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 sha256:0d54a5e221c7da38eae77766025096e7277233d75838fd16e829e31df74686d9
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9100557274f5ced62b903796b3e6dcc4357a06e572e350be2f3b8d9aa3629970
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c46a9816ae82727592be1dea28c7cd85e96165924216904e1163030df8366a5c
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b243fc52329205eb43efd01439c641a91798f8a02a13af8c360fad457b60600
+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 +1,502 @@
+{"current_steps": 1, "total_steps": 1500, "loss": 1.2436, "lr": 6.666666666666668e-08, "epoch": 0.002, "percentage": 0.07, "elapsed_time": "0:01:02", "remaining_time": "1 day, 1:58:26"}
+{"current_steps": 2, "total_steps": 1500, "loss": 1.2722, "lr": 1.3333333333333336e-07, "epoch": 0.004, "percentage": 0.13, "elapsed_time": "0:01:57", "remaining_time": "1 day, 0:27:18"}
+{"current_steps": 3, "total_steps": 1500, "loss": 1.2199, "lr": 2.0000000000000002e-07, "epoch": 0.006, "percentage": 0.2, "elapsed_time": "0:03:05", "remaining_time": "1 day, 1:42:18"}
+{"current_steps": 4, "total_steps": 1500, "loss": 1.2601, "lr": 2.666666666666667e-07, "epoch": 0.008, "percentage": 0.27, "elapsed_time": "0:04:04", "remaining_time": "1 day, 1:21:32"}
+{"current_steps": 5, "total_steps": 1500, "loss": 1.279, "lr": 3.3333333333333335e-07, "epoch": 0.01, "percentage": 0.33, "elapsed_time": "0:04:59", "remaining_time": "1 day, 0:54:13"}
+{"current_steps": 6, "total_steps": 1500, "loss": 1.2378, "lr": 4.0000000000000003e-07, "epoch": 0.012, "percentage": 0.4, "elapsed_time": "0:05:47", "remaining_time": "1 day, 0:01:09"}
+{"current_steps": 7, "total_steps": 1500, "loss": 1.2796, "lr": 4.666666666666667e-07, "epoch": 0.014, "percentage": 0.47, "elapsed_time": "0:06:50", "remaining_time": "1 day, 0:19:52"}
+{"current_steps": 8, "total_steps": 1500, "loss": 1.2348, "lr": 5.333333333333335e-07, "epoch": 0.016, "percentage": 0.53, "elapsed_time": "0:07:54", "remaining_time": "1 day, 0:35:02"}
+{"current_steps": 9, "total_steps": 1500, "loss": 1.2454, "lr": 6.000000000000001e-07, "epoch": 0.018, "percentage": 0.6, "elapsed_time": "0:08:55", "remaining_time": "1 day, 0:39:34"}
+{"current_steps": 10, "total_steps": 1500, "loss": 1.2112, "lr": 6.666666666666667e-07, "epoch": 0.02, "percentage": 0.67, "elapsed_time": "0:09:45", "remaining_time": "1 day, 0:14:36"}
+{"current_steps": 11, "total_steps": 1500, "loss": 1.2471, "lr": 7.333333333333334e-07, "epoch": 0.022, "percentage": 0.73, "elapsed_time": "0:10:42", "remaining_time": "1 day, 0:09:18"}
+{"current_steps": 12, "total_steps": 1500, "loss": 1.2422, "lr": 8.000000000000001e-07, "epoch": 0.024, "percentage": 0.8, "elapsed_time": "0:11:43", "remaining_time": "1 day, 0:13:46"}
+{"current_steps": 13, "total_steps": 1500, "loss": 1.2303, "lr": 8.666666666666668e-07, "epoch": 0.026, "percentage": 0.87, "elapsed_time": "0:12:38", "remaining_time": "1 day, 0:06:21"}
+{"current_steps": 14, "total_steps": 1500, "loss": 1.2304, "lr": 9.333333333333334e-07, "epoch": 0.028, "percentage": 0.93, "elapsed_time": "0:13:31", "remaining_time": "23:55:03"}
+{"current_steps": 15, "total_steps": 1500, "loss": 1.2018, "lr": 1.0000000000000002e-06, "epoch": 0.03, "percentage": 1.0, "elapsed_time": "0:14:34", "remaining_time": "1 day, 0:02:49"}
+{"current_steps": 16, "total_steps": 1500, "loss": 1.2221, "lr": 1.066666666666667e-06, "epoch": 0.032, "percentage": 1.07, "elapsed_time": "0:15:28", "remaining_time": "23:55:03"}
+{"current_steps": 17, "total_steps": 1500, "loss": 1.139, "lr": 1.1333333333333334e-06, "epoch": 0.034, "percentage": 1.13, "elapsed_time": "0:16:23", "remaining_time": "23:49:55"}
+{"current_steps": 18, "total_steps": 1500, "loss": 1.1654, "lr": 1.2000000000000002e-06, "epoch": 0.036, "percentage": 1.2, "elapsed_time": "0:17:10", "remaining_time": "23:34:13"}
+{"current_steps": 19, "total_steps": 1500, "loss": 1.1389, "lr": 1.2666666666666669e-06, "epoch": 0.038, "percentage": 1.27, "elapsed_time": "0:18:03", "remaining_time": "23:28:12"}
+{"current_steps": 20, "total_steps": 1500, "loss": 1.1583, "lr": 1.3333333333333334e-06, "epoch": 0.04, "percentage": 1.33, "elapsed_time": "0:19:01", "remaining_time": "23:27:55"}
+{"current_steps": 21, "total_steps": 1500, "loss": 1.1342, "lr": 1.4000000000000001e-06, "epoch": 0.042, "percentage": 1.4, "elapsed_time": "0:19:47", "remaining_time": "23:14:02"}
+{"current_steps": 22, "total_steps": 1500, "loss": 1.159, "lr": 1.4666666666666669e-06, "epoch": 0.044, "percentage": 1.47, "elapsed_time": "0:20:43", "remaining_time": "23:12:35"}
+{"current_steps": 23, "total_steps": 1500, "loss": 1.1454, "lr": 1.5333333333333334e-06, "epoch": 0.046, "percentage": 1.53, "elapsed_time": "0:21:37", "remaining_time": "23:08:54"}
+{"current_steps": 24, "total_steps": 1500, "loss": 1.1222, "lr": 1.6000000000000001e-06, "epoch": 0.048, "percentage": 1.6, "elapsed_time": "0:22:33", "remaining_time": "23:06:51"}
+{"current_steps": 25, "total_steps": 1500, "loss": 1.123, "lr": 1.6666666666666667e-06, "epoch": 0.05, "percentage": 1.67, "elapsed_time": "0:23:29", "remaining_time": "23:06:15"}
+{"current_steps": 26, "total_steps": 1500, "loss": 1.1258, "lr": 1.7333333333333336e-06, "epoch": 0.052, "percentage": 1.73, "elapsed_time": "0:24:26", "remaining_time": "23:05:55"}
+{"current_steps": 27, "total_steps": 1500, "loss": 1.0751, "lr": 1.8000000000000001e-06, "epoch": 0.054, "percentage": 1.8, "elapsed_time": "0:25:08", "remaining_time": "22:51:54"}
+{"current_steps": 28, "total_steps": 1500, "loss": 1.0648, "lr": 1.8666666666666669e-06, "epoch": 0.056, "percentage": 1.87, "elapsed_time": "0:26:02", "remaining_time": "22:49:25"}
+{"current_steps": 29, "total_steps": 1500, "loss": 1.0898, "lr": 1.9333333333333336e-06, "epoch": 0.058, "percentage": 1.93, "elapsed_time": "0:26:58", "remaining_time": "22:48:28"}
+{"current_steps": 30, "total_steps": 1500, "loss": 1.0935, "lr": 2.0000000000000003e-06, "epoch": 0.06, "percentage": 2.0, "elapsed_time": "0:28:02", "remaining_time": "22:54:10"}
+{"current_steps": 31, "total_steps": 1500, "loss": 1.0833, "lr": 2.0666666666666666e-06, "epoch": 0.062, "percentage": 2.07, "elapsed_time": "0:28:50", "remaining_time": "22:46:37"}
+{"current_steps": 32, "total_steps": 1500, "loss": 1.0441, "lr": 2.133333333333334e-06, "epoch": 0.064, "percentage": 2.13, "elapsed_time": "0:29:51", "remaining_time": "22:50:06"}
+{"current_steps": 33, "total_steps": 1500, "loss": 1.0262, "lr": 2.2e-06, "epoch": 0.066, "percentage": 2.2, "elapsed_time": "0:30:44", "remaining_time": "22:46:57"}
+{"current_steps": 34, "total_steps": 1500, "loss": 1.0388, "lr": 2.266666666666667e-06, "epoch": 0.068, "percentage": 2.27, "elapsed_time": "0:31:37", "remaining_time": "22:43:42"}
+{"current_steps": 35, "total_steps": 1500, "loss": 1.037, "lr": 2.3333333333333336e-06, "epoch": 0.07, "percentage": 2.33, "elapsed_time": "0:32:27", "remaining_time": "22:38:19"}
+{"current_steps": 36, "total_steps": 1500, "loss": 1.0476, "lr": 2.4000000000000003e-06, "epoch": 0.072, "percentage": 2.4, "elapsed_time": "0:33:26", "remaining_time": "22:39:57"}
+{"current_steps": 37, "total_steps": 1500, "loss": 1.0496, "lr": 2.466666666666667e-06, "epoch": 0.074, "percentage": 2.47, "elapsed_time": "0:34:14", "remaining_time": "22:34:11"}
+{"current_steps": 38, "total_steps": 1500, "loss": 1.0307, "lr": 2.5333333333333338e-06, "epoch": 0.076, "percentage": 2.53, "elapsed_time": "0:35:04", "remaining_time": "22:29:40"}
+{"current_steps": 39, "total_steps": 1500, "loss": 1.0279, "lr": 2.6e-06, "epoch": 0.078, "percentage": 2.6, "elapsed_time": "0:35:58", "remaining_time": "22:27:25"}
+{"current_steps": 40, "total_steps": 1500, "loss": 1.0201, "lr": 2.666666666666667e-06, "epoch": 0.08, "percentage": 2.67, "elapsed_time": "0:37:02", "remaining_time": "22:31:57"}
+{"current_steps": 41, "total_steps": 1500, "loss": 1.0028, "lr": 2.7333333333333336e-06, "epoch": 0.082, "percentage": 2.73, "elapsed_time": "0:37:59", "remaining_time": "22:31:57"}
+{"current_steps": 42, "total_steps": 1500, "loss": 1.0085, "lr": 2.8000000000000003e-06, "epoch": 0.084, "percentage": 2.8, "elapsed_time": "0:38:56", "remaining_time": "22:31:59"}
+{"current_steps": 43, "total_steps": 1500, "loss": 1.0024, "lr": 2.866666666666667e-06, "epoch": 0.086, "percentage": 2.87, "elapsed_time": "0:39:52", "remaining_time": "22:31:09"}
+{"current_steps": 44, "total_steps": 1500, "loss": 0.9924, "lr": 2.9333333333333338e-06, "epoch": 0.088, "percentage": 2.93, "elapsed_time": "0:40:38", "remaining_time": "22:25:07"}
+{"current_steps": 45, "total_steps": 1500, "loss": 1.0107, "lr": 3e-06, "epoch": 0.09, "percentage": 3.0, "elapsed_time": "0:41:27", "remaining_time": "22:20:42"}
+{"current_steps": 46, "total_steps": 1500, "loss": 0.9944, "lr": 3.066666666666667e-06, "epoch": 0.092, "percentage": 3.07, "elapsed_time": "0:42:20", "remaining_time": "22:18:08"}
+{"current_steps": 47, "total_steps": 1500, "loss": 0.9756, "lr": 3.133333333333334e-06, "epoch": 0.094, "percentage": 3.13, "elapsed_time": "0:43:15", "remaining_time": "22:17:29"}
+{"current_steps": 48, "total_steps": 1500, "loss": 0.9564, "lr": 3.2000000000000003e-06, "epoch": 0.096, "percentage": 3.2, "elapsed_time": "0:44:13", "remaining_time": "22:17:39"}
+{"current_steps": 49, "total_steps": 1500, "loss": 0.9787, "lr": 3.266666666666667e-06, "epoch": 0.098, "percentage": 3.27, "elapsed_time": "0:45:06", "remaining_time": "22:15:59"}
+{"current_steps": 50, "total_steps": 1500, "loss": 0.9744, "lr": 3.3333333333333333e-06, "epoch": 0.1, "percentage": 3.33, "elapsed_time": "0:46:07", "remaining_time": "22:17:31"}
+{"current_steps": 51, "total_steps": 1500, "loss": 0.9686, "lr": 3.4000000000000005e-06, "epoch": 0.102, "percentage": 3.4, "elapsed_time": "0:47:03", "remaining_time": "22:16:48"}
+{"current_steps": 52, "total_steps": 1500, "loss": 0.9644, "lr": 3.4666666666666672e-06, "epoch": 0.104, "percentage": 3.47, "elapsed_time": "0:47:54", "remaining_time": "22:14:02"}
+{"current_steps": 53, "total_steps": 1500, "loss": 0.9486, "lr": 3.5333333333333335e-06, "epoch": 0.106, "percentage": 3.53, "elapsed_time": "0:48:42", "remaining_time": "22:09:56"}
+{"current_steps": 54, "total_steps": 1500, "loss": 0.9473, "lr": 3.6000000000000003e-06, "epoch": 0.108, "percentage": 3.6, "elapsed_time": "0:49:31", "remaining_time": "22:05:59"}
+{"current_steps": 55, "total_steps": 1500, "loss": 0.9851, "lr": 3.6666666666666666e-06, "epoch": 0.11, "percentage": 3.67, "elapsed_time": "0:50:30", "remaining_time": "22:07:11"}
+{"current_steps": 56, "total_steps": 1500, "loss": 0.9226, "lr": 3.7333333333333337e-06, "epoch": 0.112, "percentage": 3.73, "elapsed_time": "0:51:24", "remaining_time": "22:05:34"}
+{"current_steps": 57, "total_steps": 1500, "loss": 0.9252, "lr": 3.8000000000000005e-06, "epoch": 0.114, "percentage": 3.8, "elapsed_time": "0:52:15", "remaining_time": "22:02:48"}
+{"current_steps": 58, "total_steps": 1500, "loss": 0.9495, "lr": 3.866666666666667e-06, "epoch": 0.116, "percentage": 3.87, "elapsed_time": "0:53:16", "remaining_time": "22:04:24"}
+{"current_steps": 59, "total_steps": 1500, "loss": 0.9693, "lr": 3.9333333333333335e-06, "epoch": 0.118, "percentage": 3.93, "elapsed_time": "0:54:06", "remaining_time": "22:01:25"}
+{"current_steps": 60, "total_steps": 1500, "loss": 0.9473, "lr": 4.000000000000001e-06, "epoch": 0.12, "percentage": 4.0, "elapsed_time": "0:55:02", "remaining_time": "22:01:01"}
+{"current_steps": 61, "total_steps": 1500, "loss": 0.9563, "lr": 4.066666666666667e-06, "epoch": 0.122, "percentage": 4.07, "elapsed_time": "0:55:58", "remaining_time": "22:00:30"}
+{"current_steps": 62, "total_steps": 1500, "loss": 0.9355, "lr": 4.133333333333333e-06, "epoch": 0.124, "percentage": 4.13, "elapsed_time": "0:56:57", "remaining_time": "22:00:53"}
+{"current_steps": 63, "total_steps": 1500, "loss": 0.9706, "lr": 4.2000000000000004e-06, "epoch": 0.126, "percentage": 4.2, "elapsed_time": "0:58:05", "remaining_time": "22:04:51"}
+{"current_steps": 64, "total_steps": 1500, "loss": 0.955, "lr": 4.266666666666668e-06, "epoch": 0.128, "percentage": 4.27, "elapsed_time": "0:59:01", "remaining_time": "22:04:25"}
+{"current_steps": 65, "total_steps": 1500, "loss": 0.9396, "lr": 4.333333333333334e-06, "epoch": 0.13, "percentage": 4.33, "elapsed_time": "0:59:49", "remaining_time": "22:00:54"}
+{"current_steps": 66, "total_steps": 1500, "loss": 0.9023, "lr": 4.4e-06, "epoch": 0.132, "percentage": 4.4, "elapsed_time": "1:00:57", "remaining_time": "22:04:27"}
+{"current_steps": 67, "total_steps": 1500, "loss": 0.914, "lr": 4.4666666666666665e-06, "epoch": 0.134, "percentage": 4.47, "elapsed_time": "1:01:55", "remaining_time": "22:04:23"}
+{"current_steps": 68, "total_steps": 1500, "loss": 0.9292, "lr": 4.533333333333334e-06, "epoch": 0.136, "percentage": 4.53, "elapsed_time": "1:02:46", "remaining_time": "22:01:59"}
+{"current_steps": 69, "total_steps": 1500, "loss": 0.9057, "lr": 4.600000000000001e-06, "epoch": 0.138, "percentage": 4.6, "elapsed_time": "1:03:35", "remaining_time": "21:58:51"}
+{"current_steps": 70, "total_steps": 1500, "loss": 0.9358, "lr": 4.666666666666667e-06, "epoch": 0.14, "percentage": 4.67, "elapsed_time": "1:04:33", "remaining_time": "21:58:46"}
+{"current_steps": 71, "total_steps": 1500, "loss": 0.9357, "lr": 4.7333333333333335e-06, "epoch": 0.142, "percentage": 4.73, "elapsed_time": "1:05:22", "remaining_time": "21:55:55"}
+{"current_steps": 72, "total_steps": 1500, "loss": 0.9601, "lr": 4.800000000000001e-06, "epoch": 0.144, "percentage": 4.8, "elapsed_time": "1:06:20", "remaining_time": "21:55:48"}
+{"current_steps": 73, "total_steps": 1500, "loss": 0.9408, "lr": 4.866666666666667e-06, "epoch": 0.146, "percentage": 4.87, "elapsed_time": "1:07:22", "remaining_time": "21:56:54"}
+{"current_steps": 74, "total_steps": 1500, "loss": 0.9143, "lr": 4.933333333333334e-06, "epoch": 0.148, "percentage": 4.93, "elapsed_time": "1:08:10", "remaining_time": "21:53:53"}
+{"current_steps": 75, "total_steps": 1500, "loss": 0.9461, "lr": 5e-06, "epoch": 0.15, "percentage": 5.0, "elapsed_time": "1:09:15", "remaining_time": "21:55:46"}
+{"current_steps": 76, "total_steps": 1500, "loss": 0.9324, "lr": 5.0666666666666676e-06, "epoch": 0.152, "percentage": 5.07, "elapsed_time": "1:10:05", "remaining_time": "21:53:15"}
+{"current_steps": 77, "total_steps": 1500, "loss": 0.9284, "lr": 5.133333333333334e-06, "epoch": 0.154, "percentage": 5.13, "elapsed_time": "1:11:00", "remaining_time": "21:52:11"}
+{"current_steps": 78, "total_steps": 1500, "loss": 0.9189, "lr": 5.2e-06, "epoch": 0.156, "percentage": 5.2, "elapsed_time": "1:12:06", "remaining_time": "21:54:33"}
+{"current_steps": 79, "total_steps": 1500, "loss": 0.8917, "lr": 5.2666666666666665e-06, "epoch": 0.158, "percentage": 5.27, "elapsed_time": "1:12:59", "remaining_time": "21:53:01"}
+{"current_steps": 80, "total_steps": 1500, "loss": 0.9101, "lr": 5.333333333333334e-06, "epoch": 0.16, "percentage": 5.33, "elapsed_time": "1:14:00", "remaining_time": "21:53:45"}
+{"current_steps": 81, "total_steps": 1500, "loss": 0.9442, "lr": 5.400000000000001e-06, "epoch": 0.162, "percentage": 5.4, "elapsed_time": "1:15:00", "remaining_time": "21:53:54"}
+{"current_steps": 82, "total_steps": 1500, "loss": 0.9173, "lr": 5.466666666666667e-06, "epoch": 0.164, "percentage": 5.47, "elapsed_time": "1:15:59", "remaining_time": "21:53:58"}
+{"current_steps": 83, "total_steps": 1500, "loss": 0.8942, "lr": 5.533333333333334e-06, "epoch": 0.166, "percentage": 5.53, "elapsed_time": "1:16:52", "remaining_time": "21:52:31"}
+{"current_steps": 84, "total_steps": 1500, "loss": 0.9025, "lr": 5.600000000000001e-06, "epoch": 0.168, "percentage": 5.6, "elapsed_time": "1:17:48", "remaining_time": "21:51:36"}
+{"current_steps": 85, "total_steps": 1500, "loss": 0.9191, "lr": 5.666666666666667e-06, "epoch": 0.17, "percentage": 5.67, "elapsed_time": "1:18:45", "remaining_time": "21:50:59"}
+{"current_steps": 86, "total_steps": 1500, "loss": 0.9195, "lr": 5.733333333333334e-06, "epoch": 0.172, "percentage": 5.73, "elapsed_time": "1:19:49", "remaining_time": "21:52:34"}
+{"current_steps": 87, "total_steps": 1500, "loss": 0.8992, "lr": 5.8e-06, "epoch": 0.174, "percentage": 5.8, "elapsed_time": "1:20:47", "remaining_time": "21:52:12"}
+{"current_steps": 88, "total_steps": 1500, "loss": 0.9046, "lr": 5.8666666666666675e-06, "epoch": 0.176, "percentage": 5.87, "elapsed_time": "1:21:40", "remaining_time": "21:50:24"}
+{"current_steps": 89, "total_steps": 1500, "loss": 0.9309, "lr": 5.933333333333335e-06, "epoch": 0.178, "percentage": 5.93, "elapsed_time": "1:22:36", "remaining_time": "21:49:43"}
+{"current_steps": 90, "total_steps": 1500, "loss": 0.9449, "lr": 6e-06, "epoch": 0.18, "percentage": 6.0, "elapsed_time": "1:23:38", "remaining_time": "21:50:30"}
+{"current_steps": 91, "total_steps": 1500, "loss": 0.959, "lr": 6.066666666666667e-06, "epoch": 0.182, "percentage": 6.07, "elapsed_time": "1:24:37", "remaining_time": "21:50:22"}
+{"current_steps": 92, "total_steps": 1500, "loss": 0.9479, "lr": 6.133333333333334e-06, "epoch": 0.184, "percentage": 6.13, "elapsed_time": "1:25:40", "remaining_time": "21:51:16"}
+{"current_steps": 93, "total_steps": 1500, "loss": 0.8791, "lr": 6.200000000000001e-06, "epoch": 0.186, "percentage": 6.2, "elapsed_time": "1:26:30", "remaining_time": "21:48:53"}
+{"current_steps": 94, "total_steps": 1500, "loss": 0.9234, "lr": 6.266666666666668e-06, "epoch": 0.188, "percentage": 6.27, "elapsed_time": "1:27:25", "remaining_time": "21:47:39"}
+{"current_steps": 95, "total_steps": 1500, "loss": 0.9209, "lr": 6.333333333333333e-06, "epoch": 0.19, "percentage": 6.33, "elapsed_time": "1:28:30", "remaining_time": "21:48:58"}
+{"current_steps": 96, "total_steps": 1500, "loss": 0.931, "lr": 6.4000000000000006e-06, "epoch": 0.192, "percentage": 6.4, "elapsed_time": "1:29:22", "remaining_time": "21:46:59"}
+{"current_steps": 97, "total_steps": 1500, "loss": 0.9381, "lr": 6.466666666666667e-06, "epoch": 0.194, "percentage": 6.47, "elapsed_time": "1:30:17", "remaining_time": "21:46:04"}
+{"current_steps": 98, "total_steps": 1500, "loss": 0.9074, "lr": 6.533333333333334e-06, "epoch": 0.196, "percentage": 6.53, "elapsed_time": "1:31:19", "remaining_time": "21:46:29"}
+{"current_steps": 99, "total_steps": 1500, "loss": 0.9359, "lr": 6.600000000000001e-06, "epoch": 0.198, "percentage": 6.6, "elapsed_time": "1:32:19", "remaining_time": "21:46:30"}
+{"current_steps": 100, "total_steps": 1500, "loss": 0.9097, "lr": 6.666666666666667e-06, "epoch": 0.2, "percentage": 6.67, "elapsed_time": "1:33:08", "remaining_time": "21:44:01"}
+{"current_steps": 101, "total_steps": 1500, "loss": 0.9355, "lr": 6.733333333333334e-06, "epoch": 0.202, "percentage": 6.73, "elapsed_time": "1:34:03", "remaining_time": "21:42:56"}
+{"current_steps": 102, "total_steps": 1500, "loss": 0.8883, "lr": 6.800000000000001e-06, "epoch": 0.204, "percentage": 6.8, "elapsed_time": "1:35:07", "remaining_time": "21:43:51"}
+{"current_steps": 103, "total_steps": 1500, "loss": 0.9054, "lr": 6.866666666666667e-06, "epoch": 0.206, "percentage": 6.87, "elapsed_time": "1:36:06", "remaining_time": "21:43:26"}
+{"current_steps": 104, "total_steps": 1500, "loss": 0.9196, "lr": 6.9333333333333344e-06, "epoch": 0.208, "percentage": 6.93, "elapsed_time": "1:37:06", "remaining_time": "21:43:25"}
+{"current_steps": 105, "total_steps": 1500, "loss": 0.9175, "lr": 7e-06, "epoch": 0.21, "percentage": 7.0, "elapsed_time": "1:38:07", "remaining_time": "21:43:43"}
+{"current_steps": 106, "total_steps": 1500, "loss": 0.9186, "lr": 7.066666666666667e-06, "epoch": 0.212, "percentage": 7.07, "elapsed_time": "1:39:08", "remaining_time": "21:43:45"}
+{"current_steps": 107, "total_steps": 1500, "loss": 0.9102, "lr": 7.133333333333334e-06, "epoch": 0.214, "percentage": 7.13, "elapsed_time": "1:40:05", "remaining_time": "21:43:01"}
+{"current_steps": 108, "total_steps": 1500, "loss": 0.9181, "lr": 7.2000000000000005e-06, "epoch": 0.216, "percentage": 7.2, "elapsed_time": "1:41:00", "remaining_time": "21:41:57"}
+{"current_steps": 109, "total_steps": 1500, "loss": 0.8487, "lr": 7.266666666666668e-06, "epoch": 0.218, "percentage": 7.27, "elapsed_time": "1:41:45", "remaining_time": "21:38:40"}
+{"current_steps": 110, "total_steps": 1500, "loss": 0.8934, "lr": 7.333333333333333e-06, "epoch": 0.22, "percentage": 7.33, "elapsed_time": "1:42:37", "remaining_time": "21:36:42"}
+{"current_steps": 111, "total_steps": 1500, "loss": 0.927, "lr": 7.4e-06, "epoch": 0.222, "percentage": 7.4, "elapsed_time": "1:43:40", "remaining_time": "21:37:21"}
+{"current_steps": 112, "total_steps": 1500, "loss": 0.9275, "lr": 7.4666666666666675e-06, "epoch": 0.224, "percentage": 7.47, "elapsed_time": "1:44:48", "remaining_time": "21:38:47"}
+{"current_steps": 113, "total_steps": 1500, "loss": 0.9078, "lr": 7.533333333333334e-06, "epoch": 0.226, "percentage": 7.53, "elapsed_time": "1:45:46", "remaining_time": "21:38:15"}
+{"current_steps": 114, "total_steps": 1500, "loss": 0.9462, "lr": 7.600000000000001e-06, "epoch": 0.228, "percentage": 7.6, "elapsed_time": "1:46:56", "remaining_time": "21:40:10"}
+{"current_steps": 115, "total_steps": 1500, "loss": 0.9008, "lr": 7.666666666666667e-06, "epoch": 0.23, "percentage": 7.67, "elapsed_time": "1:47:49", "remaining_time": "21:38:38"}
+{"current_steps": 116, "total_steps": 1500, "loss": 0.8878, "lr": 7.733333333333334e-06, "epoch": 0.232, "percentage": 7.73, "elapsed_time": "1:48:41", "remaining_time": "21:36:49"}
+{"current_steps": 117, "total_steps": 1500, "loss": 0.9021, "lr": 7.800000000000002e-06, "epoch": 0.234, "percentage": 7.8, "elapsed_time": "1:49:33", "remaining_time": "21:35:06"}
+{"current_steps": 118, "total_steps": 1500, "loss": 0.9012, "lr": 7.866666666666667e-06, "epoch": 0.236, "percentage": 7.87, "elapsed_time": "1:50:23", "remaining_time": "21:32:57"}
+{"current_steps": 119, "total_steps": 1500, "loss": 0.936, "lr": 7.933333333333334e-06, "epoch": 0.238, "percentage": 7.93, "elapsed_time": "1:51:19", "remaining_time": "21:31:56"}
+{"current_steps": 120, "total_steps": 1500, "loss": 0.8952, "lr": 8.000000000000001e-06, "epoch": 0.24, "percentage": 8.0, "elapsed_time": "1:52:23", "remaining_time": "21:32:28"}
+{"current_steps": 121, "total_steps": 1500, "loss": 0.9103, "lr": 8.066666666666667e-06, "epoch": 0.242, "percentage": 8.07, "elapsed_time": "1:53:27", "remaining_time": "21:33:07"}
+{"current_steps": 122, "total_steps": 1500, "loss": 0.9291, "lr": 8.133333333333334e-06, "epoch": 0.244, "percentage": 8.13, "elapsed_time": "1:54:33", "remaining_time": "21:33:59"}
+{"current_steps": 123, "total_steps": 1500, "loss": 0.905, "lr": 8.2e-06, "epoch": 0.246, "percentage": 8.2, "elapsed_time": "1:55:31", "remaining_time": "21:33:14"}
+{"current_steps": 124, "total_steps": 1500, "loss": 0.8933, "lr": 8.266666666666667e-06, "epoch": 0.248, "percentage": 8.27, "elapsed_time": "1:56:27", "remaining_time": "21:32:14"}
+{"current_steps": 125, "total_steps": 1500, "loss": 0.905, "lr": 8.333333333333334e-06, "epoch": 0.25, "percentage": 8.33, "elapsed_time": "1:57:31", "remaining_time": "21:32:43"}
+{"current_steps": 126, "total_steps": 1500, "loss": 0.8634, "lr": 8.400000000000001e-06, "epoch": 0.252, "percentage": 8.4, "elapsed_time": "1:58:19", "remaining_time": "21:30:15"}
+{"current_steps": 127, "total_steps": 1500, "loss": 0.8879, "lr": 8.466666666666668e-06, "epoch": 0.254, "percentage": 8.47, "elapsed_time": "1:59:15", "remaining_time": "21:29:19"}
+{"current_steps": 128, "total_steps": 1500, "loss": 0.8993, "lr": 8.533333333333335e-06, "epoch": 0.256, "percentage": 8.53, "elapsed_time": "2:00:16", "remaining_time": "21:29:08"}
+{"current_steps": 129, "total_steps": 1500, "loss": 0.8596, "lr": 8.6e-06, "epoch": 0.258, "percentage": 8.6, "elapsed_time": "2:01:03", "remaining_time": "21:26:31"}
+{"current_steps": 130, "total_steps": 1500, "loss": 0.8848, "lr": 8.666666666666668e-06, "epoch": 0.26, "percentage": 8.67, "elapsed_time": "2:01:54", "remaining_time": "21:24:47"}
+{"current_steps": 131, "total_steps": 1500, "loss": 0.8715, "lr": 8.733333333333333e-06, "epoch": 0.262, "percentage": 8.73, "elapsed_time": "2:02:52", "remaining_time": "21:24:05"}
+{"current_steps": 132, "total_steps": 1500, "loss": 0.898, "lr": 8.8e-06, "epoch": 0.264, "percentage": 8.8, "elapsed_time": "2:03:44", "remaining_time": "21:22:19"}
+{"current_steps": 133, "total_steps": 1500, "loss": 0.8785, "lr": 8.866666666666668e-06, "epoch": 0.266, "percentage": 8.87, "elapsed_time": "2:04:33", "remaining_time": "21:20:11"}
+{"current_steps": 134, "total_steps": 1500, "loss": 0.8904, "lr": 8.933333333333333e-06, "epoch": 0.268, "percentage": 8.93, "elapsed_time": "2:05:21", "remaining_time": "21:17:49"}
+{"current_steps": 135, "total_steps": 1500, "loss": 0.9241, "lr": 9e-06, "epoch": 0.27, "percentage": 9.0, "elapsed_time": "2:06:25", "remaining_time": "21:18:17"}
+{"current_steps": 136, "total_steps": 1500, "loss": 0.8931, "lr": 9.066666666666667e-06, "epoch": 0.272, "percentage": 9.07, "elapsed_time": "2:07:20", "remaining_time": "21:17:13"}
+{"current_steps": 137, "total_steps": 1500, "loss": 0.8773, "lr": 9.133333333333335e-06, "epoch": 0.274, "percentage": 9.13, "elapsed_time": "2:08:03", "remaining_time": "21:14:05"}
+{"current_steps": 138, "total_steps": 1500, "loss": 0.8808, "lr": 9.200000000000002e-06, "epoch": 0.276, "percentage": 9.2, "elapsed_time": "2:09:03", "remaining_time": "21:13:43"}
+{"current_steps": 139, "total_steps": 1500, "loss": 0.8811, "lr": 9.266666666666667e-06, "epoch": 0.278, "percentage": 9.27, "elapsed_time": "2:09:58", "remaining_time": "21:12:38"}
+{"current_steps": 140, "total_steps": 1500, "loss": 0.9266, "lr": 9.333333333333334e-06, "epoch": 0.28, "percentage": 9.33, "elapsed_time": "2:10:57", "remaining_time": "21:12:14"}
+{"current_steps": 141, "total_steps": 1500, "loss": 0.9108, "lr": 9.4e-06, "epoch": 0.282, "percentage": 9.4, "elapsed_time": "2:11:56", "remaining_time": "21:11:45"}
+{"current_steps": 142, "total_steps": 1500, "loss": 0.9115, "lr": 9.466666666666667e-06, "epoch": 0.284, "percentage": 9.47, "elapsed_time": "2:13:05", "remaining_time": "21:12:49"}
+{"current_steps": 143, "total_steps": 1500, "loss": 0.8877, "lr": 9.533333333333334e-06, "epoch": 0.286, "percentage": 9.53, "elapsed_time": "2:14:01", "remaining_time": "21:11:48"}
+{"current_steps": 144, "total_steps": 1500, "loss": 0.8841, "lr": 9.600000000000001e-06, "epoch": 0.288, "percentage": 9.6, "elapsed_time": "2:14:52", "remaining_time": "21:10:00"}
+{"current_steps": 145, "total_steps": 1500, "loss": 0.8631, "lr": 9.666666666666667e-06, "epoch": 0.29, "percentage": 9.67, "elapsed_time": "2:15:41", "remaining_time": "21:08:01"}
+{"current_steps": 146, "total_steps": 1500, "loss": 0.8906, "lr": 9.733333333333334e-06, "epoch": 0.292, "percentage": 9.73, "elapsed_time": "2:16:40", "remaining_time": "21:07:30"}
+{"current_steps": 147, "total_steps": 1500, "loss": 0.8899, "lr": 9.800000000000001e-06, "epoch": 0.294, "percentage": 9.8, "elapsed_time": "2:17:32", "remaining_time": "21:05:55"}
+{"current_steps": 148, "total_steps": 1500, "loss": 0.9043, "lr": 9.866666666666668e-06, "epoch": 0.296, "percentage": 9.87, "elapsed_time": "2:18:30", "remaining_time": "21:05:20"}
+{"current_steps": 149, "total_steps": 1500, "loss": 0.8981, "lr": 9.933333333333334e-06, "epoch": 0.298, "percentage": 9.93, "elapsed_time": "2:19:29", "remaining_time": "21:04:43"}
+{"current_steps": 150, "total_steps": 1500, "loss": 0.8739, "lr": 1e-05, "epoch": 0.3, "percentage": 10.0, "elapsed_time": "2:20:20", "remaining_time": "21:03:02"}
+{"current_steps": 151, "total_steps": 1500, "loss": 0.8607, "lr": 9.99998646145412e-06, "epoch": 0.302, "percentage": 10.07, "elapsed_time": "2:21:13", "remaining_time": "21:01:38"}
+{"current_steps": 152, "total_steps": 1500, "loss": 0.8963, "lr": 9.999945845889795e-06, "epoch": 0.304, "percentage": 10.13, "elapsed_time": "2:22:11", "remaining_time": "21:01:00"}
+{"current_steps": 153, "total_steps": 1500, "loss": 0.8965, "lr": 9.999878153526974e-06, "epoch": 0.306, "percentage": 10.2, "elapsed_time": "2:23:12", "remaining_time": "21:00:47"}
+{"current_steps": 154, "total_steps": 1500, "loss": 0.902, "lr": 9.999783384732242e-06, "epoch": 0.308, "percentage": 10.27, "elapsed_time": "2:24:11", "remaining_time": "21:00:15"}
+{"current_steps": 155, "total_steps": 1500, "loss": 0.8598, "lr": 9.999661540018812e-06, "epoch": 0.31, "percentage": 10.33, "elapsed_time": "2:25:13", "remaining_time": "21:00:14"}
+{"current_steps": 156, "total_steps": 1500, "loss": 0.9204, "lr": 9.999512620046523e-06, "epoch": 0.312, "percentage": 10.4, "elapsed_time": "2:26:12", "remaining_time": "20:59:36"}
+{"current_steps": 157, "total_steps": 1500, "loss": 0.8934, "lr": 9.999336625621836e-06, "epoch": 0.314, "percentage": 10.47, "elapsed_time": "2:27:05", "remaining_time": "20:58:16"}
+{"current_steps": 158, "total_steps": 1500, "loss": 0.8633, "lr": 9.99913355769784e-06, "epoch": 0.316, "percentage": 10.53, "elapsed_time": "2:27:51", "remaining_time": "20:55:52"}
+{"current_steps": 159, "total_steps": 1500, "loss": 0.894, "lr": 9.998903417374228e-06, "epoch": 0.318, "percentage": 10.6, "elapsed_time": "2:28:48", "remaining_time": "20:55:06"}
+{"current_steps": 160, "total_steps": 1500, "loss": 0.9034, "lr": 9.99864620589731e-06, "epoch": 0.32, "percentage": 10.67, "elapsed_time": "2:29:42", "remaining_time": "20:53:50"}
+{"current_steps": 161, "total_steps": 1500, "loss": 0.9033, "lr": 9.998361924659989e-06, "epoch": 0.322, "percentage": 10.73, "elapsed_time": "2:30:35", "remaining_time": "20:52:26"}
+{"current_steps": 162, "total_steps": 1500, "loss": 0.8633, "lr": 9.998050575201772e-06, "epoch": 0.324, "percentage": 10.8, "elapsed_time": "2:31:30", "remaining_time": "20:51:22"}
+{"current_steps": 163, "total_steps": 1500, "loss": 0.8819, "lr": 9.997712159208745e-06, "epoch": 0.326, "percentage": 10.87, "elapsed_time": "2:32:31", "remaining_time": "20:51:06"}
+{"current_steps": 164, "total_steps": 1500, "loss": 0.8876, "lr": 9.99734667851357e-06, "epoch": 0.328, "percentage": 10.93, "elapsed_time": "2:33:30", "remaining_time": "20:50:31"}
+{"current_steps": 165, "total_steps": 1500, "loss": 0.8782, "lr": 9.99695413509548e-06, "epoch": 0.33, "percentage": 11.0, "elapsed_time": "2:34:21", "remaining_time": "20:48:55"}
+{"current_steps": 166, "total_steps": 1500, "loss": 0.8554, "lr": 9.99653453108026e-06, "epoch": 0.332, "percentage": 11.07, "elapsed_time": "2:35:14", "remaining_time": "20:47:34"}
+{"current_steps": 167, "total_steps": 1500, "loss": 0.8662, "lr": 9.996087868740244e-06, "epoch": 0.334, "percentage": 11.13, "elapsed_time": "2:36:09", "remaining_time": "20:46:25"}
+{"current_steps": 168, "total_steps": 1500, "loss": 0.9093, "lr": 9.995614150494293e-06, "epoch": 0.336, "percentage": 11.2, "elapsed_time": "2:37:04", "remaining_time": "20:45:23"}
+{"current_steps": 169, "total_steps": 1500, "loss": 0.8916, "lr": 9.995113378907791e-06, "epoch": 0.338, "percentage": 11.27, "elapsed_time": "2:38:06", "remaining_time": "20:45:11"}
+{"current_steps": 170, "total_steps": 1500, "loss": 0.8586, "lr": 9.994585556692624e-06, "epoch": 0.34, "percentage": 11.33, "elapsed_time": "2:38:56", "remaining_time": "20:43:25"}
+{"current_steps": 171, "total_steps": 1500, "loss": 0.8919, "lr": 9.994030686707171e-06, "epoch": 0.342, "percentage": 11.4, "elapsed_time": "2:39:55", "remaining_time": "20:42:52"}
+{"current_steps": 172, "total_steps": 1500, "loss": 0.9085, "lr": 9.993448771956285e-06, "epoch": 0.344, "percentage": 11.47, "elapsed_time": "2:40:55", "remaining_time": "20:42:26"}
+{"current_steps": 173, "total_steps": 1500, "loss": 0.8941, "lr": 9.99283981559128e-06, "epoch": 0.346, "percentage": 11.53, "elapsed_time": "2:41:45", "remaining_time": "20:40:42"}
+{"current_steps": 174, "total_steps": 1500, "loss": 0.8901, "lr": 9.992203820909906e-06, "epoch": 0.348, "percentage": 11.6, "elapsed_time": "2:42:29", "remaining_time": "20:38:16"}
+{"current_steps": 175, "total_steps": 1500, "loss": 0.8806, "lr": 9.991540791356342e-06, "epoch": 0.35, "percentage": 11.67, "elapsed_time": "2:43:25", "remaining_time": "20:37:23"}
+{"current_steps": 176, "total_steps": 1500, "loss": 0.8689, "lr": 9.99085073052117e-06, "epoch": 0.352, "percentage": 11.73, "elapsed_time": "2:44:21", "remaining_time": "20:36:26"}
+{"current_steps": 177, "total_steps": 1500, "loss": 0.9046, "lr": 9.990133642141359e-06, "epoch": 0.354, "percentage": 11.8, "elapsed_time": "2:45:21", "remaining_time": "20:35:55"}
+{"current_steps": 178, "total_steps": 1500, "loss": 0.9127, "lr": 9.989389530100242e-06, "epoch": 0.356, "percentage": 11.87, "elapsed_time": "2:46:17", "remaining_time": "20:34:59"}
+{"current_steps": 179, "total_steps": 1500, "loss": 0.8533, "lr": 9.988618398427495e-06, "epoch": 0.358, "percentage": 11.93, "elapsed_time": "2:47:09", "remaining_time": "20:33:39"}
+{"current_steps": 180, "total_steps": 1500, "loss": 0.8705, "lr": 9.987820251299121e-06, "epoch": 0.36, "percentage": 12.0, "elapsed_time": "2:48:04", "remaining_time": "20:32:29"}
+{"current_steps": 181, "total_steps": 1500, "loss": 0.8946, "lr": 9.986995093037422e-06, "epoch": 0.362, "percentage": 12.07, "elapsed_time": "2:49:04", "remaining_time": "20:32:06"}
+{"current_steps": 182, "total_steps": 1500, "loss": 0.8988, "lr": 9.986142928110972e-06, "epoch": 0.364, "percentage": 12.13, "elapsed_time": "2:50:01", "remaining_time": "20:31:16"}
+{"current_steps": 183, "total_steps": 1500, "loss": 0.875, "lr": 9.985263761134602e-06, "epoch": 0.366, "percentage": 12.2, "elapsed_time": "2:50:59", "remaining_time": "20:30:33"}
+{"current_steps": 184, "total_steps": 1500, "loss": 0.8696, "lr": 9.984357596869369e-06, "epoch": 0.368, "percentage": 12.27, "elapsed_time": "2:51:49", "remaining_time": "20:28:52"}
+{"current_steps": 185, "total_steps": 1500, "loss": 0.9005, "lr": 9.98342444022253e-06, "epoch": 0.37, "percentage": 12.33, "elapsed_time": "2:52:50", "remaining_time": "20:28:31"}
+{"current_steps": 186, "total_steps": 1500, "loss": 0.8743, "lr": 9.982464296247523e-06, "epoch": 0.372, "percentage": 12.4, "elapsed_time": "2:53:43", "remaining_time": "20:27:19"}
+{"current_steps": 187, "total_steps": 1500, "loss": 0.8793, "lr": 9.981477170143924e-06, "epoch": 0.374, "percentage": 12.47, "elapsed_time": "2:54:45", "remaining_time": "20:26:59"}
+{"current_steps": 188, "total_steps": 1500, "loss": 0.8982, "lr": 9.980463067257437e-06, "epoch": 0.376, "percentage": 12.53, "elapsed_time": "2:55:39", "remaining_time": "20:25:51"}
+{"current_steps": 189, "total_steps": 1500, "loss": 0.8327, "lr": 9.979421993079853e-06, "epoch": 0.378, "percentage": 12.6, "elapsed_time": "2:56:21", "remaining_time": "20:23:19"}
+{"current_steps": 190, "total_steps": 1500, "loss": 0.8653, "lr": 9.978353953249023e-06, "epoch": 0.38, "percentage": 12.67, "elapsed_time": "2:57:13", "remaining_time": "20:21:56"}
+{"current_steps": 191, "total_steps": 1500, "loss": 0.8479, "lr": 9.977258953548831e-06, "epoch": 0.382, "percentage": 12.73, "elapsed_time": "2:58:02", "remaining_time": "20:20:08"}
+{"current_steps": 192, "total_steps": 1500, "loss": 0.8801, "lr": 9.976136999909156e-06, "epoch": 0.384, "percentage": 12.8, "elapsed_time": "2:58:57", "remaining_time": "20:19:09"}
+{"current_steps": 193, "total_steps": 1500, "loss": 0.8703, "lr": 9.97498809840585e-06, "epoch": 0.386, "percentage": 12.87, "elapsed_time": "2:59:48", "remaining_time": "20:17:38"}
+{"current_steps": 194, "total_steps": 1500, "loss": 0.8566, "lr": 9.973812255260692e-06, "epoch": 0.388, "percentage": 12.93, "elapsed_time": "3:00:44", "remaining_time": "20:16:44"}
+{"current_steps": 195, "total_steps": 1500, "loss": 0.9185, "lr": 9.972609476841368e-06, "epoch": 0.39, "percentage": 13.0, "elapsed_time": "3:01:42", "remaining_time": "20:16:03"}
+{"current_steps": 196, "total_steps": 1500, "loss": 0.878, "lr": 9.971379769661422e-06, "epoch": 0.392, "percentage": 13.07, "elapsed_time": "3:02:33", "remaining_time": "20:14:37"}
+{"current_steps": 197, "total_steps": 1500, "loss": 0.91, "lr": 9.970123140380237e-06, "epoch": 0.394, "percentage": 13.13, "elapsed_time": "3:03:35", "remaining_time": "20:14:21"}
+{"current_steps": 198, "total_steps": 1500, "loss": 0.8952, "lr": 9.968839595802982e-06, "epoch": 0.396, "percentage": 13.2, "elapsed_time": "3:04:33", "remaining_time": "20:13:33"}
+{"current_steps": 199, "total_steps": 1500, "loss": 0.9048, "lr": 9.967529142880592e-06, "epoch": 0.398, "percentage": 13.27, "elapsed_time": "3:05:36", "remaining_time": "20:13:26"}
+{"current_steps": 200, "total_steps": 1500, "loss": 0.9067, "lr": 9.966191788709716e-06, "epoch": 0.4, "percentage": 13.33, "elapsed_time": "3:06:42", "remaining_time": "20:13:36"}
+{"current_steps": 201, "total_steps": 1500, "loss": 0.8992, "lr": 9.964827540532685e-06, "epoch": 0.402, "percentage": 13.4, "elapsed_time": "3:07:45", "remaining_time": "20:13:23"}
+{"current_steps": 202, "total_steps": 1500, "loss": 0.8849, "lr": 9.963436405737476e-06, "epoch": 0.404, "percentage": 13.47, "elapsed_time": "3:08:40", "remaining_time": "20:12:23"}
+{"current_steps": 203, "total_steps": 1500, "loss": 0.88, "lr": 9.962018391857665e-06, "epoch": 0.406, "percentage": 13.53, "elapsed_time": "3:09:32", "remaining_time": "20:10:59"}
+{"current_steps": 204, "total_steps": 1500, "loss": 0.9101, "lr": 9.960573506572391e-06, "epoch": 0.408, "percentage": 13.6, "elapsed_time": "3:10:41", "remaining_time": "20:11:29"}
+{"current_steps": 205, "total_steps": 1500, "loss": 0.9005, "lr": 9.959101757706308e-06, "epoch": 0.41, "percentage": 13.67, "elapsed_time": "3:11:45", "remaining_time": "20:11:20"}
+{"current_steps": 206, "total_steps": 1500, "loss": 0.8975, "lr": 9.957603153229559e-06, "epoch": 0.412, "percentage": 13.73, "elapsed_time": "3:12:41", "remaining_time": "20:10:21"}
+{"current_steps": 207, "total_steps": 1500, "loss": 0.8636, "lr": 9.95607770125771e-06, "epoch": 0.414, "percentage": 13.8, "elapsed_time": "3:13:41", "remaining_time": "20:09:50"}
+{"current_steps": 208, "total_steps": 1500, "loss": 0.866, "lr": 9.95452541005172e-06, "epoch": 0.416, "percentage": 13.87, "elapsed_time": "3:14:28", "remaining_time": "20:08:00"}
+{"current_steps": 209, "total_steps": 1500, "loss": 0.8999, "lr": 9.952946288017899e-06, "epoch": 0.418, "percentage": 13.93, "elapsed_time": "3:15:27", "remaining_time": "20:07:23"}
+{"current_steps": 210, "total_steps": 1500, "loss": 0.9092, "lr": 9.951340343707852e-06, "epoch": 0.42, "percentage": 14.0, "elapsed_time": "3:16:32", "remaining_time": "20:07:17"}
+{"current_steps": 211, "total_steps": 1500, "loss": 0.8728, "lr": 9.94970758581844e-06, "epoch": 0.422, "percentage": 14.07, "elapsed_time": "3:17:28", "remaining_time": "20:06:20"}
+{"current_steps": 212, "total_steps": 1500, "loss": 0.8731, "lr": 9.948048023191728e-06, "epoch": 0.424, "percentage": 14.13, "elapsed_time": "3:18:13", "remaining_time": "20:04:16"}
+{"current_steps": 213, "total_steps": 1500, "loss": 0.9037, "lr": 9.946361664814942e-06, "epoch": 0.426, "percentage": 14.2, "elapsed_time": "3:19:15", "remaining_time": "20:03:58"}
+{"current_steps": 214, "total_steps": 1500, "loss": 0.8822, "lr": 9.94464851982042e-06, "epoch": 0.428, "percentage": 14.27, "elapsed_time": "3:20:09", "remaining_time": "20:02:50"}
+{"current_steps": 215, "total_steps": 1500, "loss": 0.8675, "lr": 9.942908597485558e-06, "epoch": 0.43, "percentage": 14.33, "elapsed_time": "3:21:12", "remaining_time": "20:02:33"}
+{"current_steps": 216, "total_steps": 1500, "loss": 0.9004, "lr": 9.941141907232766e-06, "epoch": 0.432, "percentage": 14.4, "elapsed_time": "3:22:21", "remaining_time": "20:02:55"}
+{"current_steps": 217, "total_steps": 1500, "loss": 0.8943, "lr": 9.939348458629406e-06, "epoch": 0.434, "percentage": 14.47, "elapsed_time": "3:23:25", "remaining_time": "20:02:42"}
+{"current_steps": 218, "total_steps": 1500, "loss": 0.902, "lr": 9.937528261387753e-06, "epoch": 0.436, "percentage": 14.53, "elapsed_time": "3:24:18", "remaining_time": "20:01:27"}
+{"current_steps": 219, "total_steps": 1500, "loss": 0.8586, "lr": 9.93568132536494e-06, "epoch": 0.438, "percentage": 14.6, "elapsed_time": "3:25:08", "remaining_time": "19:59:57"}
+{"current_steps": 220, "total_steps": 1500, "loss": 0.8662, "lr": 9.933807660562898e-06, "epoch": 0.44, "percentage": 14.67, "elapsed_time": "3:26:02", "remaining_time": "19:58:49"}
+{"current_steps": 221, "total_steps": 1500, "loss": 0.8689, "lr": 9.9319072771283e-06, "epoch": 0.442, "percentage": 14.73, "elapsed_time": "3:26:52", "remaining_time": "19:57:17"}
+{"current_steps": 222, "total_steps": 1500, "loss": 0.8736, "lr": 9.929980185352525e-06, "epoch": 0.444, "percentage": 14.8, "elapsed_time": "3:27:52", "remaining_time": "19:56:41"}
+{"current_steps": 223, "total_steps": 1500, "loss": 0.8675, "lr": 9.928026395671577e-06, "epoch": 0.446, "percentage": 14.87, "elapsed_time": "3:28:52", "remaining_time": "19:56:07"}
+{"current_steps": 224, "total_steps": 1500, "loss": 0.8943, "lr": 9.926045918666045e-06, "epoch": 0.448, "percentage": 14.93, "elapsed_time": "3:29:50", "remaining_time": "19:55:20"}
+{"current_steps": 225, "total_steps": 1500, "loss": 0.8695, "lr": 9.924038765061042e-06, "epoch": 0.45, "percentage": 15.0, "elapsed_time": "3:30:47", "remaining_time": "19:54:29"}
+{"current_steps": 226, "total_steps": 1500, "loss": 0.9111, "lr": 9.92200494572614e-06, "epoch": 0.452, "percentage": 15.07, "elapsed_time": "3:31:38", "remaining_time": "19:53:04"}
+{"current_steps": 227, "total_steps": 1500, "loss": 0.925, "lr": 9.919944471675328e-06, "epoch": 0.454, "percentage": 15.13, "elapsed_time": "3:32:35", "remaining_time": "19:52:12"}
+{"current_steps": 228, "total_steps": 1500, "loss": 0.8782, "lr": 9.91785735406693e-06, "epoch": 0.456, "percentage": 15.2, "elapsed_time": "3:33:41", "remaining_time": "19:52:10"}
+{"current_steps": 229, "total_steps": 1500, "loss": 0.8742, "lr": 9.915743604203563e-06, "epoch": 0.458, "percentage": 15.27, "elapsed_time": "3:34:43", "remaining_time": "19:51:46"}
+{"current_steps": 230, "total_steps": 1500, "loss": 0.8671, "lr": 9.913603233532067e-06, "epoch": 0.46, "percentage": 15.33, "elapsed_time": "3:35:31", "remaining_time": "19:50:05"}
+{"current_steps": 231, "total_steps": 1500, "loss": 0.8656, "lr": 9.911436253643445e-06, "epoch": 0.462, "percentage": 15.4, "elapsed_time": "3:36:25", "remaining_time": "19:48:58"}
+{"current_steps": 232, "total_steps": 1500, "loss": 0.8761, "lr": 9.909242676272797e-06, "epoch": 0.464, "percentage": 15.47, "elapsed_time": "3:37:20", "remaining_time": "19:47:51"}
+{"current_steps": 233, "total_steps": 1500, "loss": 0.8914, "lr": 9.907022513299264e-06, "epoch": 0.466, "percentage": 15.53, "elapsed_time": "3:38:12", "remaining_time": "19:46:34"}
+{"current_steps": 234, "total_steps": 1500, "loss": 0.8499, "lr": 9.904775776745959e-06, "epoch": 0.468, "percentage": 15.6, "elapsed_time": "3:39:02", "remaining_time": "19:45:04"}
+{"current_steps": 235, "total_steps": 1500, "loss": 0.849, "lr": 9.902502478779897e-06, "epoch": 0.47, "percentage": 15.67, "elapsed_time": "3:39:51", "remaining_time": "19:43:30"}
+{"current_steps": 236, "total_steps": 1500, "loss": 0.935, "lr": 9.90020263171194e-06, "epoch": 0.472, "percentage": 15.73, "elapsed_time": "3:40:55", "remaining_time": "19:43:17"}
+{"current_steps": 237, "total_steps": 1500, "loss": 0.8422, "lr": 9.89787624799672e-06, "epoch": 0.474, "percentage": 15.8, "elapsed_time": "3:41:55", "remaining_time": "19:42:38"}
+{"current_steps": 238, "total_steps": 1500, "loss": 0.8999, "lr": 9.89552334023258e-06, "epoch": 0.476, "percentage": 15.87, "elapsed_time": "3:42:55", "remaining_time": "19:42:06"}
+{"current_steps": 239, "total_steps": 1500, "loss": 0.849, "lr": 9.893143921161501e-06, "epoch": 0.478, "percentage": 15.93, "elapsed_time": "3:43:50", "remaining_time": "19:41:02"}
+{"current_steps": 240, "total_steps": 1500, "loss": 0.8743, "lr": 9.890738003669029e-06, "epoch": 0.48, "percentage": 16.0, "elapsed_time": "3:44:48", "remaining_time": "19:40:12"}
+{"current_steps": 241, "total_steps": 1500, "loss": 0.8796, "lr": 9.888305600784217e-06, "epoch": 0.482, "percentage": 16.07, "elapsed_time": "3:45:46", "remaining_time": "19:39:28"}
+{"current_steps": 242, "total_steps": 1500, "loss": 0.8423, "lr": 9.88584672567954e-06, "epoch": 0.484, "percentage": 16.13, "elapsed_time": "3:46:33", "remaining_time": "19:37:44"}
+{"current_steps": 243, "total_steps": 1500, "loss": 0.9044, "lr": 9.883361391670841e-06, "epoch": 0.486, "percentage": 16.2, "elapsed_time": "3:47:27", "remaining_time": "19:36:33"}
+{"current_steps": 244, "total_steps": 1500, "loss": 0.8751, "lr": 9.880849612217238e-06, "epoch": 0.488, "percentage": 16.27, "elapsed_time": "3:48:28", "remaining_time": "19:36:04"}
+{"current_steps": 245, "total_steps": 1500, "loss": 0.8748, "lr": 9.878311400921072e-06, "epoch": 0.49, "percentage": 16.33, "elapsed_time": "3:49:18", "remaining_time": "19:34:39"}
+{"current_steps": 246, "total_steps": 1500, "loss": 0.8951, "lr": 9.875746771527817e-06, "epoch": 0.492, "percentage": 16.4, "elapsed_time": "3:50:20", "remaining_time": "19:34:11"}
+{"current_steps": 247, "total_steps": 1500, "loss": 0.8619, "lr": 9.873155737926014e-06, "epoch": 0.494, "percentage": 16.47, "elapsed_time": "3:51:07", "remaining_time": "19:32:29"}
+{"current_steps": 248, "total_steps": 1500, "loss": 0.8861, "lr": 9.870538314147194e-06, "epoch": 0.496, "percentage": 16.53, "elapsed_time": "3:52:02", "remaining_time": "19:31:26"}
+{"current_steps": 249, "total_steps": 1500, "loss": 0.8435, "lr": 9.867894514365802e-06, "epoch": 0.498, "percentage": 16.6, "elapsed_time": "3:53:01", "remaining_time": "19:30:46"}
+{"current_steps": 250, "total_steps": 1500, "loss": 0.847, "lr": 9.86522435289912e-06, "epoch": 0.5, "percentage": 16.67, "elapsed_time": "3:53:53", "remaining_time": "19:29:27"}
+{"current_steps": 251, "total_steps": 1500, "loss": 0.9026, "lr": 9.862527844207189e-06, "epoch": 0.502, "percentage": 16.73, "elapsed_time": "3:55:00", "remaining_time": "19:29:22"}
+{"current_steps": 252, "total_steps": 1500, "loss": 0.8963, "lr": 9.859805002892733e-06, "epoch": 0.504, "percentage": 16.8, "elapsed_time": "3:56:03", "remaining_time": "19:29:03"}
+{"current_steps": 253, "total_steps": 1500, "loss": 0.8749, "lr": 9.857055843701073e-06, "epoch": 0.506, "percentage": 16.87, "elapsed_time": "3:57:04", "remaining_time": "19:28:31"}
+{"current_steps": 254, "total_steps": 1500, "loss": 0.8652, "lr": 9.85428038152006e-06, "epoch": 0.508, "percentage": 16.93, "elapsed_time": "3:58:00", "remaining_time": "19:27:35"}
+{"current_steps": 255, "total_steps": 1500, "loss": 0.8486, "lr": 9.851478631379982e-06, "epoch": 0.51, "percentage": 17.0, "elapsed_time": "3:58:51", "remaining_time": "19:26:13"}
+{"current_steps": 256, "total_steps": 1500, "loss": 0.8874, "lr": 9.84865060845349e-06, "epoch": 0.512, "percentage": 17.07, "elapsed_time": "3:59:49", "remaining_time": "19:25:22"}
+{"current_steps": 257, "total_steps": 1500, "loss": 0.8516, "lr": 9.845796328055505e-06, "epoch": 0.514, "percentage": 17.13, "elapsed_time": "4:00:45", "remaining_time": "19:24:26"}
+{"current_steps": 258, "total_steps": 1500, "loss": 0.8895, "lr": 9.842915805643156e-06, "epoch": 0.516, "percentage": 17.2, "elapsed_time": "4:01:40", "remaining_time": "19:23:25"}
+{"current_steps": 259, "total_steps": 1500, "loss": 0.8975, "lr": 9.840009056815674e-06, "epoch": 0.518, "percentage": 17.27, "elapsed_time": "4:02:32", "remaining_time": "19:22:06"}
+{"current_steps": 260, "total_steps": 1500, "loss": 0.85, "lr": 9.83707609731432e-06, "epoch": 0.52, "percentage": 17.33, "elapsed_time": "4:03:21", "remaining_time": "19:20:36"}
+{"current_steps": 261, "total_steps": 1500, "loss": 0.8465, "lr": 9.834116943022299e-06, "epoch": 0.522, "percentage": 17.4, "elapsed_time": "4:04:09", "remaining_time": "19:19:02"}
+{"current_steps": 262, "total_steps": 1500, "loss": 0.858, "lr": 9.831131609964664e-06, "epoch": 0.524, "percentage": 17.47, "elapsed_time": "4:05:13", "remaining_time": "19:18:41"}
+{"current_steps": 263, "total_steps": 1500, "loss": 0.9133, "lr": 9.828120114308248e-06, "epoch": 0.526, "percentage": 17.53, "elapsed_time": "4:06:09", "remaining_time": "19:17:48"}
+{"current_steps": 264, "total_steps": 1500, "loss": 0.8991, "lr": 9.825082472361558e-06, "epoch": 0.528, "percentage": 17.6, "elapsed_time": "4:07:17", "remaining_time": "19:17:45"}
+{"current_steps": 265, "total_steps": 1500, "loss": 0.875, "lr": 9.822018700574696e-06, "epoch": 0.53, "percentage": 17.67, "elapsed_time": "4:08:18", "remaining_time": "19:17:12"}
+{"current_steps": 266, "total_steps": 1500, "loss": 0.8935, "lr": 9.818928815539266e-06, "epoch": 0.532, "percentage": 17.73, "elapsed_time": "4:09:14", "remaining_time": "19:16:14"}
+{"current_steps": 267, "total_steps": 1500, "loss": 0.8636, "lr": 9.815812833988292e-06, "epoch": 0.534, "percentage": 17.8, "elapsed_time": "4:10:04", "remaining_time": "19:14:48"}
+{"current_steps": 268, "total_steps": 1500, "loss": 0.8796, "lr": 9.812670772796113e-06, "epoch": 0.536, "percentage": 17.87, "elapsed_time": "4:10:56", "remaining_time": "19:13:34"}
+{"current_steps": 269, "total_steps": 1500, "loss": 0.8507, "lr": 9.809502648978311e-06, "epoch": 0.538, "percentage": 17.93, "elapsed_time": "4:11:52", "remaining_time": "19:12:40"}
+{"current_steps": 270, "total_steps": 1500, "loss": 0.8668, "lr": 9.806308479691595e-06, "epoch": 0.54, "percentage": 18.0, "elapsed_time": "4:12:58", "remaining_time": "19:12:24"}
+{"current_steps": 271, "total_steps": 1500, "loss": 0.8753, "lr": 9.803088282233733e-06, "epoch": 0.542, "percentage": 18.07, "elapsed_time": "4:14:00", "remaining_time": "19:11:58"}
+{"current_steps": 272, "total_steps": 1500, "loss": 0.888, "lr": 9.799842074043438e-06, "epoch": 0.544, "percentage": 18.13, "elapsed_time": "4:14:53", "remaining_time": "19:10:45"}
+{"current_steps": 273, "total_steps": 1500, "loss": 0.8614, "lr": 9.796569872700287e-06, "epoch": 0.546, "percentage": 18.2, "elapsed_time": "4:15:46", "remaining_time": "19:09:34"}
+{"current_steps": 274, "total_steps": 1500, "loss": 0.8587, "lr": 9.793271695924621e-06, "epoch": 0.548, "percentage": 18.27, "elapsed_time": "4:16:45", "remaining_time": "19:08:52"}
+{"current_steps": 275, "total_steps": 1500, "loss": 0.8382, "lr": 9.789947561577445e-06, "epoch": 0.55, "percentage": 18.33, "elapsed_time": "4:17:32", "remaining_time": "19:07:15"}
+{"current_steps": 276, "total_steps": 1500, "loss": 0.8866, "lr": 9.786597487660336e-06, "epoch": 0.552, "percentage": 18.4, "elapsed_time": "4:18:39", "remaining_time": "19:07:03"}
+{"current_steps": 277, "total_steps": 1500, "loss": 0.8749, "lr": 9.78322149231535e-06, "epoch": 0.554, "percentage": 18.47, "elapsed_time": "4:19:40", "remaining_time": "19:06:30"}
+{"current_steps": 278, "total_steps": 1500, "loss": 0.8504, "lr": 9.779819593824909e-06, "epoch": 0.556, "percentage": 18.53, "elapsed_time": "4:20:52", "remaining_time": "19:06:41"}
+{"current_steps": 279, "total_steps": 1500, "loss": 0.8585, "lr": 9.776391810611719e-06, "epoch": 0.558, "percentage": 18.6, "elapsed_time": "4:21:46", "remaining_time": "19:05:38"}
+{"current_steps": 280, "total_steps": 1500, "loss": 0.8724, "lr": 9.77293816123866e-06, "epoch": 0.56, "percentage": 18.67, "elapsed_time": "4:22:49", "remaining_time": "19:05:10"}
+{"current_steps": 281, "total_steps": 1500, "loss": 0.8534, "lr": 9.769458664408689e-06, "epoch": 0.562, "percentage": 18.73, "elapsed_time": "4:23:54", "remaining_time": "19:04:49"}
+{"current_steps": 282, "total_steps": 1500, "loss": 0.8765, "lr": 9.765953338964736e-06, "epoch": 0.564, "percentage": 18.8, "elapsed_time": "4:24:47", "remaining_time": "19:03:39"}
+{"current_steps": 283, "total_steps": 1500, "loss": 0.852, "lr": 9.762422203889604e-06, "epoch": 0.566, "percentage": 18.87, "elapsed_time": "4:25:49", "remaining_time": "19:03:06"}
+{"current_steps": 284, "total_steps": 1500, "loss": 0.8427, "lr": 9.75886527830587e-06, "epoch": 0.568, "percentage": 18.93, "elapsed_time": "4:26:38", "remaining_time": "19:01:41"}
+{"current_steps": 285, "total_steps": 1500, "loss": 0.8853, "lr": 9.755282581475769e-06, "epoch": 0.57, "percentage": 19.0, "elapsed_time": "4:27:39", "remaining_time": "19:01:02"}
+{"current_steps": 286, "total_steps": 1500, "loss": 0.8583, "lr": 9.751674132801106e-06, "epoch": 0.572, "percentage": 19.07, "elapsed_time": "4:28:32", "remaining_time": "18:59:54"}
+{"current_steps": 287, "total_steps": 1500, "loss": 0.8584, "lr": 9.748039951823141e-06, "epoch": 0.574, "percentage": 19.13, "elapsed_time": "4:29:23", "remaining_time": "18:58:33"}
+{"current_steps": 288, "total_steps": 1500, "loss": 0.8574, "lr": 9.744380058222483e-06, "epoch": 0.576, "percentage": 19.2, "elapsed_time": "4:30:12", "remaining_time": "18:57:09"}
+{"current_steps": 289, "total_steps": 1500, "loss": 0.8507, "lr": 9.740694471818988e-06, "epoch": 0.578, "percentage": 19.27, "elapsed_time": "4:31:00", "remaining_time": "18:55:36"}
+{"current_steps": 290, "total_steps": 1500, "loss": 0.8638, "lr": 9.736983212571646e-06, "epoch": 0.58, "percentage": 19.33, "elapsed_time": "4:31:55", "remaining_time": "18:54:34"}
+{"current_steps": 291, "total_steps": 1500, "loss": 0.8778, "lr": 9.733246300578482e-06, "epoch": 0.582, "percentage": 19.4, "elapsed_time": "4:32:51", "remaining_time": "18:53:36"}
+{"current_steps": 292, "total_steps": 1500, "loss": 0.8586, "lr": 9.729483756076436e-06, "epoch": 0.584, "percentage": 19.47, "elapsed_time": "4:33:46", "remaining_time": "18:52:36"}
+{"current_steps": 293, "total_steps": 1500, "loss": 0.8799, "lr": 9.72569559944126e-06, "epoch": 0.586, "percentage": 19.53, "elapsed_time": "4:34:47", "remaining_time": "18:51:58"}
+{"current_steps": 294, "total_steps": 1500, "loss": 0.8669, "lr": 9.721881851187406e-06, "epoch": 0.588, "percentage": 19.6, "elapsed_time": "4:35:42", "remaining_time": "18:50:59"}
+{"current_steps": 295, "total_steps": 1500, "loss": 0.8898, "lr": 9.718042531967918e-06, "epoch": 0.59, "percentage": 19.67, "elapsed_time": "4:36:41", "remaining_time": "18:50:12"}
+{"current_steps": 296, "total_steps": 1500, "loss": 0.8883, "lr": 9.714177662574316e-06, "epoch": 0.592, "percentage": 19.73, "elapsed_time": "4:37:45", "remaining_time": "18:49:49"}
+{"current_steps": 297, "total_steps": 1500, "loss": 0.864, "lr": 9.710287263936485e-06, "epoch": 0.594, "percentage": 19.8, "elapsed_time": "4:38:43", "remaining_time": "18:48:56"}
+{"current_steps": 298, "total_steps": 1500, "loss": 0.8742, "lr": 9.70637135712256e-06, "epoch": 0.596, "percentage": 19.87, "elapsed_time": "4:39:35", "remaining_time": "18:47:43"}
+{"current_steps": 299, "total_steps": 1500, "loss": 0.8753, "lr": 9.702429963338812e-06, "epoch": 0.598, "percentage": 19.93, "elapsed_time": "4:40:31", "remaining_time": "18:46:49"}
+{"current_steps": 300, "total_steps": 1500, "loss": 0.8638, "lr": 9.698463103929542e-06, "epoch": 0.6, "percentage": 20.0, "elapsed_time": "4:41:26", "remaining_time": "18:45:47"}
+{"current_steps": 301, "total_steps": 1500, "loss": 0.8452, "lr": 9.694470800376951e-06, "epoch": 0.602, "percentage": 20.07, "elapsed_time": "4:42:22", "remaining_time": "18:44:47"}
+{"current_steps": 302, "total_steps": 1500, "loss": 0.865, "lr": 9.690453074301035e-06, "epoch": 0.604, "percentage": 20.13, "elapsed_time": "4:43:17", "remaining_time": "18:43:48"}
+{"current_steps": 303, "total_steps": 1500, "loss": 0.8554, "lr": 9.68640994745946e-06, "epoch": 0.606, "percentage": 20.2, "elapsed_time": "4:44:11", "remaining_time": "18:42:42"}
+{"current_steps": 304, "total_steps": 1500, "loss": 0.9091, "lr": 9.682341441747446e-06, "epoch": 0.608, "percentage": 20.27, "elapsed_time": "4:45:12", "remaining_time": "18:42:04"}
+{"current_steps": 305, "total_steps": 1500, "loss": 0.8649, "lr": 9.678247579197658e-06, "epoch": 0.61, "percentage": 20.33, "elapsed_time": "4:46:08", "remaining_time": "18:41:06"}
+{"current_steps": 306, "total_steps": 1500, "loss": 0.8565, "lr": 9.674128381980073e-06, "epoch": 0.612, "percentage": 20.4, "elapsed_time": "4:47:04", "remaining_time": "18:40:09"}
+{"current_steps": 307, "total_steps": 1500, "loss": 0.8861, "lr": 9.669983872401868e-06, "epoch": 0.614, "percentage": 20.47, "elapsed_time": "4:48:03", "remaining_time": "18:39:22"}
+{"current_steps": 308, "total_steps": 1500, "loss": 0.8731, "lr": 9.665814072907293e-06, "epoch": 0.616, "percentage": 20.53, "elapsed_time": "4:49:06", "remaining_time": "18:38:53"}
+{"current_steps": 309, "total_steps": 1500, "loss": 0.8891, "lr": 9.661619006077562e-06, "epoch": 0.618, "percentage": 20.6, "elapsed_time": "4:50:11", "remaining_time": "18:38:31"}
+{"current_steps": 310, "total_steps": 1500, "loss": 0.8677, "lr": 9.657398694630713e-06, "epoch": 0.62, "percentage": 20.67, "elapsed_time": "4:51:06", "remaining_time": "18:37:27"}
+{"current_steps": 311, "total_steps": 1500, "loss": 0.8357, "lr": 9.653153161421497e-06, "epoch": 0.622, "percentage": 20.73, "elapsed_time": "4:51:57", "remaining_time": "18:36:13"}
+{"current_steps": 312, "total_steps": 1500, "loss": 0.864, "lr": 9.648882429441258e-06, "epoch": 0.624, "percentage": 20.8, "elapsed_time": "4:52:48", "remaining_time": "18:34:56"}
+{"current_steps": 313, "total_steps": 1500, "loss": 0.8651, "lr": 9.644586521817792e-06, "epoch": 0.626, "percentage": 20.87, "elapsed_time": "4:53:42", "remaining_time": "18:33:51"}
+{"current_steps": 314, "total_steps": 1500, "loss": 0.8742, "lr": 9.640265461815235e-06, "epoch": 0.628, "percentage": 20.93, "elapsed_time": "4:54:43", "remaining_time": "18:33:12"}
+{"current_steps": 315, "total_steps": 1500, "loss": 0.8796, "lr": 9.635919272833938e-06, "epoch": 0.63, "percentage": 21.0, "elapsed_time": "4:55:29", "remaining_time": "18:31:37"}
+{"current_steps": 316, "total_steps": 1500, "loss": 0.8631, "lr": 9.63154797841033e-06, "epoch": 0.632, "percentage": 21.07, "elapsed_time": "4:56:26", "remaining_time": "18:30:41"}
+{"current_steps": 317, "total_steps": 1500, "loss": 0.852, "lr": 9.627151602216801e-06, "epoch": 0.634, "percentage": 21.13, "elapsed_time": "4:57:30", "remaining_time": "18:30:14"}
+{"current_steps": 318, "total_steps": 1500, "loss": 0.8714, "lr": 9.622730168061568e-06, "epoch": 0.636, "percentage": 21.2, "elapsed_time": "4:58:22", "remaining_time": "18:29:02"}
+{"current_steps": 319, "total_steps": 1500, "loss": 0.9097, "lr": 9.618283699888543e-06, "epoch": 0.638, "percentage": 21.27, "elapsed_time": "4:59:28", "remaining_time": "18:28:41"}
+{"current_steps": 320, "total_steps": 1500, "loss": 0.864, "lr": 9.613812221777212e-06, "epoch": 0.64, "percentage": 21.33, "elapsed_time": "5:00:21", "remaining_time": "18:27:34"}
+{"current_steps": 321, "total_steps": 1500, "loss": 0.8618, "lr": 9.609315757942504e-06, "epoch": 0.642, "percentage": 21.4, "elapsed_time": "5:01:18", "remaining_time": "18:26:38"}
+{"current_steps": 322, "total_steps": 1500, "loss": 0.8578, "lr": 9.604794332734647e-06, "epoch": 0.644, "percentage": 21.47, "elapsed_time": "5:02:15", "remaining_time": "18:25:46"}
+{"current_steps": 323, "total_steps": 1500, "loss": 0.8508, "lr": 9.600247970639053e-06, "epoch": 0.646, "percentage": 21.53, "elapsed_time": "5:03:10", "remaining_time": "18:24:46"}
+{"current_steps": 324, "total_steps": 1500, "loss": 0.9102, "lr": 9.595676696276173e-06, "epoch": 0.648, "percentage": 21.6, "elapsed_time": "5:04:06", "remaining_time": "18:23:47"}
+{"current_steps": 325, "total_steps": 1500, "loss": 0.8651, "lr": 9.591080534401371e-06, "epoch": 0.65, "percentage": 21.67, "elapsed_time": "5:05:08", "remaining_time": "18:23:11"}
+{"current_steps": 326, "total_steps": 1500, "loss": 0.8864, "lr": 9.586459509904786e-06, "epoch": 0.652, "percentage": 21.73, "elapsed_time": "5:06:02", "remaining_time": "18:22:07"}
+{"current_steps": 327, "total_steps": 1500, "loss": 0.8762, "lr": 9.581813647811199e-06, "epoch": 0.654, "percentage": 21.8, "elapsed_time": "5:07:00", "remaining_time": "18:21:16"}
+{"current_steps": 328, "total_steps": 1500, "loss": 0.8404, "lr": 9.577142973279896e-06, "epoch": 0.656, "percentage": 21.87, "elapsed_time": "5:07:46", "remaining_time": "18:19:42"}
+{"current_steps": 329, "total_steps": 1500, "loss": 0.8472, "lr": 9.572447511604536e-06, "epoch": 0.658, "percentage": 21.93, "elapsed_time": "5:08:40", "remaining_time": "18:18:40"}
+{"current_steps": 330, "total_steps": 1500, "loss": 0.9202, "lr": 9.567727288213005e-06, "epoch": 0.66, "percentage": 22.0, "elapsed_time": "5:09:47", "remaining_time": "18:18:22"}
+{"current_steps": 331, "total_steps": 1500, "loss": 0.876, "lr": 9.56298232866729e-06, "epoch": 0.662, "percentage": 22.07, "elapsed_time": "5:10:41", "remaining_time": "18:17:16"}
+{"current_steps": 332, "total_steps": 1500, "loss": 0.8977, "lr": 9.55821265866333e-06, "epoch": 0.664, "percentage": 22.13, "elapsed_time": "5:11:32", "remaining_time": "18:16:00"}
+{"current_steps": 333, "total_steps": 1500, "loss": 0.8066, "lr": 9.553418304030886e-06, "epoch": 0.666, "percentage": 22.2, "elapsed_time": "5:12:24", "remaining_time": "18:14:51"}
+{"current_steps": 334, "total_steps": 1500, "loss": 0.8716, "lr": 9.548599290733393e-06, "epoch": 0.668, "percentage": 22.27, "elapsed_time": "5:13:17", "remaining_time": "18:13:41"}
+{"current_steps": 335, "total_steps": 1500, "loss": 0.8626, "lr": 9.543755644867823e-06, "epoch": 0.67, "percentage": 22.33, "elapsed_time": "5:14:22", "remaining_time": "18:13:15"}
+{"current_steps": 336, "total_steps": 1500, "loss": 0.8535, "lr": 9.538887392664544e-06, "epoch": 0.672, "percentage": 22.4, "elapsed_time": "5:15:17", "remaining_time": "18:12:14"}
+{"current_steps": 337, "total_steps": 1500, "loss": 0.858, "lr": 9.53399456048718e-06, "epoch": 0.674, "percentage": 22.47, "elapsed_time": "5:16:18", "remaining_time": "18:11:34"}
+{"current_steps": 338, "total_steps": 1500, "loss": 0.8794, "lr": 9.529077174832466e-06, "epoch": 0.676, "percentage": 22.53, "elapsed_time": "5:17:16", "remaining_time": "18:10:43"}
+{"current_steps": 339, "total_steps": 1500, "loss": 0.8652, "lr": 9.524135262330098e-06, "epoch": 0.678, "percentage": 22.6, "elapsed_time": "5:18:15", "remaining_time": "18:09:57"}
+{"current_steps": 340, "total_steps": 1500, "loss": 0.8828, "lr": 9.519168849742603e-06, "epoch": 0.68, "percentage": 22.67, "elapsed_time": "5:19:17", "remaining_time": "18:09:21"}
+{"current_steps": 341, "total_steps": 1500, "loss": 0.85, "lr": 9.514177963965181e-06, "epoch": 0.682, "percentage": 22.73, "elapsed_time": "5:20:21", "remaining_time": "18:08:50"}
+{"current_steps": 342, "total_steps": 1500, "loss": 0.8623, "lr": 9.50916263202557e-06, "epoch": 0.684, "percentage": 22.8, "elapsed_time": "5:21:17", "remaining_time": "18:07:51"}
+{"current_steps": 343, "total_steps": 1500, "loss": 0.8575, "lr": 9.504122881083886e-06, "epoch": 0.686, "percentage": 22.87, "elapsed_time": "5:22:13", "remaining_time": "18:06:53"}
+{"current_steps": 344, "total_steps": 1500, "loss": 0.8873, "lr": 9.499058738432492e-06, "epoch": 0.688, "percentage": 22.93, "elapsed_time": "5:23:06", "remaining_time": "18:05:46"}
+{"current_steps": 345, "total_steps": 1500, "loss": 0.8831, "lr": 9.493970231495836e-06, "epoch": 0.69, "percentage": 23.0, "elapsed_time": "5:24:04", "remaining_time": "18:04:55"}
+{"current_steps": 346, "total_steps": 1500, "loss": 0.8665, "lr": 9.488857387830315e-06, "epoch": 0.692, "percentage": 23.07, "elapsed_time": "5:25:01", "remaining_time": "18:04:03"}
+{"current_steps": 347, "total_steps": 1500, "loss": 0.8505, "lr": 9.483720235124113e-06, "epoch": 0.694, "percentage": 23.13, "elapsed_time": "5:25:59", "remaining_time": "18:03:13"}
+{"current_steps": 348, "total_steps": 1500, "loss": 0.8524, "lr": 9.478558801197065e-06, "epoch": 0.696, "percentage": 23.2, "elapsed_time": "5:26:54", "remaining_time": "18:02:11"}
+{"current_steps": 349, "total_steps": 1500, "loss": 0.8424, "lr": 9.473373114000493e-06, "epoch": 0.698, "percentage": 23.27, "elapsed_time": "5:27:42", "remaining_time": "18:00:46"}
+{"current_steps": 350, "total_steps": 1500, "loss": 0.8715, "lr": 9.468163201617063e-06, "epoch": 0.7, "percentage": 23.33, "elapsed_time": "5:28:34", "remaining_time": "17:59:36"}
+{"current_steps": 351, "total_steps": 1500, "loss": 0.8191, "lr": 9.46292909226063e-06, "epoch": 0.702, "percentage": 23.4, "elapsed_time": "5:29:29", "remaining_time": "17:58:36"}
+{"current_steps": 352, "total_steps": 1500, "loss": 0.8458, "lr": 9.457670814276083e-06, "epoch": 0.704, "percentage": 23.47, "elapsed_time": "5:30:19", "remaining_time": "17:57:18"}
+{"current_steps": 353, "total_steps": 1500, "loss": 0.8391, "lr": 9.452388396139202e-06, "epoch": 0.706, "percentage": 23.53, "elapsed_time": "5:31:19", "remaining_time": "17:56:34"}
+{"current_steps": 354, "total_steps": 1500, "loss": 0.8578, "lr": 9.44708186645649e-06, "epoch": 0.708, "percentage": 23.6, "elapsed_time": "5:32:18", "remaining_time": "17:55:46"}
+{"current_steps": 355, "total_steps": 1500, "loss": 0.857, "lr": 9.441751253965022e-06, "epoch": 0.71, "percentage": 23.67, "elapsed_time": "5:33:11", "remaining_time": "17:54:38"}
+{"current_steps": 356, "total_steps": 1500, "loss": 0.86, "lr": 9.436396587532297e-06, "epoch": 0.712, "percentage": 23.73, "elapsed_time": "5:34:08", "remaining_time": "17:53:45"}
+{"current_steps": 357, "total_steps": 1500, "loss": 0.8748, "lr": 9.431017896156074e-06, "epoch": 0.714, "percentage": 23.8, "elapsed_time": "5:35:07", "remaining_time": "17:52:57"}
+{"current_steps": 358, "total_steps": 1500, "loss": 0.8504, "lr": 9.425615208964217e-06, "epoch": 0.716, "percentage": 23.87, "elapsed_time": "5:36:04", "remaining_time": "17:52:03"}
+{"current_steps": 359, "total_steps": 1500, "loss": 0.886, "lr": 9.420188555214537e-06, "epoch": 0.718, "percentage": 23.93, "elapsed_time": "5:37:03", "remaining_time": "17:51:16"}
+{"current_steps": 360, "total_steps": 1500, "loss": 0.8224, "lr": 9.414737964294636e-06, "epoch": 0.72, "percentage": 24.0, "elapsed_time": "5:37:50", "remaining_time": "17:49:50"}
+{"current_steps": 361, "total_steps": 1500, "loss": 0.8578, "lr": 9.40926346572174e-06, "epoch": 0.722, "percentage": 24.07, "elapsed_time": "5:38:43", "remaining_time": "17:48:43"}
+{"current_steps": 362, "total_steps": 1500, "loss": 0.8292, "lr": 9.403765089142554e-06, "epoch": 0.724, "percentage": 24.13, "elapsed_time": "5:39:45", "remaining_time": "17:48:06"}
+{"current_steps": 363, "total_steps": 1500, "loss": 0.8582, "lr": 9.398242864333084e-06, "epoch": 0.726, "percentage": 24.2, "elapsed_time": "5:40:44", "remaining_time": "17:47:16"}
+{"current_steps": 364, "total_steps": 1500, "loss": 0.8679, "lr": 9.392696821198488e-06, "epoch": 0.728, "percentage": 24.27, "elapsed_time": "5:41:36", "remaining_time": "17:46:08"}
+{"current_steps": 365, "total_steps": 1500, "loss": 0.8497, "lr": 9.38712698977291e-06, "epoch": 0.73, "percentage": 24.33, "elapsed_time": "5:42:35", "remaining_time": "17:45:18"}
+{"current_steps": 366, "total_steps": 1500, "loss": 0.8666, "lr": 9.381533400219319e-06, "epoch": 0.732, "percentage": 24.4, "elapsed_time": "5:43:34", "remaining_time": "17:44:30"}
+{"current_steps": 367, "total_steps": 1500, "loss": 0.8702, "lr": 9.375916082829341e-06, "epoch": 0.734, "percentage": 24.47, "elapsed_time": "5:44:26", "remaining_time": "17:43:22"}
+{"current_steps": 368, "total_steps": 1500, "loss": 0.8897, "lr": 9.370275068023097e-06, "epoch": 0.736, "percentage": 24.53, "elapsed_time": "5:45:14", "remaining_time": "17:41:59"}
+{"current_steps": 369, "total_steps": 1500, "loss": 0.8448, "lr": 9.364610386349048e-06, "epoch": 0.738, "percentage": 24.6, "elapsed_time": "5:46:07", "remaining_time": "17:40:51"}
+{"current_steps": 370, "total_steps": 1500, "loss": 0.8737, "lr": 9.358922068483813e-06, "epoch": 0.74, "percentage": 24.67, "elapsed_time": "5:47:02", "remaining_time": "17:39:51"}
+{"current_steps": 371, "total_steps": 1500, "loss": 0.9002, "lr": 9.35321014523201e-06, "epoch": 0.742, "percentage": 24.73, "elapsed_time": "5:47:53", "remaining_time": "17:38:40"}
+{"current_steps": 372, "total_steps": 1500, "loss": 0.8579, "lr": 9.347474647526095e-06, "epoch": 0.744, "percentage": 24.8, "elapsed_time": "5:48:44", "remaining_time": "17:37:29"}
+{"current_steps": 373, "total_steps": 1500, "loss": 0.8494, "lr": 9.34171560642619e-06, "epoch": 0.746, "percentage": 24.87, "elapsed_time": "5:49:42", "remaining_time": "17:36:38"}
+{"current_steps": 374, "total_steps": 1500, "loss": 0.8598, "lr": 9.335933053119906e-06, "epoch": 0.748, "percentage": 24.93, "elapsed_time": "5:50:33", "remaining_time": "17:35:25"}
+{"current_steps": 375, "total_steps": 1500, "loss": 0.8739, "lr": 9.330127018922195e-06, "epoch": 0.75, "percentage": 25.0, "elapsed_time": "5:51:24", "remaining_time": "17:34:12"}
+{"current_steps": 376, "total_steps": 1500, "loss": 0.8205, "lr": 9.324297535275156e-06, "epoch": 0.752, "percentage": 25.07, "elapsed_time": "5:52:07", "remaining_time": "17:32:38"}
+{"current_steps": 377, "total_steps": 1500, "loss": 0.858, "lr": 9.318444633747884e-06, "epoch": 0.754, "percentage": 25.13, "elapsed_time": "5:53:09", "remaining_time": "17:31:58"}
+{"current_steps": 378, "total_steps": 1500, "loss": 0.8341, "lr": 9.312568346036288e-06, "epoch": 0.756, "percentage": 25.2, "elapsed_time": "5:53:57", "remaining_time": "17:30:39"}
+{"current_steps": 379, "total_steps": 1500, "loss": 0.8241, "lr": 9.306668703962927e-06, "epoch": 0.758, "percentage": 25.27, "elapsed_time": "5:54:47", "remaining_time": "17:29:23"}
+{"current_steps": 380, "total_steps": 1500, "loss": 0.8986, "lr": 9.30074573947683e-06, "epoch": 0.76, "percentage": 25.33, "elapsed_time": "5:55:44", "remaining_time": "17:28:30"}
+{"current_steps": 381, "total_steps": 1500, "loss": 0.8645, "lr": 9.294799484653323e-06, "epoch": 0.762, "percentage": 25.4, "elapsed_time": "5:56:42", "remaining_time": "17:27:38"}
+{"current_steps": 382, "total_steps": 1500, "loss": 0.8283, "lr": 9.288829971693869e-06, "epoch": 0.764, "percentage": 25.47, "elapsed_time": "5:57:38", "remaining_time": "17:26:42"}
+{"current_steps": 383, "total_steps": 1500, "loss": 0.86, "lr": 9.282837232925876e-06, "epoch": 0.766, "percentage": 25.53, "elapsed_time": "5:58:35", "remaining_time": "17:25:49"}
+{"current_steps": 384, "total_steps": 1500, "loss": 0.8746, "lr": 9.276821300802535e-06, "epoch": 0.768, "percentage": 25.6, "elapsed_time": "5:59:27", "remaining_time": "17:24:41"}
+{"current_steps": 385, "total_steps": 1500, "loss": 0.8813, "lr": 9.27078220790263e-06, "epoch": 0.77, "percentage": 25.67, "elapsed_time": "6:00:28", "remaining_time": "17:23:58"}
+{"current_steps": 386, "total_steps": 1500, "loss": 0.8761, "lr": 9.264719986930376e-06, "epoch": 0.772, "percentage": 25.73, "elapsed_time": "6:01:22", "remaining_time": "17:22:54"}
+{"current_steps": 387, "total_steps": 1500, "loss": 0.8337, "lr": 9.25863467071524e-06, "epoch": 0.774, "percentage": 25.8, "elapsed_time": "6:02:13", "remaining_time": "17:21:45"}
+{"current_steps": 388, "total_steps": 1500, "loss": 0.8527, "lr": 9.25252629221175e-06, "epoch": 0.776, "percentage": 25.87, "elapsed_time": "6:03:10", "remaining_time": "17:20:51"}
+{"current_steps": 389, "total_steps": 1500, "loss": 0.8725, "lr": 9.246394884499334e-06, "epoch": 0.778, "percentage": 25.93, "elapsed_time": "6:04:13", "remaining_time": "17:20:15"}
+{"current_steps": 390, "total_steps": 1500, "loss": 0.8426, "lr": 9.24024048078213e-06, "epoch": 0.78, "percentage": 26.0, "elapsed_time": "6:05:06", "remaining_time": "17:19:08"}
+{"current_steps": 391, "total_steps": 1500, "loss": 0.8545, "lr": 9.234063114388809e-06, "epoch": 0.782, "percentage": 26.07, "elapsed_time": "6:06:00", "remaining_time": "17:18:07"}
+{"current_steps": 392, "total_steps": 1500, "loss": 0.8346, "lr": 9.227862818772392e-06, "epoch": 0.784, "percentage": 26.13, "elapsed_time": "6:06:54", "remaining_time": "17:17:04"}
+{"current_steps": 393, "total_steps": 1500, "loss": 0.8622, "lr": 9.221639627510076e-06, "epoch": 0.786, "percentage": 26.2, "elapsed_time": "6:07:54", "remaining_time": "17:16:18"}
+{"current_steps": 394, "total_steps": 1500, "loss": 0.8601, "lr": 9.215393574303043e-06, "epoch": 0.788, "percentage": 26.27, "elapsed_time": "6:08:48", "remaining_time": "17:15:18"}
+{"current_steps": 395, "total_steps": 1500, "loss": 0.8487, "lr": 9.209124692976287e-06, "epoch": 0.79, "percentage": 26.33, "elapsed_time": "6:09:52", "remaining_time": "17:14:41"}
+{"current_steps": 396, "total_steps": 1500, "loss": 0.844, "lr": 9.202833017478421e-06, "epoch": 0.792, "percentage": 26.4, "elapsed_time": "6:10:35", "remaining_time": "17:13:11"}
+{"current_steps": 397, "total_steps": 1500, "loss": 0.8448, "lr": 9.196518581881502e-06, "epoch": 0.794, "percentage": 26.47, "elapsed_time": "6:11:33", "remaining_time": "17:12:19"}
+{"current_steps": 398, "total_steps": 1500, "loss": 0.8535, "lr": 9.190181420380838e-06, "epoch": 0.796, "percentage": 26.53, "elapsed_time": "6:12:28", "remaining_time": "17:11:18"}
+{"current_steps": 399, "total_steps": 1500, "loss": 0.8507, "lr": 9.18382156729481e-06, "epoch": 0.798, "percentage": 26.6, "elapsed_time": "6:13:14", "remaining_time": "17:09:56"}
+{"current_steps": 400, "total_steps": 1500, "loss": 0.8617, "lr": 9.177439057064684e-06, "epoch": 0.8, "percentage": 26.67, "elapsed_time": "6:14:10", "remaining_time": "17:08:59"}
+{"current_steps": 401, "total_steps": 1500, "loss": 0.8766, "lr": 9.171033924254421e-06, "epoch": 0.802, "percentage": 26.73, "elapsed_time": "6:15:03", "remaining_time": "17:07:53"}
+{"current_steps": 402, "total_steps": 1500, "loss": 0.8283, "lr": 9.164606203550498e-06, "epoch": 0.804, "percentage": 26.8, "elapsed_time": "6:16:03", "remaining_time": "17:07:07"}
+{"current_steps": 403, "total_steps": 1500, "loss": 0.8593, "lr": 9.15815592976171e-06, "epoch": 0.806, "percentage": 26.87, "elapsed_time": "6:16:56", "remaining_time": "17:06:03"}
+{"current_steps": 404, "total_steps": 1500, "loss": 0.8423, "lr": 9.151683137818989e-06, "epoch": 0.808, "percentage": 26.93, "elapsed_time": "6:17:45", "remaining_time": "17:04:47"}
+{"current_steps": 405, "total_steps": 1500, "loss": 0.8409, "lr": 9.145187862775208e-06, "epoch": 0.81, "percentage": 27.0, "elapsed_time": "6:18:39", "remaining_time": "17:03:48"}
+{"current_steps": 406, "total_steps": 1500, "loss": 0.8458, "lr": 9.138670139805004e-06, "epoch": 0.812, "percentage": 27.07, "elapsed_time": "6:19:36", "remaining_time": "17:02:53"}
+{"current_steps": 407, "total_steps": 1500, "loss": 0.8556, "lr": 9.132130004204569e-06, "epoch": 0.814, "percentage": 27.13, "elapsed_time": "6:20:21", "remaining_time": "17:01:28"}
+{"current_steps": 408, "total_steps": 1500, "loss": 0.8907, "lr": 9.125567491391476e-06, "epoch": 0.816, "percentage": 27.2, "elapsed_time": "6:21:16", "remaining_time": "17:00:28"}
+{"current_steps": 409, "total_steps": 1500, "loss": 0.8372, "lr": 9.118982636904476e-06, "epoch": 0.818, "percentage": 27.27, "elapsed_time": "6:21:58", "remaining_time": "16:58:53"}
+{"current_steps": 410, "total_steps": 1500, "loss": 0.8363, "lr": 9.112375476403313e-06, "epoch": 0.82, "percentage": 27.33, "elapsed_time": "6:22:58", "remaining_time": "16:58:07"}
+{"current_steps": 411, "total_steps": 1500, "loss": 0.8808, "lr": 9.10574604566852e-06, "epoch": 0.822, "percentage": 27.4, "elapsed_time": "6:23:50", "remaining_time": "16:57:02"}
+{"current_steps": 412, "total_steps": 1500, "loss": 0.8395, "lr": 9.099094380601244e-06, "epoch": 0.824, "percentage": 27.47, "elapsed_time": "6:24:51", "remaining_time": "16:56:20"}
+{"current_steps": 413, "total_steps": 1500, "loss": 0.8495, "lr": 9.09242051722303e-06, "epoch": 0.826, "percentage": 27.53, "elapsed_time": "6:25:41", "remaining_time": "16:55:08"}
+{"current_steps": 414, "total_steps": 1500, "loss": 0.8482, "lr": 9.085724491675642e-06, "epoch": 0.828, "percentage": 27.6, "elapsed_time": "6:26:33", "remaining_time": "16:54:00"}
+{"current_steps": 415, "total_steps": 1500, "loss": 0.8646, "lr": 9.079006340220862e-06, "epoch": 0.83, "percentage": 27.67, "elapsed_time": "6:27:34", "remaining_time": "16:53:18"}
+{"current_steps": 416, "total_steps": 1500, "loss": 0.8486, "lr": 9.072266099240286e-06, "epoch": 0.832, "percentage": 27.73, "elapsed_time": "6:28:30", "remaining_time": "16:52:21"}
+{"current_steps": 417, "total_steps": 1500, "loss": 0.8797, "lr": 9.065503805235139e-06, "epoch": 0.834, "percentage": 27.8, "elapsed_time": "6:29:26", "remaining_time": "16:51:25"}
+{"current_steps": 418, "total_steps": 1500, "loss": 0.8313, "lr": 9.058719494826076e-06, "epoch": 0.836, "percentage": 27.87, "elapsed_time": "6:30:16", "remaining_time": "16:50:13"}
+{"current_steps": 419, "total_steps": 1500, "loss": 0.8577, "lr": 9.051913204752972e-06, "epoch": 0.838, "percentage": 27.93, "elapsed_time": "6:31:17", "remaining_time": "16:49:31"}
+{"current_steps": 420, "total_steps": 1500, "loss": 0.8476, "lr": 9.045084971874738e-06, "epoch": 0.84, "percentage": 28.0, "elapsed_time": "6:32:21", "remaining_time": "16:48:54"}
+{"current_steps": 421, "total_steps": 1500, "loss": 0.8365, "lr": 9.03823483316911e-06, "epoch": 0.842, "percentage": 28.07, "elapsed_time": "6:33:10", "remaining_time": "16:47:41"}
+{"current_steps": 422, "total_steps": 1500, "loss": 0.9092, "lr": 9.031362825732456e-06, "epoch": 0.844, "percentage": 28.13, "elapsed_time": "6:34:15", "remaining_time": "16:47:07"}
+{"current_steps": 423, "total_steps": 1500, "loss": 0.8825, "lr": 9.02446898677957e-06, "epoch": 0.846, "percentage": 28.2, "elapsed_time": "6:35:24", "remaining_time": "16:46:45"}
+{"current_steps": 424, "total_steps": 1500, "loss": 0.8157, "lr": 9.017553353643479e-06, "epoch": 0.848, "percentage": 28.27, "elapsed_time": "6:36:17", "remaining_time": "16:45:42"}
+{"current_steps": 425, "total_steps": 1500, "loss": 0.8614, "lr": 9.01061596377522e-06, "epoch": 0.85, "percentage": 28.33, "elapsed_time": "6:37:12", "remaining_time": "16:44:43"}
+{"current_steps": 426, "total_steps": 1500, "loss": 0.8652, "lr": 9.003656854743667e-06, "epoch": 0.852, "percentage": 28.4, "elapsed_time": "6:38:08", "remaining_time": "16:43:45"}
+{"current_steps": 427, "total_steps": 1500, "loss": 0.8033, "lr": 8.996676064235308e-06, "epoch": 0.854, "percentage": 28.47, "elapsed_time": "6:38:57", "remaining_time": "16:42:31"}
+{"current_steps": 428, "total_steps": 1500, "loss": 0.8576, "lr": 8.989673630054044e-06, "epoch": 0.856, "percentage": 28.53, "elapsed_time": "6:39:56", "remaining_time": "16:41:43"}
+{"current_steps": 429, "total_steps": 1500, "loss": 0.8649, "lr": 8.982649590120982e-06, "epoch": 0.858, "percentage": 28.6, "elapsed_time": "6:40:51", "remaining_time": "16:40:43"}
+{"current_steps": 430, "total_steps": 1500, "loss": 0.8683, "lr": 8.97560398247424e-06, "epoch": 0.86, "percentage": 28.67, "elapsed_time": "6:41:43", "remaining_time": "16:39:38"}
+{"current_steps": 431, "total_steps": 1500, "loss": 0.8796, "lr": 8.96853684526873e-06, "epoch": 0.862, "percentage": 28.73, "elapsed_time": "6:42:47", "remaining_time": "16:39:03"}
+{"current_steps": 432, "total_steps": 1500, "loss": 0.8421, "lr": 8.961448216775955e-06, "epoch": 0.864, "percentage": 28.8, "elapsed_time": "6:43:47", "remaining_time": "16:38:14"}
+{"current_steps": 433, "total_steps": 1500, "loss": 0.8413, "lr": 8.954338135383804e-06, "epoch": 0.866, "percentage": 28.87, "elapsed_time": "6:44:40", "remaining_time": "16:37:11"}
+{"current_steps": 434, "total_steps": 1500, "loss": 0.8446, "lr": 8.947206639596346e-06, "epoch": 0.868, "percentage": 28.93, "elapsed_time": "6:45:29", "remaining_time": "16:35:59"}
+{"current_steps": 435, "total_steps": 1500, "loss": 0.8391, "lr": 8.94005376803361e-06, "epoch": 0.87, "percentage": 29.0, "elapsed_time": "6:46:26", "remaining_time": "16:35:05"}
+{"current_steps": 436, "total_steps": 1500, "loss": 0.8555, "lr": 8.932879559431392e-06, "epoch": 0.872, "percentage": 29.07, "elapsed_time": "6:47:15", "remaining_time": "16:33:51"}
+{"current_steps": 437, "total_steps": 1500, "loss": 0.8485, "lr": 8.925684052641027e-06, "epoch": 0.874, "percentage": 29.13, "elapsed_time": "6:48:08", "remaining_time": "16:32:48"}
+{"current_steps": 438, "total_steps": 1500, "loss": 0.8347, "lr": 8.9184672866292e-06, "epoch": 0.876, "percentage": 29.2, "elapsed_time": "6:49:08", "remaining_time": "16:32:02"}
+{"current_steps": 439, "total_steps": 1500, "loss": 0.8764, "lr": 8.911229300477716e-06, "epoch": 0.878, "percentage": 29.27, "elapsed_time": "6:50:07", "remaining_time": "16:31:11"}
+{"current_steps": 440, "total_steps": 1500, "loss": 0.8486, "lr": 8.903970133383297e-06, "epoch": 0.88, "percentage": 29.33, "elapsed_time": "6:50:57", "remaining_time": "16:30:01"}
+{"current_steps": 441, "total_steps": 1500, "loss": 0.8586, "lr": 8.896689824657371e-06, "epoch": 0.882, "percentage": 29.4, "elapsed_time": "6:51:43", "remaining_time": "16:28:41"}
+{"current_steps": 442, "total_steps": 1500, "loss": 0.8255, "lr": 8.889388413725857e-06, "epoch": 0.884, "percentage": 29.47, "elapsed_time": "6:52:47", "remaining_time": "16:28:05"}
+{"current_steps": 443, "total_steps": 1500, "loss": 0.8686, "lr": 8.882065940128946e-06, "epoch": 0.886, "percentage": 29.53, "elapsed_time": "6:53:37", "remaining_time": "16:26:54"}
+{"current_steps": 444, "total_steps": 1500, "loss": 0.8468, "lr": 8.874722443520898e-06, "epoch": 0.888, "percentage": 29.6, "elapsed_time": "6:54:29", "remaining_time": "16:25:49"}
+{"current_steps": 445, "total_steps": 1500, "loss": 0.8376, "lr": 8.867357963669821e-06, "epoch": 0.89, "percentage": 29.67, "elapsed_time": "6:55:10", "remaining_time": "16:24:17"}
+{"current_steps": 446, "total_steps": 1500, "loss": 0.826, "lr": 8.859972540457451e-06, "epoch": 0.892, "percentage": 29.73, "elapsed_time": "6:56:04", "remaining_time": "16:23:15"}
+{"current_steps": 447, "total_steps": 1500, "loss": 0.852, "lr": 8.852566213878947e-06, "epoch": 0.894, "percentage": 29.8, "elapsed_time": "6:56:54", "remaining_time": "16:22:06"}
+{"current_steps": 448, "total_steps": 1500, "loss": 0.8513, "lr": 8.845139024042664e-06, "epoch": 0.896, "percentage": 29.87, "elapsed_time": "6:57:43", "remaining_time": "16:20:53"}
+{"current_steps": 449, "total_steps": 1500, "loss": 0.8531, "lr": 8.837691011169944e-06, "epoch": 0.898, "percentage": 29.93, "elapsed_time": "6:58:40", "remaining_time": "16:20:01"}
+{"current_steps": 450, "total_steps": 1500, "loss": 0.8919, "lr": 8.83022221559489e-06, "epoch": 0.9, "percentage": 30.0, "elapsed_time": "6:59:38", "remaining_time": "16:19:09"}
+{"current_steps": 451, "total_steps": 1500, "loss": 0.9006, "lr": 8.822732677764158e-06, "epoch": 0.902, "percentage": 30.07, "elapsed_time": "7:00:39", "remaining_time": "16:18:26"}
+{"current_steps": 452, "total_steps": 1500, "loss": 0.8686, "lr": 8.815222438236726e-06, "epoch": 0.904, "percentage": 30.13, "elapsed_time": "7:01:44", "remaining_time": "16:17:50"}
+{"current_steps": 453, "total_steps": 1500, "loss": 0.8542, "lr": 8.807691537683685e-06, "epoch": 0.906, "percentage": 30.2, "elapsed_time": "7:02:41", "remaining_time": "16:16:57"}
+{"current_steps": 454, "total_steps": 1500, "loss": 0.8584, "lr": 8.800140016888009e-06, "epoch": 0.908, "percentage": 30.27, "elapsed_time": "7:03:41", "remaining_time": "16:16:10"}
+{"current_steps": 455, "total_steps": 1500, "loss": 0.8548, "lr": 8.792567916744346e-06, "epoch": 0.91, "percentage": 30.33, "elapsed_time": "7:04:40", "remaining_time": "16:15:20"}
+{"current_steps": 456, "total_steps": 1500, "loss": 0.8714, "lr": 8.784975278258783e-06, "epoch": 0.912, "percentage": 30.4, "elapsed_time": "7:05:37", "remaining_time": "16:14:26"}
+{"current_steps": 457, "total_steps": 1500, "loss": 0.8662, "lr": 8.777362142548636e-06, "epoch": 0.914, "percentage": 30.47, "elapsed_time": "7:06:34", "remaining_time": "16:13:33"}
+{"current_steps": 458, "total_steps": 1500, "loss": 0.8304, "lr": 8.769728550842217e-06, "epoch": 0.916, "percentage": 30.53, "elapsed_time": "7:07:23", "remaining_time": "16:12:21"}
+{"current_steps": 459, "total_steps": 1500, "loss": 0.8184, "lr": 8.762074544478622e-06, "epoch": 0.918, "percentage": 30.6, "elapsed_time": "7:08:15", "remaining_time": "16:11:16"}
+{"current_steps": 460, "total_steps": 1500, "loss": 0.854, "lr": 8.754400164907496e-06, "epoch": 0.92, "percentage": 30.67, "elapsed_time": "7:09:05", "remaining_time": "16:10:08"}
+{"current_steps": 461, "total_steps": 1500, "loss": 0.82, "lr": 8.746705453688815e-06, "epoch": 0.922, "percentage": 30.73, "elapsed_time": "7:09:47", "remaining_time": "16:08:39"}
+{"current_steps": 462, "total_steps": 1500, "loss": 0.8278, "lr": 8.73899045249266e-06, "epoch": 0.924, "percentage": 30.8, "elapsed_time": "7:10:36", "remaining_time": "16:07:27"}
+{"current_steps": 463, "total_steps": 1500, "loss": 0.8389, "lr": 8.73125520309899e-06, "epoch": 0.926, "percentage": 30.87, "elapsed_time": "7:11:30", "remaining_time": "16:06:27"}
+{"current_steps": 464, "total_steps": 1500, "loss": 0.8692, "lr": 8.723499747397415e-06, "epoch": 0.928, "percentage": 30.93, "elapsed_time": "7:12:25", "remaining_time": "16:05:30"}
+{"current_steps": 465, "total_steps": 1500, "loss": 0.8682, "lr": 8.715724127386971e-06, "epoch": 0.93, "percentage": 31.0, "elapsed_time": "7:13:23", "remaining_time": "16:04:38"}
+{"current_steps": 466, "total_steps": 1500, "loss": 0.8257, "lr": 8.707928385175898e-06, "epoch": 0.932, "percentage": 31.07, "elapsed_time": "7:14:18", "remaining_time": "16:03:39"}
+{"current_steps": 467, "total_steps": 1500, "loss": 0.8548, "lr": 8.700112562981398e-06, "epoch": 0.934, "percentage": 31.13, "elapsed_time": "7:15:09", "remaining_time": "16:02:34"}
+{"current_steps": 468, "total_steps": 1500, "loss": 0.8589, "lr": 8.692276703129421e-06, "epoch": 0.936, "percentage": 31.2, "elapsed_time": "7:16:05", "remaining_time": "16:01:38"}
+{"current_steps": 469, "total_steps": 1500, "loss": 0.8387, "lr": 8.68442084805442e-06, "epoch": 0.938, "percentage": 31.27, "elapsed_time": "7:16:56", "remaining_time": "16:00:32"}
+{"current_steps": 470, "total_steps": 1500, "loss": 0.8649, "lr": 8.676545040299145e-06, "epoch": 0.94, "percentage": 31.33, "elapsed_time": "7:17:48", "remaining_time": "15:59:27"}
+{"current_steps": 471, "total_steps": 1500, "loss": 0.867, "lr": 8.668649322514382e-06, "epoch": 0.942, "percentage": 31.4, "elapsed_time": "7:18:42", "remaining_time": "15:58:27"}
+{"current_steps": 472, "total_steps": 1500, "loss": 0.8723, "lr": 8.660733737458751e-06, "epoch": 0.944, "percentage": 31.47, "elapsed_time": "7:19:38", "remaining_time": "15:57:30"}
+{"current_steps": 473, "total_steps": 1500, "loss": 0.8535, "lr": 8.652798327998458e-06, "epoch": 0.946, "percentage": 31.53, "elapsed_time": "7:20:35", "remaining_time": "15:56:37"}
+{"current_steps": 474, "total_steps": 1500, "loss": 0.8747, "lr": 8.644843137107058e-06, "epoch": 0.948, "percentage": 31.6, "elapsed_time": "7:21:30", "remaining_time": "15:55:39"}
+{"current_steps": 475, "total_steps": 1500, "loss": 0.8493, "lr": 8.636868207865244e-06, "epoch": 0.95, "percentage": 31.67, "elapsed_time": "7:22:24", "remaining_time": "15:54:39"}
+{"current_steps": 476, "total_steps": 1500, "loss": 0.8726, "lr": 8.628873583460593e-06, "epoch": 0.952, "percentage": 31.73, "elapsed_time": "7:23:21", "remaining_time": "15:53:47"}
+{"current_steps": 477, "total_steps": 1500, "loss": 0.8766, "lr": 8.620859307187339e-06, "epoch": 0.954, "percentage": 31.8, "elapsed_time": "7:24:23", "remaining_time": "15:53:04"}
+{"current_steps": 478, "total_steps": 1500, "loss": 0.8589, "lr": 8.61282542244614e-06, "epoch": 0.956, "percentage": 31.87, "elapsed_time": "7:25:19", "remaining_time": "15:52:08"}
+{"current_steps": 479, "total_steps": 1500, "loss": 0.8483, "lr": 8.604771972743848e-06, "epoch": 0.958, "percentage": 31.93, "elapsed_time": "7:26:08", "remaining_time": "15:50:57"}
+{"current_steps": 480, "total_steps": 1500, "loss": 0.8311, "lr": 8.596699001693257e-06, "epoch": 0.96, "percentage": 32.0, "elapsed_time": "7:26:55", "remaining_time": "15:49:43"}
+{"current_steps": 481, "total_steps": 1500, "loss": 0.8645, "lr": 8.588606553012884e-06, "epoch": 0.962, "percentage": 32.07, "elapsed_time": "7:27:55", "remaining_time": "15:48:55"}
+{"current_steps": 482, "total_steps": 1500, "loss": 0.8759, "lr": 8.580494670526725e-06, "epoch": 0.964, "percentage": 32.13, "elapsed_time": "7:29:02", "remaining_time": "15:48:23"}
+{"current_steps": 483, "total_steps": 1500, "loss": 0.8299, "lr": 8.572363398164017e-06, "epoch": 0.966, "percentage": 32.2, "elapsed_time": "7:29:53", "remaining_time": "15:47:17"}
+{"current_steps": 484, "total_steps": 1500, "loss": 0.8398, "lr": 8.564212779959003e-06, "epoch": 0.968, "percentage": 32.27, "elapsed_time": "7:30:57", "remaining_time": "15:46:38"}
+{"current_steps": 485, "total_steps": 1500, "loss": 0.8251, "lr": 8.556042860050686e-06, "epoch": 0.97, "percentage": 32.33, "elapsed_time": "7:31:43", "remaining_time": "15:45:22"}
+{"current_steps": 486, "total_steps": 1500, "loss": 0.8587, "lr": 8.547853682682605e-06, "epoch": 0.972, "percentage": 32.4, "elapsed_time": "7:32:39", "remaining_time": "15:44:26"}
+{"current_steps": 487, "total_steps": 1500, "loss": 0.8724, "lr": 8.539645292202579e-06, "epoch": 0.974, "percentage": 32.47, "elapsed_time": "7:33:31", "remaining_time": "15:43:22"}
+{"current_steps": 488, "total_steps": 1500, "loss": 0.8646, "lr": 8.531417733062476e-06, "epoch": 0.976, "percentage": 32.53, "elapsed_time": "7:34:27", "remaining_time": "15:42:27"}
+{"current_steps": 489, "total_steps": 1500, "loss": 0.8562, "lr": 8.523171049817974e-06, "epoch": 0.978, "percentage": 32.6, "elapsed_time": "7:35:22", "remaining_time": "15:41:27"}
+{"current_steps": 490, "total_steps": 1500, "loss": 0.8098, "lr": 8.51490528712831e-06, "epoch": 0.98, "percentage": 32.67, "elapsed_time": "7:36:01", "remaining_time": "15:39:58"}
+{"current_steps": 491, "total_steps": 1500, "loss": 0.8605, "lr": 8.506620489756045e-06, "epoch": 0.982, "percentage": 32.73, "elapsed_time": "7:37:01", "remaining_time": "15:39:11"}
+{"current_steps": 492, "total_steps": 1500, "loss": 0.8312, "lr": 8.498316702566828e-06, "epoch": 0.984, "percentage": 32.8, "elapsed_time": "7:37:53", "remaining_time": "15:38:06"}
+{"current_steps": 493, "total_steps": 1500, "loss": 0.857, "lr": 8.489993970529137e-06, "epoch": 0.986, "percentage": 32.87, "elapsed_time": "7:38:46", "remaining_time": "15:37:05"}
+{"current_steps": 494, "total_steps": 1500, "loss": 0.8571, "lr": 8.481652338714048e-06, "epoch": 0.988, "percentage": 32.93, "elapsed_time": "7:39:46", "remaining_time": "15:36:18"}
+{"current_steps": 495, "total_steps": 1500, "loss": 0.8774, "lr": 8.473291852294986e-06, "epoch": 0.99, "percentage": 33.0, "elapsed_time": "7:40:33", "remaining_time": "15:35:03"}
+{"current_steps": 496, "total_steps": 1500, "loss": 0.8824, "lr": 8.464912556547486e-06, "epoch": 0.992, "percentage": 33.07, "elapsed_time": "7:41:32", "remaining_time": "15:34:14"}
+{"current_steps": 497, "total_steps": 1500, "loss": 0.8418, "lr": 8.456514496848938e-06, "epoch": 0.994, "percentage": 33.13, "elapsed_time": "7:42:22", "remaining_time": "15:33:06"}
+{"current_steps": 498, "total_steps": 1500, "loss": 0.8599, "lr": 8.44809771867835e-06, "epoch": 0.996, "percentage": 33.2, "elapsed_time": "7:43:20", "remaining_time": "15:32:16"}
+{"current_steps": 499, "total_steps": 1500, "loss": 0.8589, "lr": 8.439662267616093e-06, "epoch": 0.998, "percentage": 33.27, "elapsed_time": "7:44:13", "remaining_time": "15:31:14"}
+{"current_steps": 500, "total_steps": 1500, "loss": 0.8247, "lr": 8.43120818934367e-06, "epoch": 1.0, "percentage": 33.33, "elapsed_time": "7:44:59", "remaining_time": "15:29:58"}
+{"current_steps": 501, "total_steps": 1500, "loss": 0.7995, "lr": 8.422735529643445e-06, "epoch": 1.002, "percentage": 33.4, "elapsed_time": "7:46:50", "remaining_time": "15:30:52"}
+{"current_steps": 502, "total_steps": 1500, "loss": 0.7948, "lr": 8.414244334398418e-06, "epoch": 1.004, "percentage": 33.47, "elapsed_time": "7:47:36", "remaining_time": "15:29:38"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:195cb58904593f507a9d38f0146e12e440fd6252282e301faabf7c032767fd7a
+size 7416