}
+{"current_steps": 156, "total_steps": 1479, "loss": 0.5109, "lr": 9.999108644983555e-06, "epoch": 0.31643002028397565, "percentage": 10.55, "elapsed_time": "0:34:21", "remaining_time": "4:51:22"}
+{"current_steps": 157, "total_steps": 1479, "loss": 0.5157, "lr": 9.998871887710965e-06, "epoch": 0.3184584178498986, "percentage": 10.62, "elapsed_time": "0:34:34", "remaining_time": "4:51:05"}
+{"current_steps": 158, "total_steps": 1479, "loss": 0.5052, "lr": 9.998607281064248e-06, "epoch": 0.3204868154158215, "percentage": 10.68, "elapsed_time": "0:34:47", "remaining_time": "4:50:48"}
+{"current_steps": 159, "total_steps": 1479, "loss": 0.52, "lr": 9.998314826517564e-06, "epoch": 0.3225152129817444, "percentage": 10.75, "elapsed_time": "0:34:59", "remaining_time": "4:50:31"}
+{"current_steps": 160, "total_steps": 1479, "loss": 0.5042, "lr": 9.997994525700213e-06, "epoch": 0.32454361054766734, "percentage": 10.82, "elapsed_time": "0:35:12", "remaining_time": "4:50:14"}
+{"current_steps": 161, "total_steps": 1479, "loss": 0.5312, "lr": 9.997646380396633e-06, "epoch": 0.3265720081135903, "percentage": 10.89, "elapsed_time": "0:35:25", "remaining_time": "4:50:00"}
+{"current_steps": 162, "total_steps": 1479, "loss": 0.5177, "lr": 9.997270392546389e-06, "epoch": 0.3286004056795132, "percentage": 10.95, "elapsed_time": "0:35:38", "remaining_time": "4:49:44"}
+{"current_steps": 163, "total_steps": 1479, "loss": 0.5107, "lr": 9.996866564244158e-06, "epoch": 0.3306288032454361, "percentage": 11.02, "elapsed_time": "0:35:51", "remaining_time": "4:49:27"}
+{"current_steps": 164, "total_steps": 1479, "loss": 0.5153, "lr": 9.996434897739722e-06, "epoch": 0.332657200811359, "percentage": 11.09, "elapsed_time": "0:36:03", "remaining_time": "4:49:10"}
+{"current_steps": 165, "total_steps": 1479, "loss": 0.4971, "lr": 9.995975395437952e-06, "epoch": 0.33468559837728196, "percentage": 11.16, "elapsed_time": "0:36:16", "remaining_time": "4:48:53"}
+{"current_steps": 166, "total_steps": 1479, "loss": 0.5143, "lr": 9.995488059898792e-06, "epoch": 0.3367139959432049, "percentage": 11.22, "elapsed_time": "0:36:29", "remaining_time": "4:48:37"}
+{"current_steps": 167, "total_steps": 1479, "loss": 0.5152, "lr": 9.994972893837259e-06, "epoch": 0.33874239350912777, "percentage": 11.29, "elapsed_time": "0:36:42", "remaining_time": "4:48:20"}
+{"current_steps": 168, "total_steps": 1479, "loss": 0.5155, "lr": 9.994429900123405e-06, "epoch": 0.3407707910750507, "percentage": 11.36, "elapsed_time": "0:36:54", "remaining_time": "4:48:04"}
+{"current_steps": 169, "total_steps": 1479, "loss": 0.5149, "lr": 9.993859081782322e-06, "epoch": 0.34279918864097364, "percentage": 11.43, "elapsed_time": "0:37:07", "remaining_time": "4:47:47"}
+{"current_steps": 170, "total_steps": 1479, "loss": 0.4943, "lr": 9.993260441994116e-06, "epoch": 0.3448275862068966, "percentage": 11.49, "elapsed_time": "0:37:20", "remaining_time": "4:47:31"}
+{"current_steps": 171, "total_steps": 1479, "loss": 0.5113, "lr": 9.992633984093886e-06, "epoch": 0.34685598377281945, "percentage": 11.56, "elapsed_time": "0:37:33", "remaining_time": "4:47:15"}
+{"current_steps": 172, "total_steps": 1479, "loss": 0.5245, "lr": 9.99197971157171e-06, "epoch": 0.3488843813387424, "percentage": 11.63, "elapsed_time": "0:37:46", "remaining_time": "4:46:58"}
+{"current_steps": 173, "total_steps": 1479, "loss": 0.5017, "lr": 9.991297628072632e-06, "epoch": 0.3509127789046653, "percentage": 11.7, "elapsed_time": "0:37:58", "remaining_time": "4:46:42"}
+{"current_steps": 174, "total_steps": 1479, "loss": 0.5272, "lr": 9.990587737396623e-06, "epoch": 0.35294117647058826, "percentage": 11.76, "elapsed_time": "0:38:11", "remaining_time": "4:46:26"}
+{"current_steps": 175, "total_steps": 1479, "loss": 0.5001, "lr": 9.98985004349858e-06, "epoch": 0.35496957403651114, "percentage": 11.83, "elapsed_time": "0:38:24", "remaining_time": "4:46:10"}
+{"current_steps": 176, "total_steps": 1479, "loss": 0.506, "lr": 9.989084550488293e-06, "epoch": 0.35699797160243407, "percentage": 11.9, "elapsed_time": "0:38:37", "remaining_time": "4:45:53"}
+{"current_steps": 177, "total_steps": 1479, "loss": 0.5255, "lr": 9.988291262630425e-06, "epoch": 0.359026369168357, "percentage": 11.97, "elapsed_time": "0:38:49", "remaining_time": "4:45:37"}
+{"current_steps": 178, "total_steps": 1479, "loss": 0.5131, "lr": 9.987470184344485e-06, "epoch": 0.36105476673427994, "percentage": 12.04, "elapsed_time": "0:39:02", "remaining_time": "4:45:21"}
+{"current_steps": 179, "total_steps": 1479, "loss": 0.5141, "lr": 9.986621320204813e-06, "epoch": 0.3630831643002028, "percentage": 12.1, "elapsed_time": "0:39:15", "remaining_time": "4:45:05"}
+{"current_steps": 180, "total_steps": 1479, "loss": 0.5159, "lr": 9.985744674940537e-06, "epoch": 0.36511156186612576, "percentage": 12.17, "elapsed_time": "0:39:28", "remaining_time": "4:44:49"}
+{"current_steps": 181, "total_steps": 1479, "loss": 0.5193, "lr": 9.984840253435569e-06, "epoch": 0.3671399594320487, "percentage": 12.24, "elapsed_time": "0:39:41", "remaining_time": "4:44:36"}
+{"current_steps": 182, "total_steps": 1479, "loss": 0.4865, "lr": 9.983908060728554e-06, "epoch": 0.3691683569979716, "percentage": 12.31, "elapsed_time": "0:39:53", "remaining_time": "4:44:20"}
+{"current_steps": 183, "total_steps": 1479, "loss": 0.5124, "lr": 9.982948102012866e-06, "epoch": 0.3711967545638945, "percentage": 12.37, "elapsed_time": "0:40:06", "remaining_time": "4:44:04"}
+{"current_steps": 184, "total_steps": 1479, "loss": 0.4907, "lr": 9.98196038263656e-06, "epoch": 0.37322515212981744, "percentage": 12.44, "elapsed_time": "0:40:19", "remaining_time": "4:43:49"}
+{"current_steps": 185, "total_steps": 1479, "loss": 0.5197, "lr": 9.98094490810235e-06, "epoch": 0.3752535496957404, "percentage": 12.51, "elapsed_time": "0:40:32", "remaining_time": "4:43:36"}
+{"current_steps": 186, "total_steps": 1479, "loss": 0.5031, "lr": 9.97990168406758e-06, "epoch": 0.3772819472616633, "percentage": 12.58, "elapsed_time": "0:40:45", "remaining_time": "4:43:20"}
+{"current_steps": 187, "total_steps": 1479, "loss": 0.4991, "lr": 9.978830716344185e-06, "epoch": 0.3793103448275862, "percentage": 12.64, "elapsed_time": "0:40:59", "remaining_time": "4:43:10"}
+{"current_steps": 188, "total_steps": 1479, "loss": 0.5097, "lr": 9.977732010898672e-06, "epoch": 0.3813387423935091, "percentage": 12.71, "elapsed_time": "0:41:12", "remaining_time": "4:42:57"}
+{"current_steps": 189, "total_steps": 1479, "loss": 0.4913, "lr": 9.976605573852071e-06, "epoch": 0.38336713995943206, "percentage": 12.78, "elapsed_time": "0:41:26", "remaining_time": "4:42:53"}
+{"current_steps": 190, "total_steps": 1479, "loss": 0.4823, "lr": 9.975451411479912e-06, "epoch": 0.385395537525355, "percentage": 12.85, "elapsed_time": "0:41:39", "remaining_time": "4:42:40"}
+{"current_steps": 191, "total_steps": 1479, "loss": 0.4886, "lr": 9.974269530212185e-06, "epoch": 0.38742393509127787, "percentage": 12.91, "elapsed_time": "0:41:53", "remaining_time": "4:42:29"}
+{"current_steps": 192, "total_steps": 1479, "loss": 0.4874, "lr": 9.973059936633308e-06, "epoch": 0.3894523326572008, "percentage": 12.98, "elapsed_time": "0:42:06", "remaining_time": "4:42:13"}
+{"current_steps": 193, "total_steps": 1479, "loss": 0.4953, "lr": 9.971822637482085e-06, "epoch": 0.39148073022312374, "percentage": 13.05, "elapsed_time": "0:42:18", "remaining_time": "4:41:57"}
+{"current_steps": 194, "total_steps": 1479, "loss": 0.5135, "lr": 9.970557639651672e-06, "epoch": 0.3935091277890467, "percentage": 13.12, "elapsed_time": "0:42:31", "remaining_time": "4:41:42"}
+{"current_steps": 195, "total_steps": 1479, "loss": 0.5107, "lr": 9.969264950189539e-06, "epoch": 0.39553752535496955, "percentage": 13.18, "elapsed_time": "0:42:44", "remaining_time": "4:41:26"}
+{"current_steps": 196, "total_steps": 1479, "loss": 0.5033, "lr": 9.96794457629743e-06, "epoch": 0.3975659229208925, "percentage": 13.25, "elapsed_time": "0:42:57", "remaining_time": "4:41:10"}
+{"current_steps": 197, "total_steps": 1479, "loss": 0.4992, "lr": 9.966596525331324e-06, "epoch": 0.3995943204868154, "percentage": 13.32, "elapsed_time": "0:43:10", "remaining_time": "4:40:54"}
+{"current_steps": 198, "total_steps": 1479, "loss": 0.4956, "lr": 9.965220804801386e-06, "epoch": 0.40162271805273836, "percentage": 13.39, "elapsed_time": "0:43:22", "remaining_time": "4:40:38"}
+{"current_steps": 199, "total_steps": 1479, "loss": 0.4891, "lr": 9.96381742237194e-06, "epoch": 0.40365111561866124, "percentage": 13.46, "elapsed_time": "0:43:35", "remaining_time": "4:40:23"}
+{"current_steps": 200, "total_steps": 1479, "loss": 0.5074, "lr": 9.962386385861414e-06, "epoch": 0.4056795131845842, "percentage": 13.52, "elapsed_time": "0:43:48", "remaining_time": "4:40:08"}
+{"current_steps": 201, "total_steps": 1479, "loss": 0.5067, "lr": 9.960927703242298e-06, "epoch": 0.4077079107505071, "percentage": 13.59, "elapsed_time": "0:44:01", "remaining_time": "4:39:52"}
+{"current_steps": 202, "total_steps": 1479, "loss": 0.4794, "lr": 9.959441382641109e-06, "epoch": 0.40973630831643004, "percentage": 13.66, "elapsed_time": "0:44:13", "remaining_time": "4:39:37"}
+{"current_steps": 203, "total_steps": 1479, "loss": 0.5003, "lr": 9.957927432338332e-06, "epoch": 0.4117647058823529, "percentage": 13.73, "elapsed_time": "0:44:26", "remaining_time": "4:39:21"}
+{"current_steps": 204, "total_steps": 1479, "loss": 0.5173, "lr": 9.956385860768387e-06, "epoch": 0.41379310344827586, "percentage": 13.79, "elapsed_time": "0:44:39", "remaining_time": "4:39:06"}
+{"current_steps": 205, "total_steps": 1479, "loss": 0.5122, "lr": 9.954816676519569e-06, "epoch": 0.4158215010141988, "percentage": 13.86, "elapsed_time": "0:44:52", "remaining_time": "4:38:51"}
+{"current_steps": 206, "total_steps": 1479, "loss": 0.4771, "lr": 9.953219888334011e-06, "epoch": 0.4178498985801217, "percentage": 13.93, "elapsed_time": "0:45:04", "remaining_time": "4:38:35"}
+{"current_steps": 207, "total_steps": 1479, "loss": 0.4939, "lr": 9.951595505107633e-06, "epoch": 0.4198782961460446, "percentage": 14.0, "elapsed_time": "0:45:17", "remaining_time": "4:38:20"}
+{"current_steps": 208, "total_steps": 1479, "loss": 0.5004, "lr": 9.949943535890085e-06, "epoch": 0.42190669371196754, "percentage": 14.06, "elapsed_time": "0:45:30", "remaining_time": "4:38:05"}
+{"current_steps": 209, "total_steps": 1479, "loss": 0.492, "lr": 9.948263989884708e-06, "epoch": 0.4239350912778905, "percentage": 14.13, "elapsed_time": "0:45:43", "remaining_time": "4:37:49"}
+{"current_steps": 210, "total_steps": 1479, "loss": 0.4866, "lr": 9.94655687644847e-06, "epoch": 0.4259634888438134, "percentage": 14.2, "elapsed_time": "0:45:55", "remaining_time": "4:37:34"}
+{"current_steps": 211, "total_steps": 1479, "loss": 0.4921, "lr": 9.944822205091929e-06, "epoch": 0.4279918864097363, "percentage": 14.27, "elapsed_time": "0:46:08", "remaining_time": "4:37:18"}
+{"current_steps": 212, "total_steps": 1479, "loss": 0.4784, "lr": 9.943059985479167e-06, "epoch": 0.4300202839756592, "percentage": 14.33, "elapsed_time": "0:46:21", "remaining_time": "4:37:03"}
+{"current_steps": 213, "total_steps": 1479, "loss": 0.5168, "lr": 9.94127022742774e-06, "epoch": 0.43204868154158216, "percentage": 14.4, "elapsed_time": "0:46:34", "remaining_time": "4:36:48"}
+{"current_steps": 214, "total_steps": 1479, "loss": 0.508, "lr": 9.939452940908627e-06, "epoch": 0.4340770791075051, "percentage": 14.47, "elapsed_time": "0:46:47", "remaining_time": "4:36:33"}
+{"current_steps": 215, "total_steps": 1479, "loss": 0.5038, "lr": 9.937608136046171e-06, "epoch": 0.43610547667342797, "percentage": 14.54, "elapsed_time": "0:46:59", "remaining_time": "4:36:18"}
+{"current_steps": 216, "total_steps": 1479, "loss": 0.4996, "lr": 9.935735823118022e-06, "epoch": 0.4381338742393509, "percentage": 14.6, "elapsed_time": "0:47:12", "remaining_time": "4:36:03"}
+{"current_steps": 217, "total_steps": 1479, "loss": 0.4875, "lr": 9.933836012555083e-06, "epoch": 0.44016227180527384, "percentage": 14.67, "elapsed_time": "0:47:25", "remaining_time": "4:35:48"}
+{"current_steps": 218, "total_steps": 1479, "loss": 0.4889, "lr": 9.93190871494145e-06, "epoch": 0.4421906693711968, "percentage": 14.74, "elapsed_time": "0:47:38", "remaining_time": "4:35:32"}
+{"current_steps": 219, "total_steps": 1479, "loss": 0.4729, "lr": 9.929953941014349e-06, "epoch": 0.44421906693711966, "percentage": 14.81, "elapsed_time": "0:47:50", "remaining_time": "4:35:18"}
+{"current_steps": 220, "total_steps": 1479, "loss": 0.4997, "lr": 9.927971701664085e-06, "epoch": 0.4462474645030426, "percentage": 14.87, "elapsed_time": "0:48:03", "remaining_time": "4:35:03"}
+{"current_steps": 221, "total_steps": 1479, "loss": 0.4926, "lr": 9.925962007933975e-06, "epoch": 0.4482758620689655, "percentage": 14.94, "elapsed_time": "0:48:16", "remaining_time": "4:34:48"}
+{"current_steps": 222, "total_steps": 1479, "loss": 0.4977, "lr": 9.923924871020289e-06, "epoch": 0.45030425963488846, "percentage": 15.01, "elapsed_time": "0:48:29", "remaining_time": "4:34:32"}
+{"current_steps": 223, "total_steps": 1479, "loss": 0.4879, "lr": 9.921860302272184e-06, "epoch": 0.45233265720081134, "percentage": 15.08, "elapsed_time": "0:48:41", "remaining_time": "4:34:17"}
+{"current_steps": 224, "total_steps": 1479, "loss": 0.4786, "lr": 9.919768313191642e-06, "epoch": 0.4543610547667343, "percentage": 15.15, "elapsed_time": "0:48:54", "remaining_time": "4:34:02"}
+{"current_steps": 225, "total_steps": 1479, "loss": 0.4957, "lr": 9.917648915433413e-06, "epoch": 0.4563894523326572, "percentage": 15.21, "elapsed_time": "0:49:07", "remaining_time": "4:33:47"}
+{"current_steps": 226, "total_steps": 1479, "loss": 0.4829, "lr": 9.91550212080494e-06, "epoch": 0.45841784989858014, "percentage": 15.28, "elapsed_time": "0:49:20", "remaining_time": "4:33:32"}
+{"current_steps": 227, "total_steps": 1479, "loss": 0.496, "lr": 9.9133279412663e-06, "epoch": 0.460446247464503, "percentage": 15.35, "elapsed_time": "0:49:33", "remaining_time": "4:33:17"}
+{"current_steps": 228, "total_steps": 1479, "loss": 0.493, "lr": 9.911126388930133e-06, "epoch": 0.46247464503042596, "percentage": 15.42, "elapsed_time": "0:49:45", "remaining_time": "4:33:03"}
+{"current_steps": 229, "total_steps": 1479, "loss": 0.4853, "lr": 9.908897476061576e-06, "epoch": 0.4645030425963489, "percentage": 15.48, "elapsed_time": "0:49:58", "remaining_time": "4:32:48"}
+{"current_steps": 230, "total_steps": 1479, "loss": 0.4668, "lr": 9.906641215078196e-06, "epoch": 0.4665314401622718, "percentage": 15.55, "elapsed_time": "0:50:11", "remaining_time": "4:32:33"}
+{"current_steps": 231, "total_steps": 1479, "loss": 0.4916, "lr": 9.904357618549925e-06, "epoch": 0.4685598377281947, "percentage": 15.62, "elapsed_time": "0:50:24", "remaining_time": "4:32:18"}
+{"current_steps": 232, "total_steps": 1479, "loss": 0.4764, "lr": 9.902046699198976e-06, "epoch": 0.47058823529411764, "percentage": 15.69, "elapsed_time": "0:50:36", "remaining_time": "4:32:03"}
+{"current_steps": 233, "total_steps": 1479, "loss": 0.4977, "lr": 9.899708469899786e-06, "epoch": 0.4726166328600406, "percentage": 15.75, "elapsed_time": "0:50:49", "remaining_time": "4:31:48"}
+{"current_steps": 234, "total_steps": 1479, "loss": 0.5159, "lr": 9.897342943678938e-06, "epoch": 0.4746450304259635, "percentage": 15.82, "elapsed_time": "0:51:02", "remaining_time": "4:31:33"}
+{"current_steps": 235, "total_steps": 1479, "loss": 0.4866, "lr": 9.894950133715094e-06, "epoch": 0.4766734279918864, "percentage": 15.89, "elapsed_time": "0:51:15", "remaining_time": "4:31:19"}
+{"current_steps": 236, "total_steps": 1479, "loss": 0.4925, "lr": 9.892530053338909e-06, "epoch": 0.4787018255578093, "percentage": 15.96, "elapsed_time": "0:51:27", "remaining_time": "4:31:04"}
+{"current_steps": 237, "total_steps": 1479, "loss": 0.5022, "lr": 9.89008271603297e-06, "epoch": 0.48073022312373226, "percentage": 16.02, "elapsed_time": "0:51:40", "remaining_time": "4:30:49"}
+{"current_steps": 238, "total_steps": 1479, "loss": 0.5038, "lr": 9.88760813543172e-06, "epoch": 0.4827586206896552, "percentage": 16.09, "elapsed_time": "0:51:53", "remaining_time": "4:30:34"}
+{"current_steps": 239, "total_steps": 1479, "loss": 0.4939, "lr": 9.885106325321371e-06, "epoch": 0.4847870182555781, "percentage": 16.16, "elapsed_time": "0:52:06", "remaining_time": "4:30:19"}
+{"current_steps": 240, "total_steps": 1479, "loss": 0.4815, "lr": 9.882577299639836e-06, "epoch": 0.486815415821501, "percentage": 16.23, "elapsed_time": "0:52:18", "remaining_time": "4:30:04"}
+{"current_steps": 241, "total_steps": 1479, "loss": 0.5005, "lr": 9.880021072476651e-06, "epoch": 0.48884381338742394, "percentage": 16.29, "elapsed_time": "0:52:31", "remaining_time": "4:29:49"}
+{"current_steps": 242, "total_steps": 1479, "loss": 0.5105, "lr": 9.877437658072894e-06, "epoch": 0.4908722109533469, "percentage": 16.36, "elapsed_time": "0:52:44", "remaining_time": "4:29:35"}
+{"current_steps": 243, "total_steps": 1479, "loss": 0.4787, "lr": 9.874827070821112e-06, "epoch": 0.49290060851926976, "percentage": 16.43, "elapsed_time": "0:52:57", "remaining_time": "4:29:22"}
+{"current_steps": 244, "total_steps": 1479, "loss": 0.4877, "lr": 9.872189325265222e-06, "epoch": 0.4949290060851927, "percentage": 16.5, "elapsed_time": "0:53:10", "remaining_time": "4:29:07"}
+{"current_steps": 245, "total_steps": 1479, "loss": 0.4853, "lr": 9.869524436100458e-06, "epoch": 0.4969574036511156, "percentage": 16.57, "elapsed_time": "0:53:23", "remaining_time": "4:28:53"}
+{"current_steps": 246, "total_steps": 1479, "loss": 0.4634, "lr": 9.866832418173267e-06, "epoch": 0.49898580121703856, "percentage": 16.63, "elapsed_time": "0:53:35", "remaining_time": "4:28:38"}
+{"current_steps": 247, "total_steps": 1479, "loss": 0.4925, "lr": 9.864113286481237e-06, "epoch": 0.5010141987829615, "percentage": 16.7, "elapsed_time": "0:53:48", "remaining_time": "4:28:23"}
+{"current_steps": 248, "total_steps": 1479, "loss": 0.5049, "lr": 9.861367056173007e-06, "epoch": 0.5030425963488844, "percentage": 16.77, "elapsed_time": "0:54:01", "remaining_time": "4:28:09"}
+{"current_steps": 249, "total_steps": 1479, "loss": 0.5003, "lr": 9.85859374254819e-06, "epoch": 0.5050709939148073, "percentage": 16.84, "elapsed_time": "0:54:14", "remaining_time": "4:27:54"}
+{"current_steps": 250, "total_steps": 1479, "loss": 0.4764, "lr": 9.855793361057283e-06, "epoch": 0.5070993914807302, "percentage": 16.9, "elapsed_time": "0:54:26", "remaining_time": "4:27:39"}
+{"current_steps": 251, "total_steps": 1479, "loss": 0.4895, "lr": 9.852965927301573e-06, "epoch": 0.5091277890466531, "percentage": 16.97, "elapsed_time": "0:54:39", "remaining_time": "4:27:25"}
+{"current_steps": 252, "total_steps": 1479, "loss": 0.4936, "lr": 9.850111457033075e-06, "epoch": 0.5111561866125761, "percentage": 17.04, "elapsed_time": "0:54:52", "remaining_time": "4:27:10"}
+{"current_steps": 253, "total_steps": 1479, "loss": 0.4947, "lr": 9.847229966154415e-06, "epoch": 0.513184584178499, "percentage": 17.11, "elapsed_time": "0:55:05", "remaining_time": "4:26:56"}
+{"current_steps": 254, "total_steps": 1479, "loss": 0.4834, "lr": 9.84432147071876e-06, "epoch": 0.5152129817444219, "percentage": 17.17, "elapsed_time": "0:55:18", "remaining_time": "4:26:42"}
+{"current_steps": 255, "total_steps": 1479, "loss": 0.481, "lr": 9.841385986929716e-06, "epoch": 0.5172413793103449, "percentage": 17.24, "elapsed_time": "0:55:30", "remaining_time": "4:26:27"}
+{"current_steps": 256, "total_steps": 1479, "loss": 0.5018, "lr": 9.838423531141258e-06, "epoch": 0.5192697768762677, "percentage": 17.31, "elapsed_time": "0:55:43", "remaining_time": "4:26:12"}
+{"current_steps": 257, "total_steps": 1479, "loss": 0.4964, "lr": 9.835434119857612e-06, "epoch": 0.5212981744421906, "percentage": 17.38, "elapsed_time": "0:55:56", "remaining_time": "4:25:58"}
+{"current_steps": 258, "total_steps": 1479, "loss": 0.4742, "lr": 9.832417769733185e-06, "epoch": 0.5233265720081136, "percentage": 17.44, "elapsed_time": "0:56:08", "remaining_time": "4:25:43"}
+{"current_steps": 259, "total_steps": 1479, "loss": 0.5015, "lr": 9.829374497572461e-06, "epoch": 0.5253549695740365, "percentage": 17.51, "elapsed_time": "0:56:22", "remaining_time": "4:25:31"}
+{"current_steps": 260, "total_steps": 1479, "loss": 0.488, "lr": 9.826304320329909e-06, "epoch": 0.5273833671399595, "percentage": 17.58, "elapsed_time": "0:56:34", "remaining_time": "4:25:16"}
+{"current_steps": 261, "total_steps": 1479, "loss": 0.4929, "lr": 9.823207255109891e-06, "epoch": 0.5294117647058824, "percentage": 17.65, "elapsed_time": "0:56:47", "remaining_time": "4:25:02"}
+{"current_steps": 262, "total_steps": 1479, "loss": 0.4735, "lr": 9.820083319166568e-06, "epoch": 0.5314401622718052, "percentage": 17.71, "elapsed_time": "0:57:00", "remaining_time": "4:24:47"}
+{"current_steps": 263, "total_steps": 1479, "loss": 0.4668, "lr": 9.816932529903795e-06, "epoch": 0.5334685598377282, "percentage": 17.78, "elapsed_time": "0:57:13", "remaining_time": "4:24:33"}
+{"current_steps": 264, "total_steps": 1479, "loss": 0.4939, "lr": 9.813754904875038e-06, "epoch": 0.5354969574036511, "percentage": 17.85, "elapsed_time": "0:57:25", "remaining_time": "4:24:18"}
+{"current_steps": 265, "total_steps": 1479, "loss": 0.502, "lr": 9.810550461783261e-06, "epoch": 0.537525354969574, "percentage": 17.92, "elapsed_time": "0:57:38", "remaining_time": "4:24:04"}
+{"current_steps": 266, "total_steps": 1479, "loss": 0.4878, "lr": 9.807319218480845e-06, "epoch": 0.539553752535497, "percentage": 17.99, "elapsed_time": "0:57:51", "remaining_time": "4:23:50"}
+{"current_steps": 267, "total_steps": 1479, "loss": 0.4705, "lr": 9.804061192969465e-06, "epoch": 0.5415821501014199, "percentage": 18.05, "elapsed_time": "0:58:04", "remaining_time": "4:23:35"}
+{"current_steps": 268, "total_steps": 1479, "loss": 0.4759, "lr": 9.800776403400014e-06, "epoch": 0.5436105476673428, "percentage": 18.12, "elapsed_time": "0:58:16", "remaining_time": "4:23:21"}
+{"current_steps": 269, "total_steps": 1479, "loss": 0.5036, "lr": 9.797464868072489e-06, "epoch": 0.5456389452332657, "percentage": 18.19, "elapsed_time": "0:58:29", "remaining_time": "4:23:06"}
+{"current_steps": 270, "total_steps": 1479, "loss": 0.4892, "lr": 9.794126605435885e-06, "epoch": 0.5476673427991886, "percentage": 18.26, "elapsed_time": "0:58:42", "remaining_time": "4:22:52"}
+{"current_steps": 271, "total_steps": 1479, "loss": 0.4713, "lr": 9.790761634088108e-06, "epoch": 0.5496957403651116, "percentage": 18.32, "elapsed_time": "0:58:55", "remaining_time": "4:22:38"}
+{"current_steps": 272, "total_steps": 1479, "loss": 0.4901, "lr": 9.787369972775858e-06, "epoch": 0.5517241379310345, "percentage": 18.39, "elapsed_time": "0:59:07", "remaining_time": "4:22:23"}
+{"current_steps": 273, "total_steps": 1479, "loss": 0.487, "lr": 9.78395164039452e-06, "epoch": 0.5537525354969574, "percentage": 18.46, "elapsed_time": "0:59:20", "remaining_time": "4:22:09"}
+{"current_steps": 274, "total_steps": 1479, "loss": 0.4594, "lr": 9.780506655988082e-06, "epoch": 0.5557809330628803, "percentage": 18.53, "elapsed_time": "0:59:33", "remaining_time": "4:21:55"}
+{"current_steps": 275, "total_steps": 1479, "loss": 0.4867, "lr": 9.777035038749002e-06, "epoch": 0.5578093306288032, "percentage": 18.59, "elapsed_time": "0:59:46", "remaining_time": "4:21:40"}
+{"current_steps": 276, "total_steps": 1479, "loss": 0.4786, "lr": 9.773536808018116e-06, "epoch": 0.5598377281947262, "percentage": 18.66, "elapsed_time": "0:59:58", "remaining_time": "4:21:26"}
+{"current_steps": 277, "total_steps": 1479, "loss": 0.4805, "lr": 9.77001198328453e-06, "epoch": 0.5618661257606491, "percentage": 18.73, "elapsed_time": "1:00:11", "remaining_time": "4:21:12"}
+{"current_steps": 278, "total_steps": 1479, "loss": 0.4983, "lr": 9.766460584185511e-06, "epoch": 0.563894523326572, "percentage": 18.8, "elapsed_time": "1:00:24", "remaining_time": "4:20:58"}
+{"current_steps": 279, "total_steps": 1479, "loss": 0.4771, "lr": 9.762882630506366e-06, "epoch": 0.565922920892495, "percentage": 18.86, "elapsed_time": "1:00:37", "remaining_time": "4:20:43"}
+{"current_steps": 280, "total_steps": 1479, "loss": 0.4883, "lr": 9.759278142180348e-06, "epoch": 0.5679513184584178, "percentage": 18.93, "elapsed_time": "1:00:49", "remaining_time": "4:20:29"}
+{"current_steps": 281, "total_steps": 1479, "loss": 0.4722, "lr": 9.75564713928854e-06, "epoch": 0.5699797160243407, "percentage": 19.0, "elapsed_time": "1:01:02", "remaining_time": "4:20:15"}
+{"current_steps": 282, "total_steps": 1479, "loss": 0.4653, "lr": 9.751989642059731e-06, "epoch": 0.5720081135902637, "percentage": 19.07, "elapsed_time": "1:01:15", "remaining_time": "4:20:00"}
+{"current_steps": 283, "total_steps": 1479, "loss": 0.4915, "lr": 9.748305670870326e-06, "epoch": 0.5740365111561866, "percentage": 19.13, "elapsed_time": "1:01:28", "remaining_time": "4:19:46"}
+{"current_steps": 284, "total_steps": 1479, "loss": 0.4871, "lr": 9.74459524624421e-06, "epoch": 0.5760649087221096, "percentage": 19.2, "elapsed_time": "1:01:41", "remaining_time": "4:19:34"}
+{"current_steps": 285, "total_steps": 1479, "loss": 0.4721, "lr": 9.740858388852652e-06, "epoch": 0.5780933062880325, "percentage": 19.27, "elapsed_time": "1:01:54", "remaining_time": "4:19:19"}
+{"current_steps": 286, "total_steps": 1479, "loss": 0.5015, "lr": 9.737095119514172e-06, "epoch": 0.5801217038539553, "percentage": 19.34, "elapsed_time": "1:02:06", "remaining_time": "4:19:05"}
+{"current_steps": 287, "total_steps": 1479, "loss": 0.4956, "lr": 9.733305459194444e-06, "epoch": 0.5821501014198783, "percentage": 19.41, "elapsed_time": "1:02:19", "remaining_time": "4:18:51"}
+{"current_steps": 288, "total_steps": 1479, "loss": 0.4946, "lr": 9.729489429006162e-06, "epoch": 0.5841784989858012, "percentage": 19.47, "elapsed_time": "1:02:32", "remaining_time": "4:18:37"}
+{"current_steps": 289, "total_steps": 1479, "loss": 0.4914, "lr": 9.725647050208936e-06, "epoch": 0.5862068965517241, "percentage": 19.54, "elapsed_time": "1:02:45", "remaining_time": "4:18:23"}
+{"current_steps": 290, "total_steps": 1479, "loss": 0.4953, "lr": 9.72177834420916e-06, "epoch": 0.5882352941176471, "percentage": 19.61, "elapsed_time": "1:02:57", "remaining_time": "4:18:09"}
+{"current_steps": 291, "total_steps": 1479, "loss": 0.4721, "lr": 9.717883332559911e-06, "epoch": 0.59026369168357, "percentage": 19.68, "elapsed_time": "1:03:10", "remaining_time": "4:17:55"}
+{"current_steps": 292, "total_steps": 1479, "loss": 0.4676, "lr": 9.713962036960806e-06, "epoch": 0.592292089249493, "percentage": 19.74, "elapsed_time": "1:03:23", "remaining_time": "4:17:40"}
+{"current_steps": 293, "total_steps": 1479, "loss": 0.4796, "lr": 9.710014479257906e-06, "epoch": 0.5943204868154158, "percentage": 19.81, "elapsed_time": "1:03:36", "remaining_time": "4:17:26"}
+{"current_steps": 294, "total_steps": 1479, "loss": 0.4822, "lr": 9.70604068144357e-06, "epoch": 0.5963488843813387, "percentage": 19.88, "elapsed_time": "1:03:48", "remaining_time": "4:17:12"}
+{"current_steps": 295, "total_steps": 1479, "loss": 0.4847, "lr": 9.702040665656353e-06, "epoch": 0.5983772819472617, "percentage": 19.95, "elapsed_time": "1:04:01", "remaining_time": "4:16:58"}
+{"current_steps": 296, "total_steps": 1479, "loss": 0.4842, "lr": 9.698014454180868e-06, "epoch": 0.6004056795131846, "percentage": 20.01, "elapsed_time": "1:04:14", "remaining_time": "4:16:44"}
+{"current_steps": 297, "total_steps": 1479, "loss": 0.4778, "lr": 9.693962069447669e-06, "epoch": 0.6024340770791075, "percentage": 20.08, "elapsed_time": "1:04:27", "remaining_time": "4:16:30"}
+{"current_steps": 298, "total_steps": 1479, "loss": 0.4835, "lr": 9.689883534033128e-06, "epoch": 0.6044624746450304, "percentage": 20.15, "elapsed_time": "1:04:39", "remaining_time": "4:16:16"}
+{"current_steps": 299, "total_steps": 1479, "loss": 0.4889, "lr": 9.685778870659301e-06, "epoch": 0.6064908722109533, "percentage": 20.22, "elapsed_time": "1:04:52", "remaining_time": "4:16:02"}
+{"current_steps": 300, "total_steps": 1479, "loss": 0.4747, "lr": 9.681648102193812e-06, "epoch": 0.6085192697768763, "percentage": 20.28, "elapsed_time": "1:05:05", "remaining_time": "4:15:48"}
+{"current_steps": 301, "total_steps": 1479, "loss": 0.4936, "lr": 9.677491251649711e-06, "epoch": 0.6105476673427992, "percentage": 20.35, "elapsed_time": "1:05:18", "remaining_time": "4:15:34"}
+{"current_steps": 302, "total_steps": 1479, "loss": 0.4825, "lr": 9.673308342185366e-06, "epoch": 0.6125760649087221, "percentage": 20.42, "elapsed_time": "1:05:30", "remaining_time": "4:15:20"}
+{"current_steps": 303, "total_steps": 1479, "loss": 0.4773, "lr": 9.669099397104314e-06, "epoch": 0.6146044624746451, "percentage": 20.49, "elapsed_time": "1:05:43", "remaining_time": "4:15:06"}
+{"current_steps": 304, "total_steps": 1479, "loss": 0.4839, "lr": 9.664864439855144e-06, "epoch": 0.6166328600405679, "percentage": 20.55, "elapsed_time": "1:05:56", "remaining_time": "4:14:52"}
+{"current_steps": 305, "total_steps": 1479, "loss": 0.4758, "lr": 9.660603494031358e-06, "epoch": 0.6186612576064908, "percentage": 20.62, "elapsed_time": "1:06:09", "remaining_time": "4:14:38"}
+{"current_steps": 306, "total_steps": 1479, "loss": 0.4826, "lr": 9.65631658337125e-06, "epoch": 0.6206896551724138, "percentage": 20.69, "elapsed_time": "1:06:22", "remaining_time": "4:14:24"}
+{"current_steps": 307, "total_steps": 1479, "loss": 0.4966, "lr": 9.652003731757763e-06, "epoch": 0.6227180527383367, "percentage": 20.76, "elapsed_time": "1:06:34", "remaining_time": "4:14:10"}
+{"current_steps": 308, "total_steps": 1479, "loss": 0.4766, "lr": 9.647664963218363e-06, "epoch": 0.6247464503042597, "percentage": 20.82, "elapsed_time": "1:06:47", "remaining_time": "4:13:56"}
+{"current_steps": 309, "total_steps": 1479, "loss": 0.4714, "lr": 9.643300301924902e-06, "epoch": 0.6267748478701826, "percentage": 20.89, "elapsed_time": "1:07:00", "remaining_time": "4:13:42"}
+{"current_steps": 310, "total_steps": 1479, "loss": 0.4791, "lr": 9.638909772193478e-06, "epoch": 0.6288032454361054, "percentage": 20.96, "elapsed_time": "1:07:12", "remaining_time": "4:13:28"}
+{"current_steps": 311, "total_steps": 1479, "loss": 0.4688, "lr": 9.634493398484319e-06, "epoch": 0.6308316430020284, "percentage": 21.03, "elapsed_time": "1:07:25", "remaining_time": "4:13:14"}
+{"current_steps": 312, "total_steps": 1479, "loss": 0.4554, "lr": 9.630051205401618e-06, "epoch": 0.6328600405679513, "percentage": 21.1, "elapsed_time": "1:07:38", "remaining_time": "4:13:00"}
+{"current_steps": 313, "total_steps": 1479, "loss": 0.474, "lr": 9.625583217693419e-06, "epoch": 0.6348884381338742, "percentage": 21.16, "elapsed_time": "1:07:51", "remaining_time": "4:12:46"}
+{"current_steps": 314, "total_steps": 1479, "loss": 0.4865, "lr": 9.621089460251474e-06, "epoch": 0.6369168356997972, "percentage": 21.23, "elapsed_time": "1:08:03", "remaining_time": "4:12:31"}
+{"current_steps": 315, "total_steps": 1479, "loss": 0.4635, "lr": 9.616569958111097e-06, "epoch": 0.6389452332657201, "percentage": 21.3, "elapsed_time": "1:08:16", "remaining_time": "4:12:18"}
+{"current_steps": 316, "total_steps": 1479, "loss": 0.4772, "lr": 9.612024736451027e-06, "epoch": 0.640973630831643, "percentage": 21.37, "elapsed_time": "1:08:29", "remaining_time": "4:12:03"}
+{"current_steps": 317, "total_steps": 1479, "loss": 0.4866, "lr": 9.607453820593297e-06, "epoch": 0.6430020283975659, "percentage": 21.43, "elapsed_time": "1:08:42", "remaining_time": "4:11:50"}
+{"current_steps": 318, "total_steps": 1479, "loss": 0.4805, "lr": 9.602857236003078e-06, "epoch": 0.6450304259634888, "percentage": 21.5, "elapsed_time": "1:08:54", "remaining_time": "4:11:36"}
+{"current_steps": 319, "total_steps": 1479, "loss": 0.4603, "lr": 9.598235008288551e-06, "epoch": 0.6470588235294118, "percentage": 21.57, "elapsed_time": "1:09:07", "remaining_time": "4:11:22"}
+{"current_steps": 320, "total_steps": 1479, "loss": 0.4871, "lr": 9.593587163200755e-06, "epoch": 0.6490872210953347, "percentage": 21.64, "elapsed_time": "1:09:20", "remaining_time": "4:11:08"}
+{"current_steps": 321, "total_steps": 1479, "loss": 0.5087, "lr": 9.58891372663345e-06, "epoch": 0.6511156186612576, "percentage": 21.7, "elapsed_time": "1:09:33", "remaining_time": "4:10:54"}
+{"current_steps": 322, "total_steps": 1479, "loss": 0.4796, "lr": 9.584214724622962e-06, "epoch": 0.6531440162271805, "percentage": 21.77, "elapsed_time": "1:09:45", "remaining_time": "4:10:40"}
+{"current_steps": 323, "total_steps": 1479, "loss": 0.4884, "lr": 9.579490183348052e-06, "epoch": 0.6551724137931034, "percentage": 21.84, "elapsed_time": "1:09:58", "remaining_time": "4:10:26"}
+{"current_steps": 324, "total_steps": 1479, "loss": 0.4904, "lr": 9.574740129129767e-06, "epoch": 0.6572008113590264, "percentage": 21.91, "elapsed_time": "1:10:11", "remaining_time": "4:10:12"}
+{"current_steps": 325, "total_steps": 1479, "loss": 0.4847, "lr": 9.56996458843128e-06, "epoch": 0.6592292089249493, "percentage": 21.97, "elapsed_time": "1:10:24", "remaining_time": "4:09:58"}
+{"current_steps": 326, "total_steps": 1479, "loss": 0.4728, "lr": 9.565163587857757e-06, "epoch": 0.6612576064908722, "percentage": 22.04, "elapsed_time": "1:10:36", "remaining_time": "4:09:45"}
+{"current_steps": 327, "total_steps": 1479, "loss": 0.4576, "lr": 9.56033715415621e-06, "epoch": 0.6632860040567952, "percentage": 22.11, "elapsed_time": "1:10:49", "remaining_time": "4:09:31"}
+{"current_steps": 328, "total_steps": 1479, "loss": 0.4831, "lr": 9.555485314215333e-06, "epoch": 0.665314401622718, "percentage": 22.18, "elapsed_time": "1:11:02", "remaining_time": "4:09:17"}
+{"current_steps": 329, "total_steps": 1479, "loss": 0.4948, "lr": 9.550608095065367e-06, "epoch": 0.6673427991886409, "percentage": 22.24, "elapsed_time": "1:11:15", "remaining_time": "4:09:03"}
+{"current_steps": 330, "total_steps": 1479, "loss": 0.4801, "lr": 9.545705523877944e-06, "epoch": 0.6693711967545639, "percentage": 22.31, "elapsed_time": "1:11:27", "remaining_time": "4:08:49"}
+{"current_steps": 331, "total_steps": 1479, "loss": 0.4838, "lr": 9.540777627965933e-06, "epoch": 0.6713995943204868, "percentage": 22.38, "elapsed_time": "1:11:40", "remaining_time": "4:08:35"}
+{"current_steps": 332, "total_steps": 1479, "loss": 0.4837, "lr": 9.535824434783291e-06, "epoch": 0.6734279918864098, "percentage": 22.45, "elapsed_time": "1:11:53", "remaining_time": "4:08:22"}
+{"current_steps": 333, "total_steps": 1479, "loss": 0.4925, "lr": 9.53084597192491e-06, "epoch": 0.6754563894523327, "percentage": 22.52, "elapsed_time": "1:12:06", "remaining_time": "4:08:08"}
+{"current_steps": 334, "total_steps": 1479, "loss": 0.4676, "lr": 9.525842267126464e-06, "epoch": 0.6774847870182555, "percentage": 22.58, "elapsed_time": "1:12:18", "remaining_time": "4:07:54"}
+{"current_steps": 335, "total_steps": 1479, "loss": 0.4947, "lr": 9.520813348264252e-06, "epoch": 0.6795131845841785, "percentage": 22.65, "elapsed_time": "1:12:31", "remaining_time": "4:07:41"}
+{"current_steps": 336, "total_steps": 1479, "loss": 0.4873, "lr": 9.515759243355043e-06, "epoch": 0.6815415821501014, "percentage": 22.72, "elapsed_time": "1:12:44", "remaining_time": "4:07:27"}
+{"current_steps": 337, "total_steps": 1479, "loss": 0.4739, "lr": 9.510679980555922e-06, "epoch": 0.6835699797160243, "percentage": 22.79, "elapsed_time": "1:12:57", "remaining_time": "4:07:13"}
+{"current_steps": 338, "total_steps": 1479, "loss": 0.466, "lr": 9.505575588164128e-06, "epoch": 0.6855983772819473, "percentage": 22.85, "elapsed_time": "1:13:09", "remaining_time": "4:06:59"}
+{"current_steps": 339, "total_steps": 1479, "loss": 0.4921, "lr": 9.500446094616911e-06, "epoch": 0.6876267748478702, "percentage": 22.92, "elapsed_time": "1:13:22", "remaining_time": "4:06:45"}
+{"current_steps": 340, "total_steps": 1479, "loss": 0.4903, "lr": 9.49529152849135e-06, "epoch": 0.6896551724137931, "percentage": 22.99, "elapsed_time": "1:13:35", "remaining_time": "4:06:32"}
+{"current_steps": 341, "total_steps": 1479, "loss": 0.4933, "lr": 9.490111918504213e-06, "epoch": 0.691683569979716, "percentage": 23.06, "elapsed_time": "1:13:48", "remaining_time": "4:06:18"}
+{"current_steps": 342, "total_steps": 1479, "loss": 0.4766, "lr": 9.484907293511792e-06, "epoch": 0.6937119675456389, "percentage": 23.12, "elapsed_time": "1:14:01", "remaining_time": "4:06:04"}
+{"current_steps": 343, "total_steps": 1479, "loss": 0.4741, "lr": 9.479677682509737e-06, "epoch": 0.6957403651115619, "percentage": 23.19, "elapsed_time": "1:14:13", "remaining_time": "4:05:50"}
+{"current_steps": 344, "total_steps": 1479, "loss": 0.4838, "lr": 9.4744231146329e-06, "epoch": 0.6977687626774848, "percentage": 23.26, "elapsed_time": "1:14:26", "remaining_time": "4:05:36"}
+{"current_steps": 345, "total_steps": 1479, "loss": 0.4499, "lr": 9.469143619155172e-06, "epoch": 0.6997971602434077, "percentage": 23.33, "elapsed_time": "1:14:39", "remaining_time": "4:05:22"}
+{"current_steps": 346, "total_steps": 1479, "loss": 0.4791, "lr": 9.46383922548932e-06, "epoch": 0.7018255578093306, "percentage": 23.39, "elapsed_time": "1:14:51", "remaining_time": "4:05:08"}
+{"current_steps": 347, "total_steps": 1479, "loss": 0.4891, "lr": 9.458509963186815e-06, "epoch": 0.7038539553752535, "percentage": 23.46, "elapsed_time": "1:15:04", "remaining_time": "4:04:55"}
+{"current_steps": 348, "total_steps": 1479, "loss": 0.4909, "lr": 9.453155861937684e-06, "epoch": 0.7058823529411765, "percentage": 23.53, "elapsed_time": "1:15:17", "remaining_time": "4:04:41"}
+{"current_steps": 349, "total_steps": 1479, "loss": 0.4772, "lr": 9.44777695157033e-06, "epoch": 0.7079107505070994, "percentage": 23.6, "elapsed_time": "1:15:30", "remaining_time": "4:04:27"}
+{"current_steps": 350, "total_steps": 1479, "loss": 0.4697, "lr": 9.442373262051373e-06, "epoch": 0.7099391480730223, "percentage": 23.66, "elapsed_time": "1:15:42", "remaining_time": "4:04:14"}
+{"current_steps": 351, "total_steps": 1479, "loss": 0.4869, "lr": 9.436944823485475e-06, "epoch": 0.7119675456389453, "percentage": 23.73, "elapsed_time": "1:15:55", "remaining_time": "4:04:00"}
+{"current_steps": 352, "total_steps": 1479, "loss": 0.4872, "lr": 9.431491666115187e-06, "epoch": 0.7139959432048681, "percentage": 23.8, "elapsed_time": "1:16:08", "remaining_time": "4:03:46"}
+{"current_steps": 353, "total_steps": 1479, "loss": 0.4811, "lr": 9.426013820320764e-06, "epoch": 0.716024340770791, "percentage": 23.87, "elapsed_time": "1:16:21", "remaining_time": "4:03:33"}
+{"current_steps": 354, "total_steps": 1479, "loss": 0.4704, "lr": 9.420511316620008e-06, "epoch": 0.718052738336714, "percentage": 23.94, "elapsed_time": "1:16:33", "remaining_time": "4:03:19"}
+{"current_steps": 355, "total_steps": 1479, "loss": 0.4979, "lr": 9.414984185668097e-06, "epoch": 0.7200811359026369, "percentage": 24.0, "elapsed_time": "1:16:46", "remaining_time": "4:03:05"}
+{"current_steps": 356, "total_steps": 1479, "loss": 0.4716, "lr": 9.409432458257399e-06, "epoch": 0.7221095334685599, "percentage": 24.07, "elapsed_time": "1:16:59", "remaining_time": "4:02:51"}
+{"current_steps": 357, "total_steps": 1479, "loss": 0.4744, "lr": 9.403856165317322e-06, "epoch": 0.7241379310344828, "percentage": 24.14, "elapsed_time": "1:17:12", "remaining_time": "4:02:38"}
+{"current_steps": 358, "total_steps": 1479, "loss": 0.5087, "lr": 9.398255337914131e-06, "epoch": 0.7261663286004056, "percentage": 24.21, "elapsed_time": "1:17:24", "remaining_time": "4:02:24"}
+{"current_steps": 359, "total_steps": 1479, "loss": 0.4815, "lr": 9.392630007250769e-06, "epoch": 0.7281947261663286, "percentage": 24.27, "elapsed_time": "1:17:37", "remaining_time": "4:02:10"}
+{"current_steps": 360, "total_steps": 1479, "loss": 0.5075, "lr": 9.3869802046667e-06, "epoch": 0.7302231237322515, "percentage": 24.34, "elapsed_time": "1:17:50", "remaining_time": "4:01:57"}
+{"current_steps": 361, "total_steps": 1479, "loss": 0.4806, "lr": 9.381305961637713e-06, "epoch": 0.7322515212981744, "percentage": 24.41, "elapsed_time": "1:18:03", "remaining_time": "4:01:43"}
+{"current_steps": 362, "total_steps": 1479, "loss": 0.5035, "lr": 9.37560730977577e-06, "epoch": 0.7342799188640974, "percentage": 24.48, "elapsed_time": "1:18:15", "remaining_time": "4:01:29"}
+{"current_steps": 363, "total_steps": 1479, "loss": 0.4653, "lr": 9.369884280828806e-06, "epoch": 0.7363083164300203, "percentage": 24.54, "elapsed_time": "1:18:28", "remaining_time": "4:01:16"}
+{"current_steps": 364, "total_steps": 1479, "loss": 0.4743, "lr": 9.364136906680573e-06, "epoch": 0.7383367139959433, "percentage": 24.61, "elapsed_time": "1:18:41", "remaining_time": "4:01:02"}
+{"current_steps": 365, "total_steps": 1479, "loss": 0.4831, "lr": 9.358365219350448e-06, "epoch": 0.7403651115618661, "percentage": 24.68, "elapsed_time": "1:18:54", "remaining_time": "4:00:48"}
+{"current_steps": 366, "total_steps": 1479, "loss": 0.4863, "lr": 9.352569250993265e-06, "epoch": 0.742393509127789, "percentage": 24.75, "elapsed_time": "1:19:06", "remaining_time": "4:00:35"}
+{"current_steps": 367, "total_steps": 1479, "loss": 0.4874, "lr": 9.346749033899121e-06, "epoch": 0.744421906693712, "percentage": 24.81, "elapsed_time": "1:19:19", "remaining_time": "4:00:21"}
+{"current_steps": 368, "total_steps": 1479, "loss": 0.4776, "lr": 9.34090460049322e-06, "epoch": 0.7464503042596349, "percentage": 24.88, "elapsed_time": "1:19:32", "remaining_time": "4:00:07"}
+{"current_steps": 369, "total_steps": 1479, "loss": 0.4771, "lr": 9.335035983335667e-06, "epoch": 0.7484787018255578, "percentage": 24.95, "elapsed_time": "1:19:45", "remaining_time": "3:59:54"}
+{"current_steps": 370, "total_steps": 1479, "loss": 0.4575, "lr": 9.329143215121301e-06, "epoch": 0.7505070993914807, "percentage": 25.02, "elapsed_time": "1:19:57", "remaining_time": "3:59:40"}
+{"current_steps": 371, "total_steps": 1479, "loss": 0.4588, "lr": 9.323226328679512e-06, "epoch": 0.7525354969574036, "percentage": 25.08, "elapsed_time": "1:20:10", "remaining_time": "3:59:26"}
+{"current_steps": 372, "total_steps": 1479, "loss": 0.471, "lr": 9.317285356974055e-06, "epoch": 0.7545638945233266, "percentage": 25.15, "elapsed_time": "1:20:23", "remaining_time": "3:59:13"}
+{"current_steps": 373, "total_steps": 1479, "loss": 0.4583, "lr": 9.311320333102864e-06, "epoch": 0.7565922920892495, "percentage": 25.22, "elapsed_time": "1:20:36", "remaining_time": "3:59:00"}
+{"current_steps": 374, "total_steps": 1479, "loss": 0.4762, "lr": 9.305331290297876e-06, "epoch": 0.7586206896551724, "percentage": 25.29, "elapsed_time": "1:20:48", "remaining_time": "3:58:46"}
+{"current_steps": 375, "total_steps": 1479, "loss": 0.4646, "lr": 9.299318261924834e-06, "epoch": 0.7606490872210954, "percentage": 25.35, "elapsed_time": "1:21:01", "remaining_time": "3:58:32"}
+{"current_steps": 376, "total_steps": 1479, "loss": 0.478, "lr": 9.293281281483115e-06, "epoch": 0.7626774847870182, "percentage": 25.42, "elapsed_time": "1:21:14", "remaining_time": "3:58:19"}
+{"current_steps": 377, "total_steps": 1479, "loss": 0.4712, "lr": 9.287220382605532e-06, "epoch": 0.7647058823529411, "percentage": 25.49, "elapsed_time": "1:21:27", "remaining_time": "3:58:05"}
+{"current_steps": 378, "total_steps": 1479, "loss": 0.4714, "lr": 9.28113559905815e-06, "epoch": 0.7667342799188641, "percentage": 25.56, "elapsed_time": "1:21:39", "remaining_time": "3:57:52"}
+{"current_steps": 379, "total_steps": 1479, "loss": 0.4631, "lr": 9.275026964740101e-06, "epoch": 0.768762677484787, "percentage": 25.63, "elapsed_time": "1:21:52", "remaining_time": "3:57:38"}
+{"current_steps": 380, "total_steps": 1479, "loss": 0.4923, "lr": 9.26889451368339e-06, "epoch": 0.77079107505071, "percentage": 25.69, "elapsed_time": "1:22:05", "remaining_time": "3:57:24"}
+{"current_steps": 381, "total_steps": 1479, "loss": 0.4831, "lr": 9.262738280052715e-06, "epoch": 0.7728194726166329, "percentage": 25.76, "elapsed_time": "1:22:18", "remaining_time": "3:57:11"}
+{"current_steps": 382, "total_steps": 1479, "loss": 0.4898, "lr": 9.25655829814526e-06, "epoch": 0.7748478701825557, "percentage": 25.83, "elapsed_time": "1:22:30", "remaining_time": "3:56:57"}
+{"current_steps": 383, "total_steps": 1479, "loss": 0.4734, "lr": 9.250354602390523e-06, "epoch": 0.7768762677484787, "percentage": 25.9, "elapsed_time": "1:22:43", "remaining_time": "3:56:44"}
+{"current_steps": 384, "total_steps": 1479, "loss": 0.469, "lr": 9.244127227350107e-06, "epoch": 0.7789046653144016, "percentage": 25.96, "elapsed_time": "1:22:56", "remaining_time": "3:56:30"}
+{"current_steps": 385, "total_steps": 1479, "loss": 0.4877, "lr": 9.237876207717538e-06, "epoch": 0.7809330628803245, "percentage": 26.03, "elapsed_time": "1:23:09", "remaining_time": "3:56:17"}
+{"current_steps": 386, "total_steps": 1479, "loss": 0.5045, "lr": 9.231601578318075e-06, "epoch": 0.7829614604462475, "percentage": 26.1, "elapsed_time": "1:23:21", "remaining_time": "3:56:03"}
+{"current_steps": 387, "total_steps": 1479, "loss": 0.4789, "lr": 9.225303374108503e-06, "epoch": 0.7849898580121704, "percentage": 26.17, "elapsed_time": "1:23:34", "remaining_time": "3:55:50"}
+{"current_steps": 388, "total_steps": 1479, "loss": 0.4882, "lr": 9.218981630176947e-06, "epoch": 0.7870182555780934, "percentage": 26.23, "elapsed_time": "1:23:47", "remaining_time": "3:55:36"}
+{"current_steps": 389, "total_steps": 1479, "loss": 0.4833, "lr": 9.212636381742676e-06, "epoch": 0.7890466531440162, "percentage": 26.3, "elapsed_time": "1:24:00", "remaining_time": "3:55:23"}
+{"current_steps": 390, "total_steps": 1479, "loss": 0.4615, "lr": 9.206267664155906e-06, "epoch": 0.7910750507099391, "percentage": 26.37, "elapsed_time": "1:24:13", "remaining_time": "3:55:10"}
+{"current_steps": 391, "total_steps": 1479, "loss": 0.4858, "lr": 9.199875512897602e-06, "epoch": 0.7931034482758621, "percentage": 26.44, "elapsed_time": "1:24:25", "remaining_time": "3:54:56"}
+{"current_steps": 392, "total_steps": 1479, "loss": 0.4791, "lr": 9.19345996357928e-06, "epoch": 0.795131845841785, "percentage": 26.5, "elapsed_time": "1:24:38", "remaining_time": "3:54:43"}
+{"current_steps": 393, "total_steps": 1479, "loss": 0.484, "lr": 9.187021051942814e-06, "epoch": 0.7971602434077079, "percentage": 26.57, "elapsed_time": "1:24:51", "remaining_time": "3:54:29"}
+{"current_steps": 394, "total_steps": 1479, "loss": 0.4953, "lr": 9.180558813860225e-06, "epoch": 0.7991886409736308, "percentage": 26.64, "elapsed_time": "1:25:04", "remaining_time": "3:54:16"}
+{"current_steps": 395, "total_steps": 1479, "loss": 0.4782, "lr": 9.174073285333498e-06, "epoch": 0.8012170385395537, "percentage": 26.71, "elapsed_time": "1:25:17", "remaining_time": "3:54:02"}
+{"current_steps": 396, "total_steps": 1479, "loss": 0.4713, "lr": 9.167564502494361e-06, "epoch": 0.8032454361054767, "percentage": 26.77, "elapsed_time": "1:25:29", "remaining_time": "3:53:49"}
+{"current_steps": 397, "total_steps": 1479, "loss": 0.4855, "lr": 9.161032501604106e-06, "epoch": 0.8052738336713996, "percentage": 26.84, "elapsed_time": "1:25:42", "remaining_time": "3:53:35"}
+{"current_steps": 398, "total_steps": 1479, "loss": 0.4877, "lr": 9.154477319053368e-06, "epoch": 0.8073022312373225, "percentage": 26.91, "elapsed_time": "1:25:55", "remaining_time": "3:53:22"}
+{"current_steps": 399, "total_steps": 1479, "loss": 0.4774, "lr": 9.147898991361936e-06, "epoch": 0.8093306288032455, "percentage": 26.98, "elapsed_time": "1:26:08", "remaining_time": "3:53:08"}
+{"current_steps": 400, "total_steps": 1479, "loss": 0.4695, "lr": 9.141297555178537e-06, "epoch": 0.8113590263691683, "percentage": 27.05, "elapsed_time": "1:26:20", "remaining_time": "3:52:55"}
+{"current_steps": 401, "total_steps": 1479, "loss": 0.4855, "lr": 9.134673047280644e-06, "epoch": 0.8133874239350912, "percentage": 27.11, "elapsed_time": "1:26:33", "remaining_time": "3:52:42"}
+{"current_steps": 402, "total_steps": 1479, "loss": 0.4779, "lr": 9.128025504574264e-06, "epoch": 0.8154158215010142, "percentage": 27.18, "elapsed_time": "1:26:46", "remaining_time": "3:52:28"}
+{"current_steps": 403, "total_steps": 1479, "loss": 0.4615, "lr": 9.121354964093732e-06, "epoch": 0.8174442190669371, "percentage": 27.25, "elapsed_time": "1:26:59", "remaining_time": "3:52:15"}
+{"current_steps": 404, "total_steps": 1479, "loss": 0.4677, "lr": 9.11466146300151e-06, "epoch": 0.8194726166328601, "percentage": 27.32, "elapsed_time": "1:27:12", "remaining_time": "3:52:02"}
+{"current_steps": 405, "total_steps": 1479, "loss": 0.4623, "lr": 9.107945038587974e-06, "epoch": 0.821501014198783, "percentage": 27.38, "elapsed_time": "1:27:25", "remaining_time": "3:51:49"}
+{"current_steps": 406, "total_steps": 1479, "loss": 0.474, "lr": 9.10120572827121e-06, "epoch": 0.8235294117647058, "percentage": 27.45, "elapsed_time": "1:27:37", "remaining_time": "3:51:35"}
+{"current_steps": 407, "total_steps": 1479, "loss": 0.451, "lr": 9.094443569596802e-06, "epoch": 0.8255578093306288, "percentage": 27.52, "elapsed_time": "1:27:50", "remaining_time": "3:51:22"}
+{"current_steps": 408, "total_steps": 1479, "loss": 0.4812, "lr": 9.08765860023763e-06, "epoch": 0.8275862068965517, "percentage": 27.59, "elapsed_time": "1:28:03", "remaining_time": "3:51:09"}
+{"current_steps": 409, "total_steps": 1479, "loss": 0.4639, "lr": 9.08085085799365e-06, "epoch": 0.8296146044624746, "percentage": 27.65, "elapsed_time": "1:28:16", "remaining_time": "3:50:55"}
+{"current_steps": 410, "total_steps": 1479, "loss": 0.4862, "lr": 9.074020380791694e-06, "epoch": 0.8316430020283976, "percentage": 27.72, "elapsed_time": "1:28:28", "remaining_time": "3:50:42"}
+{"current_steps": 411, "total_steps": 1479, "loss": 0.4631, "lr": 9.067167206685248e-06, "epoch": 0.8336713995943205, "percentage": 27.79, "elapsed_time": "1:28:41", "remaining_time": "3:50:28"}
+{"current_steps": 412, "total_steps": 1479, "loss": 0.48, "lr": 9.060291373854252e-06, "epoch": 0.8356997971602435, "percentage": 27.86, "elapsed_time": "1:28:54", "remaining_time": "3:50:15"}
+{"current_steps": 413, "total_steps": 1479, "loss": 0.4654, "lr": 9.05339292060487e-06, "epoch": 0.8377281947261663, "percentage": 27.92, "elapsed_time": "1:29:07", "remaining_time": "3:50:01"}
+{"current_steps": 414, "total_steps": 1479, "loss": 0.4917, "lr": 9.046471885369302e-06, "epoch": 0.8397565922920892, "percentage": 27.99, "elapsed_time": "1:29:20", "remaining_time": "3:49:48"}
+{"current_steps": 415, "total_steps": 1479, "loss": 0.4778, "lr": 9.039528306705543e-06, "epoch": 0.8417849898580122, "percentage": 28.06, "elapsed_time": "1:29:33", "remaining_time": "3:49:35"}
+{"current_steps": 416, "total_steps": 1479, "loss": 0.468, "lr": 9.032562223297187e-06, "epoch": 0.8438133874239351, "percentage": 28.13, "elapsed_time": "1:29:46", "remaining_time": "3:49:22"}
+{"current_steps": 417, "total_steps": 1479, "loss": 0.4815, "lr": 9.025573673953201e-06, "epoch": 0.845841784989858, "percentage": 28.19, "elapsed_time": "1:29:59", "remaining_time": "3:49:11"}
+{"current_steps": 418, "total_steps": 1479, "loss": 0.47, "lr": 9.018562697607716e-06, "epoch": 0.847870182555781, "percentage": 28.26, "elapsed_time": "1:30:13", "remaining_time": "3:49:00"}
+{"current_steps": 419, "total_steps": 1479, "loss": 0.4787, "lr": 9.011529333319804e-06, "epoch": 0.8498985801217038, "percentage": 28.33, "elapsed_time": "1:30:26", "remaining_time": "3:48:48"}
+{"current_steps": 420, "total_steps": 1479, "loss": 0.4729, "lr": 9.004473620273263e-06, "epoch": 0.8519269776876268, "percentage": 28.4, "elapsed_time": "1:30:39", "remaining_time": "3:48:36"}
+{"current_steps": 421, "total_steps": 1479, "loss": 0.4842, "lr": 8.997395597776404e-06, "epoch": 0.8539553752535497, "percentage": 28.47, "elapsed_time": "1:30:53", "remaining_time": "3:48:25"}
+{"current_steps": 422, "total_steps": 1479, "loss": 0.4881, "lr": 8.990295305261823e-06, "epoch": 0.8559837728194726, "percentage": 28.53, "elapsed_time": "1:31:06", "remaining_time": "3:48:11"}
+{"current_steps": 423, "total_steps": 1479, "loss": 0.4777, "lr": 8.98317278228618e-06, "epoch": 0.8580121703853956, "percentage": 28.6, "elapsed_time": "1:31:19", "remaining_time": "3:47:58"}
+{"current_steps": 424, "total_steps": 1479, "loss": 0.4715, "lr": 8.976028068529997e-06, "epoch": 0.8600405679513184, "percentage": 28.67, "elapsed_time": "1:31:31", "remaining_time": "3:47:44"}
+{"current_steps": 425, "total_steps": 1479, "loss": 0.4685, "lr": 8.96886120379741e-06, "epoch": 0.8620689655172413, "percentage": 28.74, "elapsed_time": "1:31:44", "remaining_time": "3:47:31"}
+{"current_steps": 426, "total_steps": 1479, "loss": 0.4809, "lr": 8.96167222801597e-06, "epoch": 0.8640973630831643, "percentage": 28.8, "elapsed_time": "1:31:57", "remaining_time": "3:47:18"}
+{"current_steps": 427, "total_steps": 1479, "loss": 0.476, "lr": 8.954461181236406e-06, "epoch": 0.8661257606490872, "percentage": 28.87, "elapsed_time": "1:32:10", "remaining_time": "3:47:04"}
+{"current_steps": 428, "total_steps": 1479, "loss": 0.4928, "lr": 8.947228103632414e-06, "epoch": 0.8681541582150102, "percentage": 28.94, "elapsed_time": "1:32:22", "remaining_time": "3:46:51"}
+{"current_steps": 429, "total_steps": 1479, "loss": 0.4988, "lr": 8.939973035500418e-06, "epoch": 0.8701825557809331, "percentage": 29.01, "elapsed_time": "1:32:35", "remaining_time": "3:46:37"}
+{"current_steps": 430, "total_steps": 1479, "loss": 0.4729, "lr": 8.932696017259363e-06, "epoch": 0.8722109533468559, "percentage": 29.07, "elapsed_time": "1:32:48", "remaining_time": "3:46:24"}
+{"current_steps": 431, "total_steps": 1479, "loss": 0.4777, "lr": 8.925397089450473e-06, "epoch": 0.8742393509127789, "percentage": 29.14, "elapsed_time": "1:33:01", "remaining_time": "3:46:11"}
+{"current_steps": 432, "total_steps": 1479, "loss": 0.4673, "lr": 8.91807629273704e-06, "epoch": 0.8762677484787018, "percentage": 29.21, "elapsed_time": "1:33:13", "remaining_time": "3:45:57"}
+{"current_steps": 433, "total_steps": 1479, "loss": 0.4845, "lr": 8.910733667904186e-06, "epoch": 0.8782961460446247, "percentage": 29.28, "elapsed_time": "1:33:26", "remaining_time": "3:45:44"}
+{"current_steps": 434, "total_steps": 1479, "loss": 0.4535, "lr": 8.90336925585864e-06, "epoch": 0.8803245436105477, "percentage": 29.34, "elapsed_time": "1:33:39", "remaining_time": "3:45:30"}
+{"current_steps": 435, "total_steps": 1479, "loss": 0.4594, "lr": 8.895983097628515e-06, "epoch": 0.8823529411764706, "percentage": 29.41, "elapsed_time": "1:33:52", "remaining_time": "3:45:17"}
+{"current_steps": 436, "total_steps": 1479, "loss": 0.4762, "lr": 8.888575234363071e-06, "epoch": 0.8843813387423936, "percentage": 29.48, "elapsed_time": "1:34:05", "remaining_time": "3:45:04"}
+{"current_steps": 437, "total_steps": 1479, "loss": 0.4635, "lr": 8.88114570733249e-06, "epoch": 0.8864097363083164, "percentage": 29.55, "elapsed_time": "1:34:17", "remaining_time": "3:44:50"}
+{"current_steps": 438, "total_steps": 1479, "loss": 0.486, "lr": 8.873694557927648e-06, "epoch": 0.8884381338742393, "percentage": 29.61, "elapsed_time": "1:34:30", "remaining_time": "3:44:37"}
+{"current_steps": 439, "total_steps": 1479, "loss": 0.4803, "lr": 8.866221827659876e-06, "epoch": 0.8904665314401623, "percentage": 29.68, "elapsed_time": "1:34:43", "remaining_time": "3:44:24"}
+{"current_steps": 440, "total_steps": 1479, "loss": 0.477, "lr": 8.858727558160744e-06, "epoch": 0.8924949290060852, "percentage": 29.75, "elapsed_time": "1:34:56", "remaining_time": "3:44:10"}
+{"current_steps": 441, "total_steps": 1479, "loss": 0.4775, "lr": 8.851211791181813e-06, "epoch": 0.8945233265720081, "percentage": 29.82, "elapsed_time": "1:35:08", "remaining_time": "3:43:57"}
+{"current_steps": 442, "total_steps": 1479, "loss": 0.4719, "lr": 8.843674568594412e-06, "epoch": 0.896551724137931, "percentage": 29.89, "elapsed_time": "1:35:21", "remaining_time": "3:43:43"}
+{"current_steps": 443, "total_steps": 1479, "loss": 0.4676, "lr": 8.8361159323894e-06, "epoch": 0.8985801217038539, "percentage": 29.95, "elapsed_time": "1:35:34", "remaining_time": "3:43:30"}
+{"current_steps": 444, "total_steps": 1479, "loss": 0.4875, "lr": 8.828535924676938e-06, "epoch": 0.9006085192697769, "percentage": 30.02, "elapsed_time": "1:35:47", "remaining_time": "3:43:17"}
+{"current_steps": 445, "total_steps": 1479, "loss": 0.4919, "lr": 8.820934587686247e-06, "epoch": 0.9026369168356998, "percentage": 30.09, "elapsed_time": "1:35:59", "remaining_time": "3:43:03"}
+{"current_steps": 446, "total_steps": 1479, "loss": 0.4785, "lr": 8.813311963765376e-06, "epoch": 0.9046653144016227, "percentage": 30.16, "elapsed_time": "1:36:12", "remaining_time": "3:42:50"}
+{"current_steps": 447, "total_steps": 1479, "loss": 0.4678, "lr": 8.805668095380969e-06, "epoch": 0.9066937119675457, "percentage": 30.22, "elapsed_time": "1:36:25", "remaining_time": "3:42:37"}
+{"current_steps": 448, "total_steps": 1479, "loss": 0.4622, "lr": 8.798003025118021e-06, "epoch": 0.9087221095334685, "percentage": 30.29, "elapsed_time": "1:36:38", "remaining_time": "3:42:23"}
+{"current_steps": 449, "total_steps": 1479, "loss": 0.4826, "lr": 8.790316795679654e-06, "epoch": 0.9107505070993914, "percentage": 30.36, "elapsed_time": "1:36:50", "remaining_time": "3:42:10"}
+{"current_steps": 450, "total_steps": 1479, "loss": 0.475, "lr": 8.782609449886862e-06, "epoch": 0.9127789046653144, "percentage": 30.43, "elapsed_time": "1:37:03", "remaining_time": "3:41:56"}
+{"current_steps": 451, "total_steps": 1479, "loss": 0.4501, "lr": 8.774881030678284e-06, "epoch": 0.9148073022312373, "percentage": 30.49, "elapsed_time": "1:37:16", "remaining_time": "3:41:43"}
+{"current_steps": 452, "total_steps": 1479, "loss": 0.4699, "lr": 8.767131581109961e-06, "epoch": 0.9168356997971603, "percentage": 30.56, "elapsed_time": "1:37:29", "remaining_time": "3:41:30"}
+{"current_steps": 453, "total_steps": 1479, "loss": 0.481, "lr": 8.759361144355103e-06, "epoch": 0.9188640973630832, "percentage": 30.63, "elapsed_time": "1:37:41", "remaining_time": "3:41:16"}
+{"current_steps": 454, "total_steps": 1479, "loss": 0.4863, "lr": 8.751569763703833e-06, "epoch": 0.920892494929006, "percentage": 30.7, "elapsed_time": "1:37:54", "remaining_time": "3:41:03"}
+{"current_steps": 455, "total_steps": 1479, "loss": 0.466, "lr": 8.74375748256296e-06, "epoch": 0.922920892494929, "percentage": 30.76, "elapsed_time": "1:38:07", "remaining_time": "3:40:50"}
+{"current_steps": 456, "total_steps": 1479, "loss": 0.4928, "lr": 8.735924344455732e-06, "epoch": 0.9249492900608519, "percentage": 30.83, "elapsed_time": "1:38:20", "remaining_time": "3:40:36"}
+{"current_steps": 457, "total_steps": 1479, "loss": 0.4712, "lr": 8.728070393021595e-06, "epoch": 0.9269776876267748, "percentage": 30.9, "elapsed_time": "1:38:33", "remaining_time": "3:40:23"}
+{"current_steps": 458, "total_steps": 1479, "loss": 0.4712, "lr": 8.720195672015945e-06, "epoch": 0.9290060851926978, "percentage": 30.97, "elapsed_time": "1:38:45", "remaining_time": "3:40:09"}
+{"current_steps": 459, "total_steps": 1479, "loss": 0.4739, "lr": 8.712300225309894e-06, "epoch": 0.9310344827586207, "percentage": 31.03, "elapsed_time": "1:38:58", "remaining_time": "3:39:56"}
+{"current_steps": 460, "total_steps": 1479, "loss": 0.4757, "lr": 8.704384096890014e-06, "epoch": 0.9330628803245437, "percentage": 31.1, "elapsed_time": "1:39:11", "remaining_time": "3:39:43"}
+{"current_steps": 461, "total_steps": 1479, "loss": 0.4762, "lr": 8.6964473308581e-06, "epoch": 0.9350912778904665, "percentage": 31.17, "elapsed_time": "1:39:24", "remaining_time": "3:39:30"}
+{"current_steps": 462, "total_steps": 1479, "loss": 0.4672, "lr": 8.688489971430923e-06, "epoch": 0.9371196754563894, "percentage": 31.24, "elapsed_time": "1:39:36", "remaining_time": "3:39:16"}
+{"current_steps": 463, "total_steps": 1479, "loss": 0.4863, "lr": 8.680512062939976e-06, "epoch": 0.9391480730223124, "percentage": 31.3, "elapsed_time": "1:39:49", "remaining_time": "3:39:03"}
+{"current_steps": 464, "total_steps": 1479, "loss": 0.4587, "lr": 8.672513649831247e-06, "epoch": 0.9411764705882353, "percentage": 31.37, "elapsed_time": "1:40:02", "remaining_time": "3:38:50"}
+{"current_steps": 465, "total_steps": 1479, "loss": 0.4606, "lr": 8.664494776664942e-06, "epoch": 0.9432048681541582, "percentage": 31.44, "elapsed_time": "1:40:15", "remaining_time": "3:38:36"}
+{"current_steps": 466, "total_steps": 1479, "loss": 0.4499, "lr": 8.656455488115265e-06, "epoch": 0.9452332657200812, "percentage": 31.51, "elapsed_time": "1:40:27", "remaining_time": "3:38:23"}
+{"current_steps": 467, "total_steps": 1479, "loss": 0.4742, "lr": 8.64839582897015e-06, "epoch": 0.947261663286004, "percentage": 31.58, "elapsed_time": "1:40:40", "remaining_time": "3:38:10"}
+{"current_steps": 468, "total_steps": 1479, "loss": 0.4697, "lr": 8.640315844131021e-06, "epoch": 0.949290060851927, "percentage": 31.64, "elapsed_time": "1:40:53", "remaining_time": "3:37:56"}
+{"current_steps": 469, "total_steps": 1479, "loss": 0.4404, "lr": 8.63221557861254e-06, "epoch": 0.9513184584178499, "percentage": 31.71, "elapsed_time": "1:41:06", "remaining_time": "3:37:43"}
+{"current_steps": 470, "total_steps": 1479, "loss": 0.4649, "lr": 8.62409507754235e-06, "epoch": 0.9533468559837728, "percentage": 31.78, "elapsed_time": "1:41:18", "remaining_time": "3:37:30"}
+{"current_steps": 471, "total_steps": 1479, "loss": 0.468, "lr": 8.615954386160836e-06, "epoch": 0.9553752535496958, "percentage": 31.85, "elapsed_time": "1:41:31", "remaining_time": "3:37:16"}
+{"current_steps": 472, "total_steps": 1479, "loss": 0.4695, "lr": 8.607793549820862e-06, "epoch": 0.9574036511156186, "percentage": 31.91, "elapsed_time": "1:41:44", "remaining_time": "3:37:03"}
+{"current_steps": 473, "total_steps": 1479, "loss": 0.469, "lr": 8.599612613987522e-06, "epoch": 0.9594320486815415, "percentage": 31.98, "elapsed_time": "1:41:57", "remaining_time": "3:36:50"}
+{"current_steps": 474, "total_steps": 1479, "loss": 0.4909, "lr": 8.591411624237887e-06, "epoch": 0.9614604462474645, "percentage": 32.05, "elapsed_time": "1:42:09", "remaining_time": "3:36:36"}
+{"current_steps": 475, "total_steps": 1479, "loss": 0.4735, "lr": 8.583190626260754e-06, "epoch": 0.9634888438133874, "percentage": 32.12, "elapsed_time": "1:42:22", "remaining_time": "3:36:23"}
+{"current_steps": 476, "total_steps": 1479, "loss": 0.469, "lr": 8.574949665856385e-06, "epoch": 0.9655172413793104, "percentage": 32.18, "elapsed_time": "1:42:35", "remaining_time": "3:36:10"}
+{"current_steps": 477, "total_steps": 1479, "loss": 0.4522, "lr": 8.566688788936254e-06, "epoch": 0.9675456389452333, "percentage": 32.25, "elapsed_time": "1:42:48", "remaining_time": "3:35:57"}
+{"current_steps": 478, "total_steps": 1479, "loss": 0.4825, "lr": 8.558408041522801e-06, "epoch": 0.9695740365111561, "percentage": 32.32, "elapsed_time": "1:43:00", "remaining_time": "3:35:43"}
+{"current_steps": 479, "total_steps": 1479, "loss": 0.4934, "lr": 8.550107469749159e-06, "epoch": 0.9716024340770791, "percentage": 32.39, "elapsed_time": "1:43:13", "remaining_time": "3:35:30"}
+{"current_steps": 480, "total_steps": 1479, "loss": 0.4792, "lr": 8.541787119858904e-06, "epoch": 0.973630831643002, "percentage": 32.45, "elapsed_time": "1:43:26", "remaining_time": "3:35:17"}
+{"current_steps": 481, "total_steps": 1479, "loss": 0.4823, "lr": 8.533447038205805e-06, "epoch": 0.9756592292089249, "percentage": 32.52, "elapsed_time": "1:43:39", "remaining_time": "3:35:04"}
+{"current_steps": 482, "total_steps": 1479, "loss": 0.473, "lr": 8.525087271253553e-06, "epoch": 0.9776876267748479, "percentage": 32.59, "elapsed_time": "1:43:52", "remaining_time": "3:34:50"}
+{"current_steps": 483, "total_steps": 1479, "loss": 0.461, "lr": 8.516707865575515e-06, "epoch": 0.9797160243407708, "percentage": 32.66, "elapsed_time": "1:44:04", "remaining_time": "3:34:37"}
+{"current_steps": 484, "total_steps": 1479, "loss": 0.4741, "lr": 8.508308867854465e-06, "epoch": 0.9817444219066938, "percentage": 32.72, "elapsed_time": "1:44:17", "remaining_time": "3:34:24"}
+{"current_steps": 485, "total_steps": 1479, "loss": 0.4712, "lr": 8.499890324882323e-06, "epoch": 0.9837728194726166, "percentage": 32.79, "elapsed_time": "1:44:30", "remaining_time": "3:34:11"}
+{"current_steps": 486, "total_steps": 1479, "loss": 0.472, "lr": 8.491452283559904e-06, "epoch": 0.9858012170385395, "percentage": 32.86, "elapsed_time": "1:44:43", "remaining_time": "3:33:57"}
+{"current_steps": 487, "total_steps": 1479, "loss": 0.4778, "lr": 8.482994790896645e-06, "epoch": 0.9878296146044625, "percentage": 32.93, "elapsed_time": "1:44:55", "remaining_time": "3:33:44"}
+{"current_steps": 488, "total_steps": 1479, "loss": 0.4657, "lr": 8.474517894010356e-06, "epoch": 0.9898580121703854, "percentage": 33.0, "elapsed_time": "1:45:08", "remaining_time": "3:33:31"}
+{"current_steps": 489, "total_steps": 1479, "loss": 0.4796, "lr": 8.466021640126946e-06, "epoch": 0.9918864097363083, "percentage": 33.06, "elapsed_time": "1:45:21", "remaining_time": "3:33:18"}
+{"current_steps": 490, "total_steps": 1479, "loss": 0.4894, "lr": 8.457506076580164e-06, "epoch": 0.9939148073022313, "percentage": 33.13, "elapsed_time": "1:45:34", "remaining_time": "3:33:05"}
+{"current_steps": 491, "total_steps": 1479, "loss": 0.4583, "lr": 8.448971250811337e-06, "epoch": 0.9959432048681541, "percentage": 33.2, "elapsed_time": "1:45:47", "remaining_time": "3:32:51"}
+{"current_steps": 492, "total_steps": 1479, "loss": 0.4627, "lr": 8.440417210369104e-06, "epoch": 0.9979716024340771, "percentage": 33.27, "elapsed_time": "1:46:00", "remaining_time": "3:32:38"}
+{"current_steps": 493, "total_steps": 1479, "loss": 0.4536, "lr": 8.431844002909153e-06, "epoch": 1.0, "percentage": 33.33, "elapsed_time": "1:46:12", "remaining_time": "3:32:25"}
+{"current_steps": 494, "total_steps": 1479, "loss": 0.4587, "lr": 8.423251676193951e-06, "epoch": 1.002028397565923, "percentage": 33.4, "elapsed_time": "1:47:53", "remaining_time": "3:35:06"}
+{"current_steps": 495, "total_steps": 1479, "loss": 0.4595, "lr": 8.414640278092485e-06, "epoch": 1.0040567951318458, "percentage": 33.47, "elapsed_time": "1:48:05", "remaining_time": "3:34:53"}
+{"current_steps": 496, "total_steps": 1479, "loss": 0.4559, "lr": 8.406009856579986e-06, "epoch": 1.0060851926977687, "percentage": 33.54, "elapsed_time": "1:48:18", "remaining_time": "3:34:39"}
+{"current_steps": 497, "total_steps": 1479, "loss": 0.4551, "lr": 8.397360459737673e-06, "epoch": 1.0081135902636917, "percentage": 33.6, "elapsed_time": "1:48:31", "remaining_time": "3:34:25"}
+{"current_steps": 498, "total_steps": 1479, "loss": 0.4398, "lr": 8.388692135752473e-06, "epoch": 1.0101419878296145, "percentage": 33.67, "elapsed_time": "1:48:44", "remaining_time": "3:34:11"}
+{"current_steps": 499, "total_steps": 1479, "loss": 0.4493, "lr": 8.38000493291676e-06, "epoch": 1.0121703853955375, "percentage": 33.74, "elapsed_time": "1:48:56", "remaining_time": "3:33:58"}
+{"current_steps": 500, "total_steps": 1479, "loss": 0.4549, "lr": 8.371298899628091e-06, "epoch": 1.0141987829614605, "percentage": 33.81, "elapsed_time": "1:49:09", "remaining_time": "3:33:44"}
+{"current_steps": 501, "total_steps": 1479, "loss": 0.4794, "lr": 8.362574084388921e-06, "epoch": 1.0162271805273835, "percentage": 33.87, "elapsed_time": "1:49:22", "remaining_time": "3:33:30"}
+{"current_steps": 502, "total_steps": 1479, "loss": 0.446, "lr": 8.353830535806347e-06, "epoch": 1.0182555780933062, "percentage": 33.94, "elapsed_time": "1:49:35", "remaining_time": "3:33:16"}
+{"current_steps": 503, "total_steps": 1479, "loss": 0.4656, "lr": 8.34506830259183e-06, "epoch": 1.0202839756592292, "percentage": 34.01, "elapsed_time": "1:49:48", "remaining_time": "3:33:03"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e0716ee604bdfd9ddf2c5c92b131e3a5ec41df2d4df3e91542744b4be1eefa
+size 7224