Training in progress, epoch 0

.gitattributes

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

added_tokens.json

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

config.json

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

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3ff49f3b67b59203ec8d8ee192f9bea18e2080f9d075140416043daa5362b62
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02e4386df21b4a70303bdb30eb76f47423709d5c9bc81bcba73cf0919e540710
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:557199f10896181d25751b1b65efa3a1eb76d0607341eb9ee5ffb460a04e1f35
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:003f048daa5f325d2403815dcc3c02a22a43f6c532718ce758320d729de9ab53
+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 +1,444 @@
+{"current_steps": 1, "total_steps": 2195, "loss": 0.8913, "lr": 1.8181818181818183e-07, "epoch": 0.0022753128555176336, "percentage": 0.05, "elapsed_time": "0:00:43", "remaining_time": "1 day, 2:45:29"}
+{"current_steps": 2, "total_steps": 2195, "loss": 0.8872, "lr": 3.6363636363636366e-07, "epoch": 0.004550625711035267, "percentage": 0.09, "elapsed_time": "0:01:20", "remaining_time": "1 day, 0:32:22"}
+{"current_steps": 3, "total_steps": 2195, "loss": 0.8818, "lr": 5.454545454545455e-07, "epoch": 0.006825938566552901, "percentage": 0.14, "elapsed_time": "0:01:57", "remaining_time": "23:54:52"}
+{"current_steps": 4, "total_steps": 2195, "loss": 0.9099, "lr": 7.272727272727273e-07, "epoch": 0.009101251422070534, "percentage": 0.18, "elapsed_time": "0:02:31", "remaining_time": "23:00:01"}
+{"current_steps": 5, "total_steps": 2195, "loss": 0.9119, "lr": 9.090909090909091e-07, "epoch": 0.011376564277588168, "percentage": 0.23, "elapsed_time": "0:03:12", "remaining_time": "23:21:58"}
+{"current_steps": 6, "total_steps": 2195, "loss": 0.8553, "lr": 1.090909090909091e-06, "epoch": 0.013651877133105802, "percentage": 0.27, "elapsed_time": "0:03:54", "remaining_time": "23:45:10"}
+{"current_steps": 7, "total_steps": 2195, "loss": 0.8673, "lr": 1.2727272727272728e-06, "epoch": 0.015927189988623434, "percentage": 0.32, "elapsed_time": "0:04:33", "remaining_time": "23:45:13"}
+{"current_steps": 8, "total_steps": 2195, "loss": 0.8509, "lr": 1.4545454545454546e-06, "epoch": 0.01820250284414107, "percentage": 0.36, "elapsed_time": "0:05:14", "remaining_time": "23:54:31"}
+{"current_steps": 9, "total_steps": 2195, "loss": 0.8566, "lr": 1.6363636363636365e-06, "epoch": 0.020477815699658702, "percentage": 0.41, "elapsed_time": "0:05:52", "remaining_time": "23:47:28"}
+{"current_steps": 10, "total_steps": 2195, "loss": 0.8302, "lr": 1.8181818181818183e-06, "epoch": 0.022753128555176336, "percentage": 0.46, "elapsed_time": "0:06:34", "remaining_time": "23:58:18"}
+{"current_steps": 11, "total_steps": 2195, "loss": 0.7387, "lr": 2.0000000000000003e-06, "epoch": 0.02502844141069397, "percentage": 0.5, "elapsed_time": "0:07:11", "remaining_time": "23:49:10"}
+{"current_steps": 12, "total_steps": 2195, "loss": 0.7747, "lr": 2.181818181818182e-06, "epoch": 0.027303754266211604, "percentage": 0.55, "elapsed_time": "0:07:45", "remaining_time": "23:29:59"}
+{"current_steps": 13, "total_steps": 2195, "loss": 0.7314, "lr": 2.363636363636364e-06, "epoch": 0.02957906712172924, "percentage": 0.59, "elapsed_time": "0:08:22", "remaining_time": "23:26:42"}
+{"current_steps": 14, "total_steps": 2195, "loss": 0.7661, "lr": 2.5454545454545456e-06, "epoch": 0.03185437997724687, "percentage": 0.64, "elapsed_time": "0:09:03", "remaining_time": "23:31:28"}
+{"current_steps": 15, "total_steps": 2195, "loss": 0.8176, "lr": 2.7272727272727272e-06, "epoch": 0.034129692832764506, "percentage": 0.68, "elapsed_time": "0:09:41", "remaining_time": "23:29:23"}
+{"current_steps": 16, "total_steps": 2195, "loss": 0.7332, "lr": 2.9090909090909093e-06, "epoch": 0.03640500568828214, "percentage": 0.73, "elapsed_time": "0:10:15", "remaining_time": "23:16:19"}
+{"current_steps": 17, "total_steps": 2195, "loss": 0.8074, "lr": 3.090909090909091e-06, "epoch": 0.038680318543799774, "percentage": 0.77, "elapsed_time": "0:10:56", "remaining_time": "23:21:16"}
+{"current_steps": 18, "total_steps": 2195, "loss": 0.7319, "lr": 3.272727272727273e-06, "epoch": 0.040955631399317405, "percentage": 0.82, "elapsed_time": "0:11:34", "remaining_time": "23:19:18"}
+{"current_steps": 19, "total_steps": 2195, "loss": 0.7348, "lr": 3.454545454545455e-06, "epoch": 0.04323094425483504, "percentage": 0.87, "elapsed_time": "0:12:12", "remaining_time": "23:17:30"}
+{"current_steps": 20, "total_steps": 2195, "loss": 0.6978, "lr": 3.6363636363636366e-06, "epoch": 0.04550625711035267, "percentage": 0.91, "elapsed_time": "0:12:57", "remaining_time": "23:29:01"}
+{"current_steps": 21, "total_steps": 2195, "loss": 0.6676, "lr": 3.818181818181819e-06, "epoch": 0.04778156996587031, "percentage": 0.96, "elapsed_time": "0:13:32", "remaining_time": "23:21:10"}
+{"current_steps": 22, "total_steps": 2195, "loss": 0.6623, "lr": 4.000000000000001e-06, "epoch": 0.05005688282138794, "percentage": 1.0, "elapsed_time": "0:14:06", "remaining_time": "23:13:24"}
+{"current_steps": 23, "total_steps": 2195, "loss": 0.6766, "lr": 4.181818181818182e-06, "epoch": 0.05233219567690557, "percentage": 1.05, "elapsed_time": "0:14:46", "remaining_time": "23:15:41"}
+{"current_steps": 24, "total_steps": 2195, "loss": 0.6282, "lr": 4.363636363636364e-06, "epoch": 0.05460750853242321, "percentage": 1.09, "elapsed_time": "0:15:26", "remaining_time": "23:16:33"}
+{"current_steps": 25, "total_steps": 2195, "loss": 0.6829, "lr": 4.5454545454545455e-06, "epoch": 0.05688282138794084, "percentage": 1.14, "elapsed_time": "0:16:03", "remaining_time": "23:14:02"}
+{"current_steps": 26, "total_steps": 2195, "loss": 0.6547, "lr": 4.727272727272728e-06, "epoch": 0.05915813424345848, "percentage": 1.18, "elapsed_time": "0:16:43", "remaining_time": "23:15:48"}
+{"current_steps": 27, "total_steps": 2195, "loss": 0.6336, "lr": 4.90909090909091e-06, "epoch": 0.06143344709897611, "percentage": 1.23, "elapsed_time": "0:17:19", "remaining_time": "23:11:39"}
+{"current_steps": 28, "total_steps": 2195, "loss": 0.6571, "lr": 5.090909090909091e-06, "epoch": 0.06370875995449374, "percentage": 1.28, "elapsed_time": "0:17:53", "remaining_time": "23:04:53"}
+{"current_steps": 29, "total_steps": 2195, "loss": 0.6185, "lr": 5.272727272727273e-06, "epoch": 0.06598407281001138, "percentage": 1.32, "elapsed_time": "0:18:35", "remaining_time": "23:09:11"}
+{"current_steps": 30, "total_steps": 2195, "loss": 0.6272, "lr": 5.4545454545454545e-06, "epoch": 0.06825938566552901, "percentage": 1.37, "elapsed_time": "0:19:16", "remaining_time": "23:11:13"}
+{"current_steps": 31, "total_steps": 2195, "loss": 0.5826, "lr": 5.636363636363636e-06, "epoch": 0.07053469852104664, "percentage": 1.41, "elapsed_time": "0:19:53", "remaining_time": "23:09:01"}
+{"current_steps": 32, "total_steps": 2195, "loss": 0.6514, "lr": 5.8181818181818185e-06, "epoch": 0.07281001137656427, "percentage": 1.46, "elapsed_time": "0:20:31", "remaining_time": "23:07:36"}
+{"current_steps": 33, "total_steps": 2195, "loss": 0.6011, "lr": 6e-06, "epoch": 0.07508532423208192, "percentage": 1.5, "elapsed_time": "0:21:11", "remaining_time": "23:08:44"}
+{"current_steps": 34, "total_steps": 2195, "loss": 0.589, "lr": 6.181818181818182e-06, "epoch": 0.07736063708759955, "percentage": 1.55, "elapsed_time": "0:21:43", "remaining_time": "23:00:39"}
+{"current_steps": 35, "total_steps": 2195, "loss": 0.5828, "lr": 6.363636363636364e-06, "epoch": 0.07963594994311718, "percentage": 1.59, "elapsed_time": "0:22:23", "remaining_time": "23:01:48"}
+{"current_steps": 36, "total_steps": 2195, "loss": 0.5921, "lr": 6.545454545454546e-06, "epoch": 0.08191126279863481, "percentage": 1.64, "elapsed_time": "0:23:08", "remaining_time": "23:07:33"}
+{"current_steps": 37, "total_steps": 2195, "loss": 0.5817, "lr": 6.7272727272727275e-06, "epoch": 0.08418657565415244, "percentage": 1.69, "elapsed_time": "0:23:41", "remaining_time": "23:02:15"}
+{"current_steps": 38, "total_steps": 2195, "loss": 0.5872, "lr": 6.90909090909091e-06, "epoch": 0.08646188850967008, "percentage": 1.73, "elapsed_time": "0:24:22", "remaining_time": "23:03:08"}
+{"current_steps": 39, "total_steps": 2195, "loss": 0.6163, "lr": 7.0909090909090916e-06, "epoch": 0.08873720136518772, "percentage": 1.78, "elapsed_time": "0:24:59", "remaining_time": "23:01:11"}
+{"current_steps": 40, "total_steps": 2195, "loss": 0.596, "lr": 7.272727272727273e-06, "epoch": 0.09101251422070535, "percentage": 1.82, "elapsed_time": "0:25:34", "remaining_time": "22:57:59"}
+{"current_steps": 41, "total_steps": 2195, "loss": 0.5605, "lr": 7.454545454545456e-06, "epoch": 0.09328782707622298, "percentage": 1.87, "elapsed_time": "0:26:19", "remaining_time": "23:02:44"}
+{"current_steps": 42, "total_steps": 2195, "loss": 0.5656, "lr": 7.636363636363638e-06, "epoch": 0.09556313993174062, "percentage": 1.91, "elapsed_time": "0:26:57", "remaining_time": "23:01:52"}
+{"current_steps": 43, "total_steps": 2195, "loss": 0.5677, "lr": 7.81818181818182e-06, "epoch": 0.09783845278725825, "percentage": 1.96, "elapsed_time": "0:27:35", "remaining_time": "23:00:29"}
+{"current_steps": 44, "total_steps": 2195, "loss": 0.5768, "lr": 8.000000000000001e-06, "epoch": 0.10011376564277588, "percentage": 2.0, "elapsed_time": "0:28:13", "remaining_time": "22:59:39"}
+{"current_steps": 45, "total_steps": 2195, "loss": 0.571, "lr": 8.181818181818183e-06, "epoch": 0.10238907849829351, "percentage": 2.05, "elapsed_time": "0:28:52", "remaining_time": "22:59:14"}
+{"current_steps": 46, "total_steps": 2195, "loss": 0.5502, "lr": 8.363636363636365e-06, "epoch": 0.10466439135381114, "percentage": 2.1, "elapsed_time": "0:29:26", "remaining_time": "22:55:41"}
+{"current_steps": 47, "total_steps": 2195, "loss": 0.5612, "lr": 8.545454545454546e-06, "epoch": 0.10693970420932879, "percentage": 2.14, "elapsed_time": "0:30:05", "remaining_time": "22:55:36"}
+{"current_steps": 48, "total_steps": 2195, "loss": 0.5941, "lr": 8.727272727272728e-06, "epoch": 0.10921501706484642, "percentage": 2.19, "elapsed_time": "0:30:47", "remaining_time": "22:56:56"}
+{"current_steps": 49, "total_steps": 2195, "loss": 0.5478, "lr": 8.90909090909091e-06, "epoch": 0.11149032992036405, "percentage": 2.23, "elapsed_time": "0:31:24", "remaining_time": "22:55:34"}
+{"current_steps": 50, "total_steps": 2195, "loss": 0.568, "lr": 9.090909090909091e-06, "epoch": 0.11376564277588168, "percentage": 2.28, "elapsed_time": "0:31:56", "remaining_time": "22:50:24"}
+{"current_steps": 51, "total_steps": 2195, "loss": 0.5352, "lr": 9.272727272727273e-06, "epoch": 0.11604095563139932, "percentage": 2.32, "elapsed_time": "0:32:33", "remaining_time": "22:49:02"}
+{"current_steps": 52, "total_steps": 2195, "loss": 0.573, "lr": 9.454545454545456e-06, "epoch": 0.11831626848691695, "percentage": 2.37, "elapsed_time": "0:33:08", "remaining_time": "22:46:01"}
+{"current_steps": 53, "total_steps": 2195, "loss": 0.5716, "lr": 9.636363636363638e-06, "epoch": 0.12059158134243458, "percentage": 2.41, "elapsed_time": "0:33:45", "remaining_time": "22:44:23"}
+{"current_steps": 54, "total_steps": 2195, "loss": 0.5671, "lr": 9.81818181818182e-06, "epoch": 0.12286689419795221, "percentage": 2.46, "elapsed_time": "0:34:22", "remaining_time": "22:43:08"}
+{"current_steps": 55, "total_steps": 2195, "loss": 0.5649, "lr": 1e-05, "epoch": 0.12514220705346984, "percentage": 2.51, "elapsed_time": "0:34:53", "remaining_time": "22:37:36"}
+{"current_steps": 56, "total_steps": 2195, "loss": 0.5418, "lr": 1.0181818181818182e-05, "epoch": 0.12741751990898748, "percentage": 2.55, "elapsed_time": "0:35:21", "remaining_time": "22:30:32"}
+{"current_steps": 57, "total_steps": 2195, "loss": 0.5346, "lr": 1.0363636363636364e-05, "epoch": 0.1296928327645051, "percentage": 2.6, "elapsed_time": "0:35:57", "remaining_time": "22:29:00"}
+{"current_steps": 58, "total_steps": 2195, "loss": 0.5503, "lr": 1.0545454545454546e-05, "epoch": 0.13196814562002276, "percentage": 2.64, "elapsed_time": "0:36:40", "remaining_time": "22:31:01"}
+{"current_steps": 59, "total_steps": 2195, "loss": 0.5641, "lr": 1.0727272727272729e-05, "epoch": 0.1342434584755404, "percentage": 2.69, "elapsed_time": "0:37:18", "remaining_time": "22:30:40"}
+{"current_steps": 60, "total_steps": 2195, "loss": 0.5555, "lr": 1.0909090909090909e-05, "epoch": 0.13651877133105803, "percentage": 2.73, "elapsed_time": "0:37:49", "remaining_time": "22:25:39"}
+{"current_steps": 61, "total_steps": 2195, "loss": 0.544, "lr": 1.1090909090909092e-05, "epoch": 0.13879408418657566, "percentage": 2.78, "elapsed_time": "0:38:22", "remaining_time": "22:22:28"}
+{"current_steps": 62, "total_steps": 2195, "loss": 0.5564, "lr": 1.1272727272727272e-05, "epoch": 0.1410693970420933, "percentage": 2.82, "elapsed_time": "0:38:54", "remaining_time": "22:18:19"}
+{"current_steps": 63, "total_steps": 2195, "loss": 0.5765, "lr": 1.1454545454545455e-05, "epoch": 0.14334470989761092, "percentage": 2.87, "elapsed_time": "0:39:33", "remaining_time": "22:18:45"}
+{"current_steps": 64, "total_steps": 2195, "loss": 0.5618, "lr": 1.1636363636363637e-05, "epoch": 0.14562002275312855, "percentage": 2.92, "elapsed_time": "0:40:13", "remaining_time": "22:19:32"}
+{"current_steps": 65, "total_steps": 2195, "loss": 0.53, "lr": 1.181818181818182e-05, "epoch": 0.14789533560864618, "percentage": 2.96, "elapsed_time": "0:40:47", "remaining_time": "22:16:58"}
+{"current_steps": 66, "total_steps": 2195, "loss": 0.5719, "lr": 1.2e-05, "epoch": 0.15017064846416384, "percentage": 3.01, "elapsed_time": "0:41:24", "remaining_time": "22:15:46"}
+{"current_steps": 67, "total_steps": 2195, "loss": 0.5512, "lr": 1.2181818181818184e-05, "epoch": 0.15244596131968147, "percentage": 3.05, "elapsed_time": "0:42:00", "remaining_time": "22:14:07"}
+{"current_steps": 68, "total_steps": 2195, "loss": 0.5624, "lr": 1.2363636363636364e-05, "epoch": 0.1547212741751991, "percentage": 3.1, "elapsed_time": "0:42:36", "remaining_time": "22:12:44"}
+{"current_steps": 69, "total_steps": 2195, "loss": 0.5481, "lr": 1.2545454545454547e-05, "epoch": 0.15699658703071673, "percentage": 3.14, "elapsed_time": "0:43:06", "remaining_time": "22:08:21"}
+{"current_steps": 70, "total_steps": 2195, "loss": 0.5932, "lr": 1.2727272727272728e-05, "epoch": 0.15927189988623436, "percentage": 3.19, "elapsed_time": "0:43:41", "remaining_time": "22:06:28"}
+{"current_steps": 71, "total_steps": 2195, "loss": 0.5337, "lr": 1.2909090909090912e-05, "epoch": 0.161547212741752, "percentage": 3.23, "elapsed_time": "0:44:19", "remaining_time": "22:05:54"}
+{"current_steps": 72, "total_steps": 2195, "loss": 0.5842, "lr": 1.3090909090909092e-05, "epoch": 0.16382252559726962, "percentage": 3.28, "elapsed_time": "0:45:03", "remaining_time": "22:08:37"}
+{"current_steps": 73, "total_steps": 2195, "loss": 0.5668, "lr": 1.3272727272727275e-05, "epoch": 0.16609783845278725, "percentage": 3.33, "elapsed_time": "0:45:39", "remaining_time": "22:07:16"}
+{"current_steps": 74, "total_steps": 2195, "loss": 0.5796, "lr": 1.3454545454545455e-05, "epoch": 0.16837315130830488, "percentage": 3.37, "elapsed_time": "0:46:20", "remaining_time": "22:08:20"}
+{"current_steps": 75, "total_steps": 2195, "loss": 0.5466, "lr": 1.3636363636363637e-05, "epoch": 0.17064846416382254, "percentage": 3.42, "elapsed_time": "0:47:01", "remaining_time": "22:09:05"}
+{"current_steps": 76, "total_steps": 2195, "loss": 0.5137, "lr": 1.381818181818182e-05, "epoch": 0.17292377701934017, "percentage": 3.46, "elapsed_time": "0:47:41", "remaining_time": "22:09:53"}
+{"current_steps": 77, "total_steps": 2195, "loss": 0.5271, "lr": 1.4e-05, "epoch": 0.1751990898748578, "percentage": 3.51, "elapsed_time": "0:48:16", "remaining_time": "22:07:53"}
+{"current_steps": 78, "total_steps": 2195, "loss": 0.5275, "lr": 1.4181818181818183e-05, "epoch": 0.17747440273037543, "percentage": 3.55, "elapsed_time": "0:48:54", "remaining_time": "22:07:15"}
+{"current_steps": 79, "total_steps": 2195, "loss": 0.5706, "lr": 1.4363636363636365e-05, "epoch": 0.17974971558589306, "percentage": 3.6, "elapsed_time": "0:49:33", "remaining_time": "22:07:35"}
+{"current_steps": 80, "total_steps": 2195, "loss": 0.5656, "lr": 1.4545454545454546e-05, "epoch": 0.1820250284414107, "percentage": 3.64, "elapsed_time": "0:50:10", "remaining_time": "22:06:34"}
+{"current_steps": 81, "total_steps": 2195, "loss": 0.5197, "lr": 1.4727272727272728e-05, "epoch": 0.18430034129692832, "percentage": 3.69, "elapsed_time": "0:50:50", "remaining_time": "22:06:59"}
+{"current_steps": 82, "total_steps": 2195, "loss": 0.4971, "lr": 1.4909090909090911e-05, "epoch": 0.18657565415244595, "percentage": 3.74, "elapsed_time": "0:51:25", "remaining_time": "22:04:57"}
+{"current_steps": 83, "total_steps": 2195, "loss": 0.5307, "lr": 1.5090909090909091e-05, "epoch": 0.18885096700796358, "percentage": 3.78, "elapsed_time": "0:52:06", "remaining_time": "22:05:58"}
+{"current_steps": 84, "total_steps": 2195, "loss": 0.5316, "lr": 1.5272727272727276e-05, "epoch": 0.19112627986348124, "percentage": 3.83, "elapsed_time": "0:52:41", "remaining_time": "22:04:10"}
+{"current_steps": 85, "total_steps": 2195, "loss": 0.5406, "lr": 1.5454545454545454e-05, "epoch": 0.19340159271899887, "percentage": 3.87, "elapsed_time": "0:53:14", "remaining_time": "22:01:39"}
+{"current_steps": 86, "total_steps": 2195, "loss": 0.5359, "lr": 1.563636363636364e-05, "epoch": 0.1956769055745165, "percentage": 3.92, "elapsed_time": "0:53:46", "remaining_time": "21:58:54"}
+{"current_steps": 87, "total_steps": 2195, "loss": 0.5494, "lr": 1.5818181818181818e-05, "epoch": 0.19795221843003413, "percentage": 3.96, "elapsed_time": "0:54:24", "remaining_time": "21:58:22"}
+{"current_steps": 88, "total_steps": 2195, "loss": 0.5519, "lr": 1.6000000000000003e-05, "epoch": 0.20022753128555176, "percentage": 4.01, "elapsed_time": "0:55:06", "remaining_time": "21:59:34"}
+{"current_steps": 89, "total_steps": 2195, "loss": 0.5491, "lr": 1.6181818181818184e-05, "epoch": 0.2025028441410694, "percentage": 4.05, "elapsed_time": "0:55:46", "remaining_time": "21:59:48"}
+{"current_steps": 90, "total_steps": 2195, "loss": 0.57, "lr": 1.6363636363636366e-05, "epoch": 0.20477815699658702, "percentage": 4.1, "elapsed_time": "0:56:25", "remaining_time": "21:59:33"}
+{"current_steps": 91, "total_steps": 2195, "loss": 0.5323, "lr": 1.6545454545454548e-05, "epoch": 0.20705346985210465, "percentage": 4.15, "elapsed_time": "0:57:02", "remaining_time": "21:58:41"}
+{"current_steps": 92, "total_steps": 2195, "loss": 0.5033, "lr": 1.672727272727273e-05, "epoch": 0.20932878270762229, "percentage": 4.19, "elapsed_time": "0:57:37", "remaining_time": "21:57:19"}
+{"current_steps": 93, "total_steps": 2195, "loss": 0.5387, "lr": 1.690909090909091e-05, "epoch": 0.21160409556313994, "percentage": 4.24, "elapsed_time": "0:58:13", "remaining_time": "21:56:06"}
+{"current_steps": 94, "total_steps": 2195, "loss": 0.4999, "lr": 1.7090909090909092e-05, "epoch": 0.21387940841865757, "percentage": 4.28, "elapsed_time": "0:58:51", "remaining_time": "21:55:32"}
+{"current_steps": 95, "total_steps": 2195, "loss": 0.5254, "lr": 1.7272727272727274e-05, "epoch": 0.2161547212741752, "percentage": 4.33, "elapsed_time": "0:59:27", "remaining_time": "21:54:10"}
+{"current_steps": 96, "total_steps": 2195, "loss": 0.5215, "lr": 1.7454545454545456e-05, "epoch": 0.21843003412969283, "percentage": 4.37, "elapsed_time": "1:00:09", "remaining_time": "21:55:17"}
+{"current_steps": 97, "total_steps": 2195, "loss": 0.5283, "lr": 1.7636363636363637e-05, "epoch": 0.22070534698521047, "percentage": 4.42, "elapsed_time": "1:00:44", "remaining_time": "21:53:39"}
+{"current_steps": 98, "total_steps": 2195, "loss": 0.5077, "lr": 1.781818181818182e-05, "epoch": 0.2229806598407281, "percentage": 4.46, "elapsed_time": "1:01:21", "remaining_time": "21:53:00"}
+{"current_steps": 99, "total_steps": 2195, "loss": 0.5314, "lr": 1.8e-05, "epoch": 0.22525597269624573, "percentage": 4.51, "elapsed_time": "1:01:54", "remaining_time": "21:50:51"}
+{"current_steps": 100, "total_steps": 2195, "loss": 0.5608, "lr": 1.8181818181818182e-05, "epoch": 0.22753128555176336, "percentage": 4.56, "elapsed_time": "1:02:30", "remaining_time": "21:49:34"}
+{"current_steps": 101, "total_steps": 2195, "loss": 0.5294, "lr": 1.8363636363636367e-05, "epoch": 0.229806598407281, "percentage": 4.6, "elapsed_time": "1:03:00", "remaining_time": "21:46:14"}
+{"current_steps": 102, "total_steps": 2195, "loss": 0.5217, "lr": 1.8545454545454545e-05, "epoch": 0.23208191126279865, "percentage": 4.65, "elapsed_time": "1:03:38", "remaining_time": "21:45:50"}
+{"current_steps": 103, "total_steps": 2195, "loss": 0.5735, "lr": 1.872727272727273e-05, "epoch": 0.23435722411831628, "percentage": 4.69, "elapsed_time": "1:04:09", "remaining_time": "21:43:00"}
+{"current_steps": 104, "total_steps": 2195, "loss": 0.4806, "lr": 1.8909090909090912e-05, "epoch": 0.2366325369738339, "percentage": 4.74, "elapsed_time": "1:04:44", "remaining_time": "21:41:39"}
+{"current_steps": 105, "total_steps": 2195, "loss": 0.5464, "lr": 1.9090909090909094e-05, "epoch": 0.23890784982935154, "percentage": 4.78, "elapsed_time": "1:05:18", "remaining_time": "21:40:05"}
+{"current_steps": 106, "total_steps": 2195, "loss": 0.5456, "lr": 1.9272727272727275e-05, "epoch": 0.24118316268486917, "percentage": 4.83, "elapsed_time": "1:05:51", "remaining_time": "21:37:55"}
+{"current_steps": 107, "total_steps": 2195, "loss": 0.5155, "lr": 1.9454545454545457e-05, "epoch": 0.2434584755403868, "percentage": 4.87, "elapsed_time": "1:06:25", "remaining_time": "21:36:16"}
+{"current_steps": 108, "total_steps": 2195, "loss": 0.5728, "lr": 1.963636363636364e-05, "epoch": 0.24573378839590443, "percentage": 4.92, "elapsed_time": "1:07:11", "remaining_time": "21:38:27"}
+{"current_steps": 109, "total_steps": 2195, "loss": 0.499, "lr": 1.981818181818182e-05, "epoch": 0.24800910125142206, "percentage": 4.97, "elapsed_time": "1:07:48", "remaining_time": "21:37:40"}
+{"current_steps": 110, "total_steps": 2195, "loss": 0.5033, "lr": 2e-05, "epoch": 0.2502844141069397, "percentage": 5.01, "elapsed_time": "1:08:22", "remaining_time": "21:36:01"}
+{"current_steps": 111, "total_steps": 2195, "loss": 0.5017, "lr": 2.0181818181818183e-05, "epoch": 0.2525597269624573, "percentage": 5.06, "elapsed_time": "1:08:59", "remaining_time": "21:35:15"}
+{"current_steps": 112, "total_steps": 2195, "loss": 0.5251, "lr": 2.0363636363636365e-05, "epoch": 0.25483503981797495, "percentage": 5.1, "elapsed_time": "1:09:35", "remaining_time": "21:34:09"}
+{"current_steps": 113, "total_steps": 2195, "loss": 0.5551, "lr": 2.054545454545455e-05, "epoch": 0.2571103526734926, "percentage": 5.15, "elapsed_time": "1:10:17", "remaining_time": "21:35:03"}
+{"current_steps": 114, "total_steps": 2195, "loss": 0.5354, "lr": 2.0727272727272728e-05, "epoch": 0.2593856655290102, "percentage": 5.19, "elapsed_time": "1:10:58", "remaining_time": "21:35:28"}
+{"current_steps": 115, "total_steps": 2195, "loss": 0.5264, "lr": 2.090909090909091e-05, "epoch": 0.2616609783845279, "percentage": 5.24, "elapsed_time": "1:11:31", "remaining_time": "21:33:40"}
+{"current_steps": 116, "total_steps": 2195, "loss": 0.5196, "lr": 2.109090909090909e-05, "epoch": 0.26393629124004553, "percentage": 5.28, "elapsed_time": "1:12:11", "remaining_time": "21:33:45"}
+{"current_steps": 117, "total_steps": 2195, "loss": 0.5304, "lr": 2.1272727272727276e-05, "epoch": 0.26621160409556316, "percentage": 5.33, "elapsed_time": "1:12:53", "remaining_time": "21:34:39"}
+{"current_steps": 118, "total_steps": 2195, "loss": 0.5259, "lr": 2.1454545454545458e-05, "epoch": 0.2684869169510808, "percentage": 5.38, "elapsed_time": "1:13:29", "remaining_time": "21:33:36"}
+{"current_steps": 119, "total_steps": 2195, "loss": 0.544, "lr": 2.1636363636363636e-05, "epoch": 0.2707622298065984, "percentage": 5.42, "elapsed_time": "1:14:05", "remaining_time": "21:32:41"}
+{"current_steps": 120, "total_steps": 2195, "loss": 0.5183, "lr": 2.1818181818181818e-05, "epoch": 0.27303754266211605, "percentage": 5.47, "elapsed_time": "1:14:46", "remaining_time": "21:32:54"}
+{"current_steps": 121, "total_steps": 2195, "loss": 0.5378, "lr": 2.2000000000000003e-05, "epoch": 0.2753128555176337, "percentage": 5.51, "elapsed_time": "1:15:22", "remaining_time": "21:31:50"}
+{"current_steps": 122, "total_steps": 2195, "loss": 0.5439, "lr": 2.2181818181818184e-05, "epoch": 0.2775881683731513, "percentage": 5.56, "elapsed_time": "1:16:02", "remaining_time": "21:32:10"}
+{"current_steps": 123, "total_steps": 2195, "loss": 0.5456, "lr": 2.2363636363636366e-05, "epoch": 0.27986348122866894, "percentage": 5.6, "elapsed_time": "1:16:37", "remaining_time": "21:30:49"}
+{"current_steps": 124, "total_steps": 2195, "loss": 0.498, "lr": 2.2545454545454544e-05, "epoch": 0.2821387940841866, "percentage": 5.65, "elapsed_time": "1:17:17", "remaining_time": "21:30:46"}
+{"current_steps": 125, "total_steps": 2195, "loss": 0.4867, "lr": 2.2727272727272733e-05, "epoch": 0.2844141069397042, "percentage": 5.69, "elapsed_time": "1:17:57", "remaining_time": "21:31:02"}
+{"current_steps": 126, "total_steps": 2195, "loss": 0.5054, "lr": 2.290909090909091e-05, "epoch": 0.28668941979522183, "percentage": 5.74, "elapsed_time": "1:18:30", "remaining_time": "21:29:16"}
+{"current_steps": 127, "total_steps": 2195, "loss": 0.5057, "lr": 2.3090909090909093e-05, "epoch": 0.28896473265073946, "percentage": 5.79, "elapsed_time": "1:19:07", "remaining_time": "21:28:17"}
+{"current_steps": 128, "total_steps": 2195, "loss": 0.4932, "lr": 2.3272727272727274e-05, "epoch": 0.2912400455062571, "percentage": 5.83, "elapsed_time": "1:19:34", "remaining_time": "21:25:00"}
+{"current_steps": 129, "total_steps": 2195, "loss": 0.5352, "lr": 2.3454545454545456e-05, "epoch": 0.2935153583617747, "percentage": 5.88, "elapsed_time": "1:20:12", "remaining_time": "21:24:36"}
+{"current_steps": 130, "total_steps": 2195, "loss": 0.5173, "lr": 2.363636363636364e-05, "epoch": 0.29579067121729236, "percentage": 5.92, "elapsed_time": "1:20:53", "remaining_time": "21:24:48"}
+{"current_steps": 131, "total_steps": 2195, "loss": 0.5169, "lr": 2.381818181818182e-05, "epoch": 0.29806598407281, "percentage": 5.97, "elapsed_time": "1:21:33", "remaining_time": "21:24:58"}
+{"current_steps": 132, "total_steps": 2195, "loss": 0.5314, "lr": 2.4e-05, "epoch": 0.3003412969283277, "percentage": 6.01, "elapsed_time": "1:22:08", "remaining_time": "21:23:44"}
+{"current_steps": 133, "total_steps": 2195, "loss": 0.5533, "lr": 2.4181818181818182e-05, "epoch": 0.3026166097838453, "percentage": 6.06, "elapsed_time": "1:22:48", "remaining_time": "21:23:55"}
+{"current_steps": 134, "total_steps": 2195, "loss": 0.5511, "lr": 2.4363636363636367e-05, "epoch": 0.30489192263936293, "percentage": 6.1, "elapsed_time": "1:23:31", "remaining_time": "21:24:35"}
+{"current_steps": 135, "total_steps": 2195, "loss": 0.5384, "lr": 2.454545454545455e-05, "epoch": 0.30716723549488056, "percentage": 6.15, "elapsed_time": "1:24:13", "remaining_time": "21:25:15"}
+{"current_steps": 136, "total_steps": 2195, "loss": 0.5401, "lr": 2.4727272727272727e-05, "epoch": 0.3094425483503982, "percentage": 6.2, "elapsed_time": "1:24:54", "remaining_time": "21:25:29"}
+{"current_steps": 137, "total_steps": 2195, "loss": 0.5068, "lr": 2.490909090909091e-05, "epoch": 0.3117178612059158, "percentage": 6.24, "elapsed_time": "1:25:30", "remaining_time": "21:24:29"}
+{"current_steps": 138, "total_steps": 2195, "loss": 0.5472, "lr": 2.5090909090909094e-05, "epoch": 0.31399317406143346, "percentage": 6.29, "elapsed_time": "1:26:06", "remaining_time": "21:23:31"}
+{"current_steps": 139, "total_steps": 2195, "loss": 0.5413, "lr": 2.5272727272727275e-05, "epoch": 0.3162684869169511, "percentage": 6.33, "elapsed_time": "1:26:39", "remaining_time": "21:21:54"}
+{"current_steps": 140, "total_steps": 2195, "loss": 0.5188, "lr": 2.5454545454545457e-05, "epoch": 0.3185437997724687, "percentage": 6.38, "elapsed_time": "1:27:15", "remaining_time": "21:20:54"}
+{"current_steps": 141, "total_steps": 2195, "loss": 0.4928, "lr": 2.563636363636364e-05, "epoch": 0.32081911262798635, "percentage": 6.42, "elapsed_time": "1:27:51", "remaining_time": "21:19:56"}
+{"current_steps": 142, "total_steps": 2195, "loss": 0.5598, "lr": 2.5818181818181824e-05, "epoch": 0.323094425483504, "percentage": 6.47, "elapsed_time": "1:28:32", "remaining_time": "21:20:02"}
+{"current_steps": 143, "total_steps": 2195, "loss": 0.5543, "lr": 2.6000000000000002e-05, "epoch": 0.3253697383390216, "percentage": 6.51, "elapsed_time": "1:29:13", "remaining_time": "21:20:27"}
+{"current_steps": 144, "total_steps": 2195, "loss": 0.4953, "lr": 2.6181818181818183e-05, "epoch": 0.32764505119453924, "percentage": 6.56, "elapsed_time": "1:29:49", "remaining_time": "21:19:18"}
+{"current_steps": 145, "total_steps": 2195, "loss": 0.5151, "lr": 2.6363636363636365e-05, "epoch": 0.32992036405005687, "percentage": 6.61, "elapsed_time": "1:30:21", "remaining_time": "21:17:22"}
+{"current_steps": 146, "total_steps": 2195, "loss": 0.5163, "lr": 2.654545454545455e-05, "epoch": 0.3321956769055745, "percentage": 6.65, "elapsed_time": "1:30:54", "remaining_time": "21:15:44"}
+{"current_steps": 147, "total_steps": 2195, "loss": 0.5566, "lr": 2.672727272727273e-05, "epoch": 0.33447098976109213, "percentage": 6.7, "elapsed_time": "1:31:32", "remaining_time": "21:15:18"}
+{"current_steps": 148, "total_steps": 2195, "loss": 0.5371, "lr": 2.690909090909091e-05, "epoch": 0.33674630261660976, "percentage": 6.74, "elapsed_time": "1:32:17", "remaining_time": "21:16:29"}
+{"current_steps": 149, "total_steps": 2195, "loss": 0.5233, "lr": 2.709090909090909e-05, "epoch": 0.3390216154721274, "percentage": 6.79, "elapsed_time": "1:32:51", "remaining_time": "21:15:07"}
+{"current_steps": 150, "total_steps": 2195, "loss": 0.516, "lr": 2.7272727272727273e-05, "epoch": 0.3412969283276451, "percentage": 6.83, "elapsed_time": "1:33:27", "remaining_time": "21:14:03"}
+{"current_steps": 151, "total_steps": 2195, "loss": 0.503, "lr": 2.7454545454545458e-05, "epoch": 0.3435722411831627, "percentage": 6.88, "elapsed_time": "1:34:08", "remaining_time": "21:14:16"}
+{"current_steps": 152, "total_steps": 2195, "loss": 0.5842, "lr": 2.763636363636364e-05, "epoch": 0.34584755403868034, "percentage": 6.92, "elapsed_time": "1:34:48", "remaining_time": "21:14:14"}
+{"current_steps": 153, "total_steps": 2195, "loss": 0.4992, "lr": 2.781818181818182e-05, "epoch": 0.34812286689419797, "percentage": 6.97, "elapsed_time": "1:35:20", "remaining_time": "21:12:30"}
+{"current_steps": 154, "total_steps": 2195, "loss": 0.544, "lr": 2.8e-05, "epoch": 0.3503981797497156, "percentage": 7.02, "elapsed_time": "1:36:02", "remaining_time": "21:12:58"}
+{"current_steps": 155, "total_steps": 2195, "loss": 0.5255, "lr": 2.8181818181818185e-05, "epoch": 0.35267349260523323, "percentage": 7.06, "elapsed_time": "1:36:38", "remaining_time": "21:12:00"}
+{"current_steps": 156, "total_steps": 2195, "loss": 0.5192, "lr": 2.8363636363636366e-05, "epoch": 0.35494880546075086, "percentage": 7.11, "elapsed_time": "1:37:15", "remaining_time": "21:11:11"}
+{"current_steps": 157, "total_steps": 2195, "loss": 0.4928, "lr": 2.8545454545454548e-05, "epoch": 0.3572241183162685, "percentage": 7.15, "elapsed_time": "1:37:53", "remaining_time": "21:10:49"}
+{"current_steps": 158, "total_steps": 2195, "loss": 0.4826, "lr": 2.872727272727273e-05, "epoch": 0.3594994311717861, "percentage": 7.2, "elapsed_time": "1:38:28", "remaining_time": "21:09:40"}
+{"current_steps": 159, "total_steps": 2195, "loss": 0.4954, "lr": 2.8909090909090914e-05, "epoch": 0.36177474402730375, "percentage": 7.24, "elapsed_time": "1:39:00", "remaining_time": "21:07:52"}
+{"current_steps": 160, "total_steps": 2195, "loss": 0.4845, "lr": 2.9090909090909093e-05, "epoch": 0.3640500568828214, "percentage": 7.29, "elapsed_time": "1:39:35", "remaining_time": "21:06:35"}
+{"current_steps": 161, "total_steps": 2195, "loss": 0.5241, "lr": 2.9272727272727274e-05, "epoch": 0.366325369738339, "percentage": 7.33, "elapsed_time": "1:40:17", "remaining_time": "21:07:07"}
+{"current_steps": 162, "total_steps": 2195, "loss": 0.4895, "lr": 2.9454545454545456e-05, "epoch": 0.36860068259385664, "percentage": 7.38, "elapsed_time": "1:40:52", "remaining_time": "21:05:55"}
+{"current_steps": 163, "total_steps": 2195, "loss": 0.5158, "lr": 2.963636363636364e-05, "epoch": 0.3708759954493743, "percentage": 7.43, "elapsed_time": "1:41:32", "remaining_time": "21:05:53"}
+{"current_steps": 164, "total_steps": 2195, "loss": 0.5131, "lr": 2.9818181818181823e-05, "epoch": 0.3731513083048919, "percentage": 7.47, "elapsed_time": "1:42:03", "remaining_time": "21:03:58"}
+{"current_steps": 165, "total_steps": 2195, "loss": 0.5551, "lr": 3.0000000000000004e-05, "epoch": 0.37542662116040953, "percentage": 7.52, "elapsed_time": "1:42:39", "remaining_time": "21:03:03"}
+{"current_steps": 166, "total_steps": 2195, "loss": 0.5128, "lr": 3.0181818181818182e-05, "epoch": 0.37770193401592717, "percentage": 7.56, "elapsed_time": "1:43:23", "remaining_time": "21:03:42"}
+{"current_steps": 167, "total_steps": 2195, "loss": 0.4988, "lr": 3.0363636363636364e-05, "epoch": 0.3799772468714448, "percentage": 7.61, "elapsed_time": "1:44:03", "remaining_time": "21:03:38"}
+{"current_steps": 168, "total_steps": 2195, "loss": 0.5443, "lr": 3.054545454545455e-05, "epoch": 0.3822525597269625, "percentage": 7.65, "elapsed_time": "1:44:44", "remaining_time": "21:03:47"}
+{"current_steps": 169, "total_steps": 2195, "loss": 0.5538, "lr": 3.072727272727273e-05, "epoch": 0.3845278725824801, "percentage": 7.7, "elapsed_time": "1:45:23", "remaining_time": "21:03:25"}
+{"current_steps": 170, "total_steps": 2195, "loss": 0.5025, "lr": 3.090909090909091e-05, "epoch": 0.38680318543799774, "percentage": 7.74, "elapsed_time": "1:45:49", "remaining_time": "21:00:37"}
+{"current_steps": 171, "total_steps": 2195, "loss": 0.5192, "lr": 3.1090909090909094e-05, "epoch": 0.3890784982935154, "percentage": 7.79, "elapsed_time": "1:46:21", "remaining_time": "20:58:50"}
+{"current_steps": 172, "total_steps": 2195, "loss": 0.5452, "lr": 3.127272727272728e-05, "epoch": 0.391353811149033, "percentage": 7.84, "elapsed_time": "1:46:55", "remaining_time": "20:57:40"}
+{"current_steps": 173, "total_steps": 2195, "loss": 0.5392, "lr": 3.145454545454546e-05, "epoch": 0.39362912400455063, "percentage": 7.88, "elapsed_time": "1:47:29", "remaining_time": "20:56:22"}
+{"current_steps": 174, "total_steps": 2195, "loss": 0.538, "lr": 3.1636363636363635e-05, "epoch": 0.39590443686006827, "percentage": 7.93, "elapsed_time": "1:47:59", "remaining_time": "20:54:16"}
+{"current_steps": 175, "total_steps": 2195, "loss": 0.5047, "lr": 3.181818181818182e-05, "epoch": 0.3981797497155859, "percentage": 7.97, "elapsed_time": "1:48:38", "remaining_time": "20:54:06"}
+{"current_steps": 176, "total_steps": 2195, "loss": 0.5, "lr": 3.2000000000000005e-05, "epoch": 0.4004550625711035, "percentage": 8.02, "elapsed_time": "1:49:18", "remaining_time": "20:53:59"}
+{"current_steps": 177, "total_steps": 2195, "loss": 0.5271, "lr": 3.2181818181818184e-05, "epoch": 0.40273037542662116, "percentage": 8.06, "elapsed_time": "1:49:58", "remaining_time": "20:53:45"}
+{"current_steps": 178, "total_steps": 2195, "loss": 0.5053, "lr": 3.236363636363637e-05, "epoch": 0.4050056882821388, "percentage": 8.11, "elapsed_time": "1:50:39", "remaining_time": "20:54:00"}
+{"current_steps": 179, "total_steps": 2195, "loss": 0.526, "lr": 3.254545454545455e-05, "epoch": 0.4072810011376564, "percentage": 8.15, "elapsed_time": "1:51:16", "remaining_time": "20:53:11"}
+{"current_steps": 180, "total_steps": 2195, "loss": 0.5486, "lr": 3.272727272727273e-05, "epoch": 0.40955631399317405, "percentage": 8.2, "elapsed_time": "1:51:52", "remaining_time": "20:52:25"}
+{"current_steps": 181, "total_steps": 2195, "loss": 0.542, "lr": 3.290909090909091e-05, "epoch": 0.4118316268486917, "percentage": 8.25, "elapsed_time": "1:52:30", "remaining_time": "20:51:49"}
+{"current_steps": 182, "total_steps": 2195, "loss": 0.5414, "lr": 3.3090909090909095e-05, "epoch": 0.4141069397042093, "percentage": 8.29, "elapsed_time": "1:53:14", "remaining_time": "20:52:33"}
+{"current_steps": 183, "total_steps": 2195, "loss": 0.4837, "lr": 3.327272727272727e-05, "epoch": 0.41638225255972694, "percentage": 8.34, "elapsed_time": "1:53:51", "remaining_time": "20:51:44"}
+{"current_steps": 184, "total_steps": 2195, "loss": 0.5704, "lr": 3.345454545454546e-05, "epoch": 0.41865756541524457, "percentage": 8.38, "elapsed_time": "1:54:27", "remaining_time": "20:50:58"}
+{"current_steps": 185, "total_steps": 2195, "loss": 0.5087, "lr": 3.363636363636364e-05, "epoch": 0.42093287827076226, "percentage": 8.43, "elapsed_time": "1:54:59", "remaining_time": "20:49:17"}
+{"current_steps": 186, "total_steps": 2195, "loss": 0.5356, "lr": 3.381818181818182e-05, "epoch": 0.4232081911262799, "percentage": 8.47, "elapsed_time": "1:55:34", "remaining_time": "20:48:23"}
+{"current_steps": 187, "total_steps": 2195, "loss": 0.5236, "lr": 3.4e-05, "epoch": 0.4254835039817975, "percentage": 8.52, "elapsed_time": "1:56:17", "remaining_time": "20:48:47"}
+{"current_steps": 188, "total_steps": 2195, "loss": 0.4978, "lr": 3.4181818181818185e-05, "epoch": 0.42775881683731515, "percentage": 8.56, "elapsed_time": "1:56:51", "remaining_time": "20:47:36"}
+{"current_steps": 189, "total_steps": 2195, "loss": 0.5361, "lr": 3.436363636363637e-05, "epoch": 0.4300341296928328, "percentage": 8.61, "elapsed_time": "1:57:32", "remaining_time": "20:47:30"}
+{"current_steps": 190, "total_steps": 2195, "loss": 0.5032, "lr": 3.454545454545455e-05, "epoch": 0.4323094425483504, "percentage": 8.66, "elapsed_time": "1:58:12", "remaining_time": "20:47:23"}
+{"current_steps": 191, "total_steps": 2195, "loss": 0.5303, "lr": 3.4727272727272726e-05, "epoch": 0.43458475540386804, "percentage": 8.7, "elapsed_time": "1:58:52", "remaining_time": "20:47:14"}
+{"current_steps": 192, "total_steps": 2195, "loss": 0.481, "lr": 3.490909090909091e-05, "epoch": 0.43686006825938567, "percentage": 8.75, "elapsed_time": "1:59:32", "remaining_time": "20:47:09"}
+{"current_steps": 193, "total_steps": 2195, "loss": 0.5371, "lr": 3.5090909090909096e-05, "epoch": 0.4391353811149033, "percentage": 8.79, "elapsed_time": "2:00:07", "remaining_time": "20:45:59"}
+{"current_steps": 194, "total_steps": 2195, "loss": 0.5298, "lr": 3.5272727272727274e-05, "epoch": 0.44141069397042093, "percentage": 8.84, "elapsed_time": "2:00:43", "remaining_time": "20:45:17"}
+{"current_steps": 195, "total_steps": 2195, "loss": 0.4723, "lr": 3.545454545454546e-05, "epoch": 0.44368600682593856, "percentage": 8.88, "elapsed_time": "2:01:17", "remaining_time": "20:44:03"}
+{"current_steps": 196, "total_steps": 2195, "loss": 0.5182, "lr": 3.563636363636364e-05, "epoch": 0.4459613196814562, "percentage": 8.93, "elapsed_time": "2:01:56", "remaining_time": "20:43:39"}
+{"current_steps": 197, "total_steps": 2195, "loss": 0.5175, "lr": 3.581818181818182e-05, "epoch": 0.4482366325369738, "percentage": 8.97, "elapsed_time": "2:02:28", "remaining_time": "20:42:11"}
+{"current_steps": 198, "total_steps": 2195, "loss": 0.5099, "lr": 3.6e-05, "epoch": 0.45051194539249145, "percentage": 9.02, "elapsed_time": "2:03:06", "remaining_time": "20:41:41"}
+{"current_steps": 199, "total_steps": 2195, "loss": 0.5269, "lr": 3.6181818181818186e-05, "epoch": 0.4527872582480091, "percentage": 9.07, "elapsed_time": "2:03:42", "remaining_time": "20:40:50"}
+{"current_steps": 200, "total_steps": 2195, "loss": 0.4923, "lr": 3.6363636363636364e-05, "epoch": 0.4550625711035267, "percentage": 9.11, "elapsed_time": "2:04:18", "remaining_time": "20:39:55"}
+{"current_steps": 201, "total_steps": 2195, "loss": 0.5067, "lr": 3.654545454545455e-05, "epoch": 0.45733788395904434, "percentage": 9.16, "elapsed_time": "2:04:51", "remaining_time": "20:38:39"}
+{"current_steps": 202, "total_steps": 2195, "loss": 0.5372, "lr": 3.6727272727272734e-05, "epoch": 0.459613196814562, "percentage": 9.2, "elapsed_time": "2:05:27", "remaining_time": "20:37:49"}
+{"current_steps": 203, "total_steps": 2195, "loss": 0.4968, "lr": 3.690909090909091e-05, "epoch": 0.46188850967007966, "percentage": 9.25, "elapsed_time": "2:06:08", "remaining_time": "20:37:49"}
+{"current_steps": 204, "total_steps": 2195, "loss": 0.4992, "lr": 3.709090909090909e-05, "epoch": 0.4641638225255973, "percentage": 9.29, "elapsed_time": "2:06:42", "remaining_time": "20:36:34"}
+{"current_steps": 205, "total_steps": 2195, "loss": 0.4772, "lr": 3.7272727272727276e-05, "epoch": 0.4664391353811149, "percentage": 9.34, "elapsed_time": "2:07:27", "remaining_time": "20:37:13"}
+{"current_steps": 206, "total_steps": 2195, "loss": 0.5144, "lr": 3.745454545454546e-05, "epoch": 0.46871444823663255, "percentage": 9.38, "elapsed_time": "2:07:59", "remaining_time": "20:35:46"}
+{"current_steps": 207, "total_steps": 2195, "loss": 0.5449, "lr": 3.763636363636364e-05, "epoch": 0.4709897610921502, "percentage": 9.43, "elapsed_time": "2:08:38", "remaining_time": "20:35:28"}
+{"current_steps": 208, "total_steps": 2195, "loss": 0.4696, "lr": 3.7818181818181824e-05, "epoch": 0.4732650739476678, "percentage": 9.48, "elapsed_time": "2:09:21", "remaining_time": "20:35:45"}
+{"current_steps": 209, "total_steps": 2195, "loss": 0.4927, "lr": 3.8e-05, "epoch": 0.47554038680318544, "percentage": 9.52, "elapsed_time": "2:10:03", "remaining_time": "20:35:47"}
+{"current_steps": 210, "total_steps": 2195, "loss": 0.521, "lr": 3.818181818181819e-05, "epoch": 0.4778156996587031, "percentage": 9.57, "elapsed_time": "2:10:37", "remaining_time": "20:34:38"}
+{"current_steps": 211, "total_steps": 2195, "loss": 0.4953, "lr": 3.8363636363636365e-05, "epoch": 0.4800910125142207, "percentage": 9.61, "elapsed_time": "2:11:12", "remaining_time": "20:33:41"}
+{"current_steps": 212, "total_steps": 2195, "loss": 0.5362, "lr": 3.854545454545455e-05, "epoch": 0.48236632536973834, "percentage": 9.66, "elapsed_time": "2:11:48", "remaining_time": "20:32:54"}
+{"current_steps": 213, "total_steps": 2195, "loss": 0.4901, "lr": 3.872727272727273e-05, "epoch": 0.48464163822525597, "percentage": 9.7, "elapsed_time": "2:12:33", "remaining_time": "20:33:24"}
+{"current_steps": 214, "total_steps": 2195, "loss": 0.5112, "lr": 3.8909090909090914e-05, "epoch": 0.4869169510807736, "percentage": 9.75, "elapsed_time": "2:13:11", "remaining_time": "20:32:53"}
+{"current_steps": 215, "total_steps": 2195, "loss": 0.4931, "lr": 3.909090909090909e-05, "epoch": 0.4891922639362912, "percentage": 9.79, "elapsed_time": "2:13:42", "remaining_time": "20:31:21"}
+{"current_steps": 216, "total_steps": 2195, "loss": 0.4868, "lr": 3.927272727272728e-05, "epoch": 0.49146757679180886, "percentage": 9.84, "elapsed_time": "2:14:20", "remaining_time": "20:30:52"}
+{"current_steps": 217, "total_steps": 2195, "loss": 0.4902, "lr": 3.9454545454545455e-05, "epoch": 0.4937428896473265, "percentage": 9.89, "elapsed_time": "2:15:00", "remaining_time": "20:30:39"}
+{"current_steps": 218, "total_steps": 2195, "loss": 0.464, "lr": 3.963636363636364e-05, "epoch": 0.4960182025028441, "percentage": 9.93, "elapsed_time": "2:15:34", "remaining_time": "20:29:25"}
+{"current_steps": 219, "total_steps": 2195, "loss": 0.4919, "lr": 3.9818181818181825e-05, "epoch": 0.49829351535836175, "percentage": 9.98, "elapsed_time": "2:16:11", "remaining_time": "20:28:53"}
+{"current_steps": 220, "total_steps": 2195, "loss": 0.4965, "lr": 4e-05, "epoch": 0.5005688282138794, "percentage": 10.02, "elapsed_time": "2:16:42", "remaining_time": "20:27:19"}
+{"current_steps": 221, "total_steps": 2195, "loss": 0.53, "lr": 3.9999974697382296e-05, "epoch": 0.502844141069397, "percentage": 10.07, "elapsed_time": "2:17:25", "remaining_time": "20:27:30"}
+{"current_steps": 222, "total_steps": 2195, "loss": 0.5027, "lr": 3.9999898789593194e-05, "epoch": 0.5051194539249146, "percentage": 10.11, "elapsed_time": "2:18:05", "remaining_time": "20:27:13"}
+{"current_steps": 223, "total_steps": 2195, "loss": 0.508, "lr": 3.999977227682476e-05, "epoch": 0.5073947667804323, "percentage": 10.16, "elapsed_time": "2:18:44", "remaining_time": "20:26:55"}
+{"current_steps": 224, "total_steps": 2195, "loss": 0.5049, "lr": 3.99995951593971e-05, "epoch": 0.5096700796359499, "percentage": 10.21, "elapsed_time": "2:19:21", "remaining_time": "20:26:13"}
+{"current_steps": 225, "total_steps": 2195, "loss": 0.5252, "lr": 3.999936743775839e-05, "epoch": 0.5119453924914675, "percentage": 10.25, "elapsed_time": "2:20:00", "remaining_time": "20:25:52"}
+{"current_steps": 226, "total_steps": 2195, "loss": 0.4751, "lr": 3.999908911248481e-05, "epoch": 0.5142207053469852, "percentage": 10.3, "elapsed_time": "2:20:33", "remaining_time": "20:24:39"}
+{"current_steps": 227, "total_steps": 2195, "loss": 0.5008, "lr": 3.999876018428059e-05, "epoch": 0.5164960182025028, "percentage": 10.34, "elapsed_time": "2:21:13", "remaining_time": "20:24:21"}
+{"current_steps": 228, "total_steps": 2195, "loss": 0.5198, "lr": 3.999838065397801e-05, "epoch": 0.5187713310580204, "percentage": 10.39, "elapsed_time": "2:21:47", "remaining_time": "20:23:15"}
+{"current_steps": 229, "total_steps": 2195, "loss": 0.5272, "lr": 3.9997950522537385e-05, "epoch": 0.5210466439135382, "percentage": 10.43, "elapsed_time": "2:22:28", "remaining_time": "20:23:10"}
+{"current_steps": 230, "total_steps": 2195, "loss": 0.5225, "lr": 3.999746979104705e-05, "epoch": 0.5233219567690558, "percentage": 10.48, "elapsed_time": "2:23:00", "remaining_time": "20:21:48"}
+{"current_steps": 231, "total_steps": 2195, "loss": 0.5182, "lr": 3.999693846072339e-05, "epoch": 0.5255972696245734, "percentage": 10.52, "elapsed_time": "2:23:40", "remaining_time": "20:21:36"}
+{"current_steps": 232, "total_steps": 2195, "loss": 0.4981, "lr": 3.9996356532910814e-05, "epoch": 0.5278725824800911, "percentage": 10.57, "elapsed_time": "2:24:12", "remaining_time": "20:20:07"}
+{"current_steps": 233, "total_steps": 2195, "loss": 0.5603, "lr": 3.9995724009081745e-05, "epoch": 0.5301478953356087, "percentage": 10.62, "elapsed_time": "2:24:49", "remaining_time": "20:19:34"}
+{"current_steps": 234, "total_steps": 2195, "loss": 0.5583, "lr": 3.9995040890836635e-05, "epoch": 0.5324232081911263, "percentage": 10.66, "elapsed_time": "2:25:24", "remaining_time": "20:18:37"}
+{"current_steps": 235, "total_steps": 2195, "loss": 0.5082, "lr": 3.999430717990395e-05, "epoch": 0.534698521046644, "percentage": 10.71, "elapsed_time": "2:26:04", "remaining_time": "20:18:18"}
+{"current_steps": 236, "total_steps": 2195, "loss": 0.4989, "lr": 3.9993522878140174e-05, "epoch": 0.5369738339021616, "percentage": 10.75, "elapsed_time": "2:26:44", "remaining_time": "20:18:02"}
+{"current_steps": 237, "total_steps": 2195, "loss": 0.5234, "lr": 3.999268798752979e-05, "epoch": 0.5392491467576792, "percentage": 10.8, "elapsed_time": "2:27:19", "remaining_time": "20:17:11"}
+{"current_steps": 238, "total_steps": 2195, "loss": 0.556, "lr": 3.9991802510185296e-05, "epoch": 0.5415244596131968, "percentage": 10.84, "elapsed_time": "2:27:54", "remaining_time": "20:16:10"}
+{"current_steps": 239, "total_steps": 2195, "loss": 0.4887, "lr": 3.999086644834719e-05, "epoch": 0.5437997724687145, "percentage": 10.89, "elapsed_time": "2:28:31", "remaining_time": "20:15:32"}
+{"current_steps": 240, "total_steps": 2195, "loss": 0.4936, "lr": 3.998987980438393e-05, "epoch": 0.5460750853242321, "percentage": 10.93, "elapsed_time": "2:29:15", "remaining_time": "20:15:51"}
+{"current_steps": 241, "total_steps": 2195, "loss": 0.4627, "lr": 3.998884258079199e-05, "epoch": 0.5483503981797497, "percentage": 10.98, "elapsed_time": "2:29:46", "remaining_time": "20:14:23"}
+{"current_steps": 242, "total_steps": 2195, "loss": 0.5042, "lr": 3.998775478019584e-05, "epoch": 0.5506257110352674, "percentage": 11.03, "elapsed_time": "2:30:27", "remaining_time": "20:14:15"}
+{"current_steps": 243, "total_steps": 2195, "loss": 0.5115, "lr": 3.998661640534787e-05, "epoch": 0.552901023890785, "percentage": 11.07, "elapsed_time": "2:31:03", "remaining_time": "20:13:23"}
+{"current_steps": 244, "total_steps": 2195, "loss": 0.5077, "lr": 3.9985427459128496e-05, "epoch": 0.5551763367463026, "percentage": 11.12, "elapsed_time": "2:31:35", "remaining_time": "20:12:06"}
+{"current_steps": 245, "total_steps": 2195, "loss": 0.5583, "lr": 3.998418794454604e-05, "epoch": 0.5574516496018203, "percentage": 11.16, "elapsed_time": "2:32:09", "remaining_time": "20:11:00"}
+{"current_steps": 246, "total_steps": 2195, "loss": 0.5342, "lr": 3.998289786473681e-05, "epoch": 0.5597269624573379, "percentage": 11.21, "elapsed_time": "2:32:42", "remaining_time": "20:09:53"}
+{"current_steps": 247, "total_steps": 2195, "loss": 0.5082, "lr": 3.998155722296504e-05, "epoch": 0.5620022753128555, "percentage": 11.25, "elapsed_time": "2:33:20", "remaining_time": "20:09:23"}
+{"current_steps": 248, "total_steps": 2195, "loss": 0.511, "lr": 3.99801660226229e-05, "epoch": 0.5642775881683731, "percentage": 11.3, "elapsed_time": "2:33:58", "remaining_time": "20:08:52"}
+{"current_steps": 249, "total_steps": 2195, "loss": 0.5457, "lr": 3.9978724267230495e-05, "epoch": 0.5665529010238908, "percentage": 11.34, "elapsed_time": "2:34:38", "remaining_time": "20:08:30"}
+{"current_steps": 250, "total_steps": 2195, "loss": 0.5201, "lr": 3.997723196043585e-05, "epoch": 0.5688282138794084, "percentage": 11.39, "elapsed_time": "2:35:19", "remaining_time": "20:08:27"}
+{"current_steps": 251, "total_steps": 2195, "loss": 0.541, "lr": 3.997568910601489e-05, "epoch": 0.571103526734926, "percentage": 11.44, "elapsed_time": "2:35:56", "remaining_time": "20:07:47"}
+{"current_steps": 252, "total_steps": 2195, "loss": 0.5242, "lr": 3.997409570787144e-05, "epoch": 0.5733788395904437, "percentage": 11.48, "elapsed_time": "2:36:34", "remaining_time": "20:07:13"}
+{"current_steps": 253, "total_steps": 2195, "loss": 0.4896, "lr": 3.997245177003721e-05, "epoch": 0.5756541524459613, "percentage": 11.53, "elapsed_time": "2:37:06", "remaining_time": "20:05:55"}
+{"current_steps": 254, "total_steps": 2195, "loss": 0.4972, "lr": 3.9970757296671795e-05, "epoch": 0.5779294653014789, "percentage": 11.57, "elapsed_time": "2:37:43", "remaining_time": "20:05:16"}
+{"current_steps": 255, "total_steps": 2195, "loss": 0.4786, "lr": 3.9969012292062655e-05, "epoch": 0.5802047781569966, "percentage": 11.62, "elapsed_time": "2:38:17", "remaining_time": "20:04:19"}
+{"current_steps": 256, "total_steps": 2195, "loss": 0.5189, "lr": 3.9967216760625113e-05, "epoch": 0.5824800910125142, "percentage": 11.66, "elapsed_time": "2:38:54", "remaining_time": "20:03:36"}
+{"current_steps": 257, "total_steps": 2195, "loss": 0.4769, "lr": 3.9965370706902324e-05, "epoch": 0.5847554038680318, "percentage": 11.71, "elapsed_time": "2:39:29", "remaining_time": "20:02:43"}
+{"current_steps": 258, "total_steps": 2195, "loss": 0.5005, "lr": 3.99634741355653e-05, "epoch": 0.5870307167235495, "percentage": 11.75, "elapsed_time": "2:40:04", "remaining_time": "20:01:50"}
+{"current_steps": 259, "total_steps": 2195, "loss": 0.5257, "lr": 3.9961527051412854e-05, "epoch": 0.5893060295790671, "percentage": 11.8, "elapsed_time": "2:40:38", "remaining_time": "20:00:46"}
+{"current_steps": 260, "total_steps": 2195, "loss": 0.5049, "lr": 3.9959529459371624e-05, "epoch": 0.5915813424345847, "percentage": 11.85, "elapsed_time": "2:41:15", "remaining_time": "20:00:06"}
+{"current_steps": 261, "total_steps": 2195, "loss": 0.5102, "lr": 3.9957481364496044e-05, "epoch": 0.5938566552901023, "percentage": 11.89, "elapsed_time": "2:41:52", "remaining_time": "19:59:30"}
+{"current_steps": 262, "total_steps": 2195, "loss": 0.5099, "lr": 3.9955382771968316e-05, "epoch": 0.59613196814562, "percentage": 11.94, "elapsed_time": "2:42:26", "remaining_time": "19:58:28"}
+{"current_steps": 263, "total_steps": 2195, "loss": 0.5309, "lr": 3.9953233687098435e-05, "epoch": 0.5984072810011376, "percentage": 11.98, "elapsed_time": "2:43:02", "remaining_time": "19:57:40"}
+{"current_steps": 264, "total_steps": 2195, "loss": 0.568, "lr": 3.9951034115324156e-05, "epoch": 0.6006825938566553, "percentage": 12.03, "elapsed_time": "2:43:44", "remaining_time": "19:57:40"}
+{"current_steps": 265, "total_steps": 2195, "loss": 0.5319, "lr": 3.994878406221097e-05, "epoch": 0.602957906712173, "percentage": 12.07, "elapsed_time": "2:44:19", "remaining_time": "19:56:47"}
+{"current_steps": 266, "total_steps": 2195, "loss": 0.5227, "lr": 3.994648353345208e-05, "epoch": 0.6052332195676906, "percentage": 12.12, "elapsed_time": "2:44:54", "remaining_time": "19:55:55"}
+{"current_steps": 267, "total_steps": 2195, "loss": 0.4714, "lr": 3.994413253486846e-05, "epoch": 0.6075085324232082, "percentage": 12.16, "elapsed_time": "2:45:32", "remaining_time": "19:55:21"}
+{"current_steps": 268, "total_steps": 2195, "loss": 0.4962, "lr": 3.994173107240872e-05, "epoch": 0.6097838452787259, "percentage": 12.21, "elapsed_time": "2:46:04", "remaining_time": "19:54:05"}
+{"current_steps": 269, "total_steps": 2195, "loss": 0.5242, "lr": 3.9939279152149216e-05, "epoch": 0.6120591581342435, "percentage": 12.26, "elapsed_time": "2:46:48", "remaining_time": "19:54:20"}
+{"current_steps": 270, "total_steps": 2195, "loss": 0.5343, "lr": 3.993677678029392e-05, "epoch": 0.6143344709897611, "percentage": 12.3, "elapsed_time": "2:47:27", "remaining_time": "19:53:57"}
+{"current_steps": 271, "total_steps": 2195, "loss": 0.5203, "lr": 3.993422396317451e-05, "epoch": 0.6166097838452788, "percentage": 12.35, "elapsed_time": "2:48:04", "remaining_time": "19:53:18"}
+{"current_steps": 272, "total_steps": 2195, "loss": 0.5501, "lr": 3.993162070725027e-05, "epoch": 0.6188850967007964, "percentage": 12.39, "elapsed_time": "2:48:44", "remaining_time": "19:52:58"}
+{"current_steps": 273, "total_steps": 2195, "loss": 0.5163, "lr": 3.992896701910813e-05, "epoch": 0.621160409556314, "percentage": 12.44, "elapsed_time": "2:49:18", "remaining_time": "19:51:58"}
+{"current_steps": 274, "total_steps": 2195, "loss": 0.5206, "lr": 3.99262629054626e-05, "epoch": 0.6234357224118316, "percentage": 12.48, "elapsed_time": "2:49:50", "remaining_time": "19:50:43"}
+{"current_steps": 275, "total_steps": 2195, "loss": 0.5023, "lr": 3.992350837315581e-05, "epoch": 0.6257110352673493, "percentage": 12.53, "elapsed_time": "2:50:28", "remaining_time": "19:50:11"}
+{"current_steps": 276, "total_steps": 2195, "loss": 0.4886, "lr": 3.9920703429157436e-05, "epoch": 0.6279863481228669, "percentage": 12.57, "elapsed_time": "2:51:07", "remaining_time": "19:49:46"}
+{"current_steps": 277, "total_steps": 2195, "loss": 0.5148, "lr": 3.991784808056473e-05, "epoch": 0.6302616609783845, "percentage": 12.62, "elapsed_time": "2:51:49", "remaining_time": "19:49:46"}
+{"current_steps": 278, "total_steps": 2195, "loss": 0.5318, "lr": 3.9914942334602464e-05, "epoch": 0.6325369738339022, "percentage": 12.67, "elapsed_time": "2:52:22", "remaining_time": "19:48:35"}
+{"current_steps": 279, "total_steps": 2195, "loss": 0.5194, "lr": 3.991198619862294e-05, "epoch": 0.6348122866894198, "percentage": 12.71, "elapsed_time": "2:52:57", "remaining_time": "19:47:48"}
+{"current_steps": 280, "total_steps": 2195, "loss": 0.5411, "lr": 3.990897968010596e-05, "epoch": 0.6370875995449374, "percentage": 12.76, "elapsed_time": "2:53:38", "remaining_time": "19:47:33"}
+{"current_steps": 281, "total_steps": 2195, "loss": 0.5076, "lr": 3.99059227866588e-05, "epoch": 0.6393629124004551, "percentage": 12.8, "elapsed_time": "2:54:14", "remaining_time": "19:46:50"}
+{"current_steps": 282, "total_steps": 2195, "loss": 0.5031, "lr": 3.9902815526016196e-05, "epoch": 0.6416382252559727, "percentage": 12.85, "elapsed_time": "2:54:46", "remaining_time": "19:45:36"}
+{"current_steps": 283, "total_steps": 2195, "loss": 0.4603, "lr": 3.989965790604033e-05, "epoch": 0.6439135381114903, "percentage": 12.89, "elapsed_time": "2:55:20", "remaining_time": "19:44:38"}
+{"current_steps": 284, "total_steps": 2195, "loss": 0.5107, "lr": 3.9896449934720814e-05, "epoch": 0.646188850967008, "percentage": 12.94, "elapsed_time": "2:55:57", "remaining_time": "19:44:02"}
+{"current_steps": 285, "total_steps": 2195, "loss": 0.5502, "lr": 3.989319162017465e-05, "epoch": 0.6484641638225256, "percentage": 12.98, "elapsed_time": "2:56:41", "remaining_time": "19:44:05"}
+{"current_steps": 286, "total_steps": 2195, "loss": 0.5083, "lr": 3.988988297064623e-05, "epoch": 0.6507394766780432, "percentage": 13.03, "elapsed_time": "2:57:16", "remaining_time": "19:43:15"}
+{"current_steps": 287, "total_steps": 2195, "loss": 0.5028, "lr": 3.98865239945073e-05, "epoch": 0.6530147895335608, "percentage": 13.08, "elapsed_time": "2:57:56", "remaining_time": "19:43:01"}
+{"current_steps": 288, "total_steps": 2195, "loss": 0.5036, "lr": 3.988311470025695e-05, "epoch": 0.6552901023890785, "percentage": 13.12, "elapsed_time": "2:58:37", "remaining_time": "19:42:45"}
+{"current_steps": 289, "total_steps": 2195, "loss": 0.4908, "lr": 3.987965509652159e-05, "epoch": 0.6575654152445961, "percentage": 13.17, "elapsed_time": "2:59:18", "remaining_time": "19:42:33"}
+{"current_steps": 290, "total_steps": 2195, "loss": 0.478, "lr": 3.987614519205493e-05, "epoch": 0.6598407281001137, "percentage": 13.21, "elapsed_time": "2:59:58", "remaining_time": "19:42:14"}
+{"current_steps": 291, "total_steps": 2195, "loss": 0.5338, "lr": 3.987258499573792e-05, "epoch": 0.6621160409556314, "percentage": 13.26, "elapsed_time": "3:00:33", "remaining_time": "19:41:25"}
+{"current_steps": 292, "total_steps": 2195, "loss": 0.5124, "lr": 3.986897451657882e-05, "epoch": 0.664391353811149, "percentage": 13.3, "elapsed_time": "3:01:04", "remaining_time": "19:40:03"}
+{"current_steps": 293, "total_steps": 2195, "loss": 0.4858, "lr": 3.986531376371307e-05, "epoch": 0.6666666666666666, "percentage": 13.35, "elapsed_time": "3:01:43", "remaining_time": "19:39:41"}
+{"current_steps": 294, "total_steps": 2195, "loss": 0.5302, "lr": 3.9861602746403336e-05, "epoch": 0.6689419795221843, "percentage": 13.39, "elapsed_time": "3:02:17", "remaining_time": "19:38:44"}
+{"current_steps": 295, "total_steps": 2195, "loss": 0.4594, "lr": 3.985784147403947e-05, "epoch": 0.6712172923777019, "percentage": 13.44, "elapsed_time": "3:02:53", "remaining_time": "19:37:56"}
+{"current_steps": 296, "total_steps": 2195, "loss": 0.4986, "lr": 3.985402995613847e-05, "epoch": 0.6734926052332195, "percentage": 13.49, "elapsed_time": "3:03:27", "remaining_time": "19:37:00"}
+{"current_steps": 297, "total_steps": 2195, "loss": 0.4866, "lr": 3.985016820234447e-05, "epoch": 0.6757679180887372, "percentage": 13.53, "elapsed_time": "3:04:09", "remaining_time": "19:36:50"}
+{"current_steps": 298, "total_steps": 2195, "loss": 0.4996, "lr": 3.984625622242872e-05, "epoch": 0.6780432309442548, "percentage": 13.58, "elapsed_time": "3:04:44", "remaining_time": "19:35:59"}
+{"current_steps": 299, "total_steps": 2195, "loss": 0.4805, "lr": 3.9842294026289565e-05, "epoch": 0.6803185437997725, "percentage": 13.62, "elapsed_time": "3:05:22", "remaining_time": "19:35:29"}
+{"current_steps": 300, "total_steps": 2195, "loss": 0.4821, "lr": 3.983828162395238e-05, "epoch": 0.6825938566552902, "percentage": 13.67, "elapsed_time": "3:06:01", "remaining_time": "19:35:05"}
+{"current_steps": 301, "total_steps": 2195, "loss": 0.527, "lr": 3.98342190255696e-05, "epoch": 0.6848691695108078, "percentage": 13.71, "elapsed_time": "3:06:45", "remaining_time": "19:35:11"}
+{"current_steps": 302, "total_steps": 2195, "loss": 0.5062, "lr": 3.9830106241420666e-05, "epoch": 0.6871444823663254, "percentage": 13.76, "elapsed_time": "3:07:26", "remaining_time": "19:34:58"}
+{"current_steps": 303, "total_steps": 2195, "loss": 0.5062, "lr": 3.9825943281912005e-05, "epoch": 0.689419795221843, "percentage": 13.8, "elapsed_time": "3:07:57", "remaining_time": "19:33:39"}
+{"current_steps": 304, "total_steps": 2195, "loss": 0.4887, "lr": 3.9821730157576975e-05, "epoch": 0.6916951080773607, "percentage": 13.85, "elapsed_time": "3:08:31", "remaining_time": "19:32:39"}
+{"current_steps": 305, "total_steps": 2195, "loss": 0.5404, "lr": 3.98174668790759e-05, "epoch": 0.6939704209328783, "percentage": 13.9, "elapsed_time": "3:09:13", "remaining_time": "19:32:36"}
+{"current_steps": 306, "total_steps": 2195, "loss": 0.4759, "lr": 3.9813153457195986e-05, "epoch": 0.6962457337883959, "percentage": 13.94, "elapsed_time": "3:09:46", "remaining_time": "19:31:32"}
+{"current_steps": 307, "total_steps": 2195, "loss": 0.5379, "lr": 3.980878990285132e-05, "epoch": 0.6985210466439136, "percentage": 13.99, "elapsed_time": "3:10:24", "remaining_time": "19:30:56"}
+{"current_steps": 308, "total_steps": 2195, "loss": 0.5152, "lr": 3.9804376227082834e-05, "epoch": 0.7007963594994312, "percentage": 14.03, "elapsed_time": "3:10:56", "remaining_time": "19:29:52"}
+{"current_steps": 309, "total_steps": 2195, "loss": 0.5232, "lr": 3.9799912441058286e-05, "epoch": 0.7030716723549488, "percentage": 14.08, "elapsed_time": "3:11:31", "remaining_time": "19:28:56"}
+{"current_steps": 310, "total_steps": 2195, "loss": 0.5045, "lr": 3.979539855607222e-05, "epoch": 0.7053469852104665, "percentage": 14.12, "elapsed_time": "3:12:00", "remaining_time": "19:27:30"}
+{"current_steps": 311, "total_steps": 2195, "loss": 0.4881, "lr": 3.9790834583545946e-05, "epoch": 0.7076222980659841, "percentage": 14.17, "elapsed_time": "3:12:35", "remaining_time": "19:26:43"}
+{"current_steps": 312, "total_steps": 2195, "loss": 0.4773, "lr": 3.978622053502751e-05, "epoch": 0.7098976109215017, "percentage": 14.21, "elapsed_time": "3:13:16", "remaining_time": "19:26:30"}
+{"current_steps": 313, "total_steps": 2195, "loss": 0.5217, "lr": 3.978155642219167e-05, "epoch": 0.7121729237770194, "percentage": 14.26, "elapsed_time": "3:13:56", "remaining_time": "19:26:08"}
+{"current_steps": 314, "total_steps": 2195, "loss": 0.5206, "lr": 3.977684225683984e-05, "epoch": 0.714448236632537, "percentage": 14.31, "elapsed_time": "3:14:35", "remaining_time": "19:25:40"}
+{"current_steps": 315, "total_steps": 2195, "loss": 0.5388, "lr": 3.9772078050900105e-05, "epoch": 0.7167235494880546, "percentage": 14.35, "elapsed_time": "3:15:15", "remaining_time": "19:25:21"}
+{"current_steps": 316, "total_steps": 2195, "loss": 0.4748, "lr": 3.9767263816427146e-05, "epoch": 0.7189988623435722, "percentage": 14.4, "elapsed_time": "3:15:53", "remaining_time": "19:24:51"}
+{"current_steps": 317, "total_steps": 2195, "loss": 0.4889, "lr": 3.9762399565602233e-05, "epoch": 0.7212741751990899, "percentage": 14.44, "elapsed_time": "3:16:37", "remaining_time": "19:24:52"}
+{"current_steps": 318, "total_steps": 2195, "loss": 0.5083, "lr": 3.97574853107332e-05, "epoch": 0.7235494880546075, "percentage": 14.49, "elapsed_time": "3:17:13", "remaining_time": "19:24:07"}
+{"current_steps": 319, "total_steps": 2195, "loss": 0.4951, "lr": 3.97525210642544e-05, "epoch": 0.7258248009101251, "percentage": 14.53, "elapsed_time": "3:17:49", "remaining_time": "19:23:24"}
+{"current_steps": 320, "total_steps": 2195, "loss": 0.4905, "lr": 3.974750683872667e-05, "epoch": 0.7281001137656428, "percentage": 14.58, "elapsed_time": "3:18:22", "remaining_time": "19:22:20"}
+{"current_steps": 321, "total_steps": 2195, "loss": 0.5322, "lr": 3.9742442646837316e-05, "epoch": 0.7303754266211604, "percentage": 14.62, "elapsed_time": "3:19:01", "remaining_time": "19:21:56"}
+{"current_steps": 322, "total_steps": 2195, "loss": 0.52, "lr": 3.973732850140007e-05, "epoch": 0.732650739476678, "percentage": 14.67, "elapsed_time": "3:19:41", "remaining_time": "19:21:32"}
+{"current_steps": 323, "total_steps": 2195, "loss": 0.532, "lr": 3.973216441535506e-05, "epoch": 0.7349260523321957, "percentage": 14.72, "elapsed_time": "3:20:21", "remaining_time": "19:21:11"}
+{"current_steps": 324, "total_steps": 2195, "loss": 0.5008, "lr": 3.972695040176877e-05, "epoch": 0.7372013651877133, "percentage": 14.76, "elapsed_time": "3:20:57", "remaining_time": "19:20:28"}
+{"current_steps": 325, "total_steps": 2195, "loss": 0.5578, "lr": 3.972168647383402e-05, "epoch": 0.7394766780432309, "percentage": 14.81, "elapsed_time": "3:21:41", "remaining_time": "19:20:28"}
+{"current_steps": 326, "total_steps": 2195, "loss": 0.5155, "lr": 3.971637264486993e-05, "epoch": 0.7417519908987485, "percentage": 14.85, "elapsed_time": "3:22:20", "remaining_time": "19:20:00"}
+{"current_steps": 327, "total_steps": 2195, "loss": 0.4719, "lr": 3.971100892832188e-05, "epoch": 0.7440273037542662, "percentage": 14.9, "elapsed_time": "3:22:59", "remaining_time": "19:19:38"}
+{"current_steps": 328, "total_steps": 2195, "loss": 0.4922, "lr": 3.970559533776147e-05, "epoch": 0.7463026166097838, "percentage": 14.94, "elapsed_time": "3:23:27", "remaining_time": "19:18:05"}
+{"current_steps": 329, "total_steps": 2195, "loss": 0.5046, "lr": 3.9700131886886506e-05, "epoch": 0.7485779294653014, "percentage": 14.99, "elapsed_time": "3:24:08", "remaining_time": "19:17:50"}
+{"current_steps": 330, "total_steps": 2195, "loss": 0.5153, "lr": 3.9694618589520945e-05, "epoch": 0.7508532423208191, "percentage": 15.03, "elapsed_time": "3:24:49", "remaining_time": "19:17:31"}
+{"current_steps": 331, "total_steps": 2195, "loss": 0.4978, "lr": 3.968905545961487e-05, "epoch": 0.7531285551763367, "percentage": 15.08, "elapsed_time": "3:25:28", "remaining_time": "19:17:09"}
+{"current_steps": 332, "total_steps": 2195, "loss": 0.5221, "lr": 3.968344251124447e-05, "epoch": 0.7554038680318543, "percentage": 15.13, "elapsed_time": "3:26:10", "remaining_time": "19:16:55"}
+{"current_steps": 333, "total_steps": 2195, "loss": 0.495, "lr": 3.967777975861196e-05, "epoch": 0.757679180887372, "percentage": 15.17, "elapsed_time": "3:26:51", "remaining_time": "19:16:38"}
+{"current_steps": 334, "total_steps": 2195, "loss": 0.538, "lr": 3.96720672160456e-05, "epoch": 0.7599544937428896, "percentage": 15.22, "elapsed_time": "3:27:28", "remaining_time": "19:15:58"}
+{"current_steps": 335, "total_steps": 2195, "loss": 0.5085, "lr": 3.966630489799959e-05, "epoch": 0.7622298065984073, "percentage": 15.26, "elapsed_time": "3:28:04", "remaining_time": "19:15:17"}
+{"current_steps": 336, "total_steps": 2195, "loss": 0.4873, "lr": 3.966049281905414e-05, "epoch": 0.764505119453925, "percentage": 15.31, "elapsed_time": "3:28:39", "remaining_time": "19:14:28"}
+{"current_steps": 337, "total_steps": 2195, "loss": 0.4971, "lr": 3.9654630993915305e-05, "epoch": 0.7667804323094426, "percentage": 15.35, "elapsed_time": "3:29:16", "remaining_time": "19:13:48"}
+{"current_steps": 338, "total_steps": 2195, "loss": 0.5122, "lr": 3.964871943741504e-05, "epoch": 0.7690557451649602, "percentage": 15.4, "elapsed_time": "3:29:39", "remaining_time": "19:11:53"}
+{"current_steps": 339, "total_steps": 2195, "loss": 0.5161, "lr": 3.964275816451115e-05, "epoch": 0.7713310580204779, "percentage": 15.44, "elapsed_time": "3:30:20", "remaining_time": "19:11:35"}
+{"current_steps": 340, "total_steps": 2195, "loss": 0.5104, "lr": 3.96367471902872e-05, "epoch": 0.7736063708759955, "percentage": 15.49, "elapsed_time": "3:31:01", "remaining_time": "19:11:20"}
+{"current_steps": 341, "total_steps": 2195, "loss": 0.5148, "lr": 3.963068652995252e-05, "epoch": 0.7758816837315131, "percentage": 15.54, "elapsed_time": "3:31:39", "remaining_time": "19:10:48"}
+{"current_steps": 342, "total_steps": 2195, "loss": 0.471, "lr": 3.962457619884218e-05, "epoch": 0.7781569965870307, "percentage": 15.58, "elapsed_time": "3:32:16", "remaining_time": "19:10:09"}
+{"current_steps": 343, "total_steps": 2195, "loss": 0.5463, "lr": 3.961841621241692e-05, "epoch": 0.7804323094425484, "percentage": 15.63, "elapsed_time": "3:32:57", "remaining_time": "19:09:51"}
+{"current_steps": 344, "total_steps": 2195, "loss": 0.5066, "lr": 3.9612206586263104e-05, "epoch": 0.782707622298066, "percentage": 15.67, "elapsed_time": "3:33:33", "remaining_time": "19:09:05"}
+{"current_steps": 345, "total_steps": 2195, "loss": 0.4976, "lr": 3.960594733609273e-05, "epoch": 0.7849829351535836, "percentage": 15.72, "elapsed_time": "3:34:14", "remaining_time": "19:08:51"}
+{"current_steps": 346, "total_steps": 2195, "loss": 0.4864, "lr": 3.959963847774332e-05, "epoch": 0.7872582480091013, "percentage": 15.76, "elapsed_time": "3:34:47", "remaining_time": "19:07:49"}
+{"current_steps": 347, "total_steps": 2195, "loss": 0.4753, "lr": 3.959328002717795e-05, "epoch": 0.7895335608646189, "percentage": 15.81, "elapsed_time": "3:35:24", "remaining_time": "19:07:11"}
+{"current_steps": 348, "total_steps": 2195, "loss": 0.4916, "lr": 3.958687200048516e-05, "epoch": 0.7918088737201365, "percentage": 15.85, "elapsed_time": "3:36:02", "remaining_time": "19:06:37"}
+{"current_steps": 349, "total_steps": 2195, "loss": 0.5161, "lr": 3.958041441387894e-05, "epoch": 0.7940841865756542, "percentage": 15.9, "elapsed_time": "3:36:36", "remaining_time": "19:05:42"}
+{"current_steps": 350, "total_steps": 2195, "loss": 0.4895, "lr": 3.957390728369867e-05, "epoch": 0.7963594994311718, "percentage": 15.95, "elapsed_time": "3:37:16", "remaining_time": "19:05:18"}
+{"current_steps": 351, "total_steps": 2195, "loss": 0.4915, "lr": 3.9567350626409094e-05, "epoch": 0.7986348122866894, "percentage": 15.99, "elapsed_time": "3:37:48", "remaining_time": "19:04:15"}
+{"current_steps": 352, "total_steps": 2195, "loss": 0.487, "lr": 3.956074445860027e-05, "epoch": 0.800910125142207, "percentage": 16.04, "elapsed_time": "3:38:24", "remaining_time": "19:03:31"}
+{"current_steps": 353, "total_steps": 2195, "loss": 0.507, "lr": 3.955408879698753e-05, "epoch": 0.8031854379977247, "percentage": 16.08, "elapsed_time": "3:38:51", "remaining_time": "19:02:04"}
+{"current_steps": 354, "total_steps": 2195, "loss": 0.5003, "lr": 3.954738365841144e-05, "epoch": 0.8054607508532423, "percentage": 16.13, "elapsed_time": "3:39:27", "remaining_time": "19:01:16"}
+{"current_steps": 355, "total_steps": 2195, "loss": 0.4729, "lr": 3.9540629059837767e-05, "epoch": 0.8077360637087599, "percentage": 16.17, "elapsed_time": "3:40:08", "remaining_time": "19:01:00"}
+{"current_steps": 356, "total_steps": 2195, "loss": 0.5229, "lr": 3.95338250183574e-05, "epoch": 0.8100113765642776, "percentage": 16.22, "elapsed_time": "3:40:47", "remaining_time": "19:00:34"}
+{"current_steps": 357, "total_steps": 2195, "loss": 0.5088, "lr": 3.952697155118635e-05, "epoch": 0.8122866894197952, "percentage": 16.26, "elapsed_time": "3:41:22", "remaining_time": "18:59:45"}
+{"current_steps": 358, "total_steps": 2195, "loss": 0.4869, "lr": 3.952006867566569e-05, "epoch": 0.8145620022753128, "percentage": 16.31, "elapsed_time": "3:42:02", "remaining_time": "18:59:23"}
+{"current_steps": 359, "total_steps": 2195, "loss": 0.5289, "lr": 3.951311640926148e-05, "epoch": 0.8168373151308305, "percentage": 16.36, "elapsed_time": "3:42:38", "remaining_time": "18:58:38"}
+{"current_steps": 360, "total_steps": 2195, "loss": 0.4777, "lr": 3.95061147695648e-05, "epoch": 0.8191126279863481, "percentage": 16.4, "elapsed_time": "3:43:08", "remaining_time": "18:57:23"}
+{"current_steps": 361, "total_steps": 2195, "loss": 0.5008, "lr": 3.949906377429162e-05, "epoch": 0.8213879408418657, "percentage": 16.45, "elapsed_time": "3:43:46", "remaining_time": "18:56:48"}
+{"current_steps": 362, "total_steps": 2195, "loss": 0.5241, "lr": 3.94919634412828e-05, "epoch": 0.8236632536973834, "percentage": 16.49, "elapsed_time": "3:44:17", "remaining_time": "18:55:43"}
+{"current_steps": 363, "total_steps": 2195, "loss": 0.4912, "lr": 3.948481378850405e-05, "epoch": 0.825938566552901, "percentage": 16.54, "elapsed_time": "3:44:56", "remaining_time": "18:55:16"}
+{"current_steps": 364, "total_steps": 2195, "loss": 0.486, "lr": 3.947761483404585e-05, "epoch": 0.8282138794084186, "percentage": 16.58, "elapsed_time": "3:45:30", "remaining_time": "18:54:22"}
+{"current_steps": 365, "total_steps": 2195, "loss": 0.4874, "lr": 3.947036659612345e-05, "epoch": 0.8304891922639362, "percentage": 16.63, "elapsed_time": "3:46:08", "remaining_time": "18:53:46"}
+{"current_steps": 366, "total_steps": 2195, "loss": 0.5323, "lr": 3.94630690930768e-05, "epoch": 0.8327645051194539, "percentage": 16.67, "elapsed_time": "3:46:52", "remaining_time": "18:53:43"}
+{"current_steps": 367, "total_steps": 2195, "loss": 0.4979, "lr": 3.945572234337046e-05, "epoch": 0.8350398179749715, "percentage": 16.72, "elapsed_time": "3:47:22", "remaining_time": "18:52:33"}
+{"current_steps": 368, "total_steps": 2195, "loss": 0.551, "lr": 3.944832636559366e-05, "epoch": 0.8373151308304891, "percentage": 16.77, "elapsed_time": "3:48:01", "remaining_time": "18:52:05"}
+{"current_steps": 369, "total_steps": 2195, "loss": 0.5008, "lr": 3.944088117846015e-05, "epoch": 0.8395904436860068, "percentage": 16.81, "elapsed_time": "3:48:35", "remaining_time": "18:51:10"}
+{"current_steps": 370, "total_steps": 2195, "loss": 0.5148, "lr": 3.94333868008082e-05, "epoch": 0.8418657565415245, "percentage": 16.86, "elapsed_time": "3:49:13", "remaining_time": "18:50:40"}
+{"current_steps": 371, "total_steps": 2195, "loss": 0.491, "lr": 3.9425843251600555e-05, "epoch": 0.8441410693970421, "percentage": 16.9, "elapsed_time": "3:49:55", "remaining_time": "18:50:26"}
+{"current_steps": 372, "total_steps": 2195, "loss": 0.5177, "lr": 3.9418250549924355e-05, "epoch": 0.8464163822525598, "percentage": 16.95, "elapsed_time": "3:50:30", "remaining_time": "18:49:38"}
+{"current_steps": 373, "total_steps": 2195, "loss": 0.5107, "lr": 3.9410608714991136e-05, "epoch": 0.8486916951080774, "percentage": 16.99, "elapsed_time": "3:51:03", "remaining_time": "18:48:40"}
+{"current_steps": 374, "total_steps": 2195, "loss": 0.5397, "lr": 3.9402917766136735e-05, "epoch": 0.850967007963595, "percentage": 17.04, "elapsed_time": "3:51:33", "remaining_time": "18:47:27"}
+{"current_steps": 375, "total_steps": 2195, "loss": 0.4829, "lr": 3.939517772282127e-05, "epoch": 0.8532423208191127, "percentage": 17.08, "elapsed_time": "3:52:13", "remaining_time": "18:47:02"}
+{"current_steps": 376, "total_steps": 2195, "loss": 0.5003, "lr": 3.938738860462907e-05, "epoch": 0.8555176336746303, "percentage": 17.13, "elapsed_time": "3:52:45", "remaining_time": "18:45:59"}
+{"current_steps": 377, "total_steps": 2195, "loss": 0.5284, "lr": 3.937955043126864e-05, "epoch": 0.8577929465301479, "percentage": 17.18, "elapsed_time": "3:53:21", "remaining_time": "18:45:19"}
+{"current_steps": 378, "total_steps": 2195, "loss": 0.5121, "lr": 3.9371663222572625e-05, "epoch": 0.8600682593856656, "percentage": 17.22, "elapsed_time": "3:53:59", "remaining_time": "18:44:46"}
+{"current_steps": 379, "total_steps": 2195, "loss": 0.4996, "lr": 3.936372699849772e-05, "epoch": 0.8623435722411832, "percentage": 17.27, "elapsed_time": "3:54:42", "remaining_time": "18:44:35"}
+{"current_steps": 380, "total_steps": 2195, "loss": 0.5245, "lr": 3.935574177912465e-05, "epoch": 0.8646188850967008, "percentage": 17.31, "elapsed_time": "3:55:12", "remaining_time": "18:43:27"}
+{"current_steps": 381, "total_steps": 2195, "loss": 0.5342, "lr": 3.93477075846581e-05, "epoch": 0.8668941979522184, "percentage": 17.36, "elapsed_time": "3:55:50", "remaining_time": "18:42:54"}
+{"current_steps": 382, "total_steps": 2195, "loss": 0.4866, "lr": 3.93396244354267e-05, "epoch": 0.8691695108077361, "percentage": 17.4, "elapsed_time": "3:56:23", "remaining_time": "18:41:54"}
+{"current_steps": 383, "total_steps": 2195, "loss": 0.5043, "lr": 3.933149235188294e-05, "epoch": 0.8714448236632537, "percentage": 17.45, "elapsed_time": "3:57:03", "remaining_time": "18:41:31"}
+{"current_steps": 384, "total_steps": 2195, "loss": 0.4805, "lr": 3.9323311354603086e-05, "epoch": 0.8737201365187713, "percentage": 17.49, "elapsed_time": "3:57:42", "remaining_time": "18:41:02"}
+{"current_steps": 385, "total_steps": 2195, "loss": 0.5067, "lr": 3.931508146428724e-05, "epoch": 0.875995449374289, "percentage": 17.54, "elapsed_time": "3:58:23", "remaining_time": "18:40:44"}
+{"current_steps": 386, "total_steps": 2195, "loss": 0.5271, "lr": 3.930680270175915e-05, "epoch": 0.8782707622298066, "percentage": 17.59, "elapsed_time": "3:59:00", "remaining_time": "18:40:08"}
+{"current_steps": 387, "total_steps": 2195, "loss": 0.5124, "lr": 3.929847508796628e-05, "epoch": 0.8805460750853242, "percentage": 17.63, "elapsed_time": "3:59:39", "remaining_time": "18:39:40"}
+{"current_steps": 388, "total_steps": 2195, "loss": 0.5205, "lr": 3.9290098643979654e-05, "epoch": 0.8828213879408419, "percentage": 17.68, "elapsed_time": "4:00:21", "remaining_time": "18:39:22"}
+{"current_steps": 389, "total_steps": 2195, "loss": 0.503, "lr": 3.928167339099387e-05, "epoch": 0.8850967007963595, "percentage": 17.72, "elapsed_time": "4:00:58", "remaining_time": "18:38:46"}
+{"current_steps": 390, "total_steps": 2195, "loss": 0.4765, "lr": 3.927319935032703e-05, "epoch": 0.8873720136518771, "percentage": 17.77, "elapsed_time": "4:01:41", "remaining_time": "18:38:34"}
+{"current_steps": 391, "total_steps": 2195, "loss": 0.485, "lr": 3.926467654342067e-05, "epoch": 0.8896473265073948, "percentage": 17.81, "elapsed_time": "4:02:26", "remaining_time": "18:38:36"}
+{"current_steps": 392, "total_steps": 2195, "loss": 0.4695, "lr": 3.9256104991839724e-05, "epoch": 0.8919226393629124, "percentage": 17.86, "elapsed_time": "4:03:02", "remaining_time": "18:37:52"}
+{"current_steps": 393, "total_steps": 2195, "loss": 0.4862, "lr": 3.924748471727246e-05, "epoch": 0.89419795221843, "percentage": 17.9, "elapsed_time": "4:03:41", "remaining_time": "18:37:25"}
+{"current_steps": 394, "total_steps": 2195, "loss": 0.4937, "lr": 3.923881574153043e-05, "epoch": 0.8964732650739476, "percentage": 17.95, "elapsed_time": "4:04:16", "remaining_time": "18:36:34"}
+{"current_steps": 395, "total_steps": 2195, "loss": 0.4581, "lr": 3.9230098086548414e-05, "epoch": 0.8987485779294653, "percentage": 18.0, "elapsed_time": "4:04:53", "remaining_time": "18:35:58"}
+{"current_steps": 396, "total_steps": 2195, "loss": 0.4966, "lr": 3.9221331774384356e-05, "epoch": 0.9010238907849829, "percentage": 18.04, "elapsed_time": "4:05:29", "remaining_time": "18:35:15"}
+{"current_steps": 397, "total_steps": 2195, "loss": 0.4709, "lr": 3.9212516827219323e-05, "epoch": 0.9032992036405005, "percentage": 18.09, "elapsed_time": "4:06:01", "remaining_time": "18:34:12"}
+{"current_steps": 398, "total_steps": 2195, "loss": 0.4943, "lr": 3.920365326735745e-05, "epoch": 0.9055745164960182, "percentage": 18.13, "elapsed_time": "4:06:37", "remaining_time": "18:33:31"}
+{"current_steps": 399, "total_steps": 2195, "loss": 0.4643, "lr": 3.919474111722585e-05, "epoch": 0.9078498293515358, "percentage": 18.18, "elapsed_time": "4:07:13", "remaining_time": "18:32:50"}
+{"current_steps": 400, "total_steps": 2195, "loss": 0.4839, "lr": 3.918578039937459e-05, "epoch": 0.9101251422070534, "percentage": 18.22, "elapsed_time": "4:07:52", "remaining_time": "18:32:19"}
+{"current_steps": 401, "total_steps": 2195, "loss": 0.4816, "lr": 3.917677113647665e-05, "epoch": 0.9124004550625711, "percentage": 18.27, "elapsed_time": "4:08:31", "remaining_time": "18:31:53"}
+{"current_steps": 402, "total_steps": 2195, "loss": 0.5565, "lr": 3.916771335132781e-05, "epoch": 0.9146757679180887, "percentage": 18.31, "elapsed_time": "4:09:11", "remaining_time": "18:31:28"}
+{"current_steps": 403, "total_steps": 2195, "loss": 0.5043, "lr": 3.915860706684664e-05, "epoch": 0.9169510807736063, "percentage": 18.36, "elapsed_time": "4:09:46", "remaining_time": "18:30:41"}
+{"current_steps": 404, "total_steps": 2195, "loss": 0.4969, "lr": 3.914945230607443e-05, "epoch": 0.919226393629124, "percentage": 18.41, "elapsed_time": "4:10:21", "remaining_time": "18:29:54"}
+{"current_steps": 405, "total_steps": 2195, "loss": 0.5155, "lr": 3.914024909217511e-05, "epoch": 0.9215017064846417, "percentage": 18.45, "elapsed_time": "4:10:53", "remaining_time": "18:28:53"}
+{"current_steps": 406, "total_steps": 2195, "loss": 0.4922, "lr": 3.9130997448435235e-05, "epoch": 0.9237770193401593, "percentage": 18.5, "elapsed_time": "4:11:32", "remaining_time": "18:28:21"}
+{"current_steps": 407, "total_steps": 2195, "loss": 0.4862, "lr": 3.9121697398263874e-05, "epoch": 0.926052332195677, "percentage": 18.54, "elapsed_time": "4:12:07", "remaining_time": "18:27:36"}
+{"current_steps": 408, "total_steps": 2195, "loss": 0.5229, "lr": 3.911234896519259e-05, "epoch": 0.9283276450511946, "percentage": 18.59, "elapsed_time": "4:12:43", "remaining_time": "18:26:56"}
+{"current_steps": 409, "total_steps": 2195, "loss": 0.5235, "lr": 3.910295217287537e-05, "epoch": 0.9306029579067122, "percentage": 18.63, "elapsed_time": "4:13:24", "remaining_time": "18:26:32"}
+{"current_steps": 410, "total_steps": 2195, "loss": 0.4991, "lr": 3.909350704508856e-05, "epoch": 0.9328782707622298, "percentage": 18.68, "elapsed_time": "4:14:00", "remaining_time": "18:25:52"}
+{"current_steps": 411, "total_steps": 2195, "loss": 0.5074, "lr": 3.90840136057308e-05, "epoch": 0.9351535836177475, "percentage": 18.72, "elapsed_time": "4:14:33", "remaining_time": "18:24:56"}
+{"current_steps": 412, "total_steps": 2195, "loss": 0.4932, "lr": 3.9074471878822975e-05, "epoch": 0.9374288964732651, "percentage": 18.77, "elapsed_time": "4:15:14", "remaining_time": "18:24:35"}
+{"current_steps": 413, "total_steps": 2195, "loss": 0.5003, "lr": 3.906488188850816e-05, "epoch": 0.9397042093287827, "percentage": 18.82, "elapsed_time": "4:15:51", "remaining_time": "18:23:57"}
+{"current_steps": 414, "total_steps": 2195, "loss": 0.4752, "lr": 3.905524365905153e-05, "epoch": 0.9419795221843004, "percentage": 18.86, "elapsed_time": "4:16:25", "remaining_time": "18:23:09"}
+{"current_steps": 415, "total_steps": 2195, "loss": 0.4711, "lr": 3.904555721484034e-05, "epoch": 0.944254835039818, "percentage": 18.91, "elapsed_time": "4:16:59", "remaining_time": "18:22:14"}
+{"current_steps": 416, "total_steps": 2195, "loss": 0.4742, "lr": 3.903582258038382e-05, "epoch": 0.9465301478953356, "percentage": 18.95, "elapsed_time": "4:17:32", "remaining_time": "18:21:22"}
+{"current_steps": 417, "total_steps": 2195, "loss": 0.506, "lr": 3.902603978031315e-05, "epoch": 0.9488054607508533, "percentage": 19.0, "elapsed_time": "4:18:06", "remaining_time": "18:20:30"}
+{"current_steps": 418, "total_steps": 2195, "loss": 0.4762, "lr": 3.901620883938137e-05, "epoch": 0.9510807736063709, "percentage": 19.04, "elapsed_time": "4:18:41", "remaining_time": "18:19:42"}
+{"current_steps": 419, "total_steps": 2195, "loss": 0.4787, "lr": 3.900632978246334e-05, "epoch": 0.9533560864618885, "percentage": 19.09, "elapsed_time": "4:19:17", "remaining_time": "18:19:04"}
+{"current_steps": 420, "total_steps": 2195, "loss": 0.4921, "lr": 3.899640263455566e-05, "epoch": 0.9556313993174061, "percentage": 19.13, "elapsed_time": "4:19:52", "remaining_time": "18:18:17"}
+{"current_steps": 421, "total_steps": 2195, "loss": 0.5105, "lr": 3.8986427420776604e-05, "epoch": 0.9579067121729238, "percentage": 19.18, "elapsed_time": "4:20:34", "remaining_time": "18:17:58"}
+{"current_steps": 422, "total_steps": 2195, "loss": 0.4889, "lr": 3.897640416636608e-05, "epoch": 0.9601820250284414, "percentage": 19.23, "elapsed_time": "4:21:12", "remaining_time": "18:17:28"}
+{"current_steps": 423, "total_steps": 2195, "loss": 0.4662, "lr": 3.8966332896685545e-05, "epoch": 0.962457337883959, "percentage": 19.27, "elapsed_time": "4:21:42", "remaining_time": "18:16:21"}
+{"current_steps": 424, "total_steps": 2195, "loss": 0.5228, "lr": 3.895621363721795e-05, "epoch": 0.9647326507394767, "percentage": 19.32, "elapsed_time": "4:22:16", "remaining_time": "18:15:31"}
+{"current_steps": 425, "total_steps": 2195, "loss": 0.5191, "lr": 3.894604641356767e-05, "epoch": 0.9670079635949943, "percentage": 19.36, "elapsed_time": "4:22:50", "remaining_time": "18:14:40"}
+{"current_steps": 426, "total_steps": 2195, "loss": 0.4918, "lr": 3.893583125146043e-05, "epoch": 0.9692832764505119, "percentage": 19.41, "elapsed_time": "4:23:23", "remaining_time": "18:13:47"}
+{"current_steps": 427, "total_steps": 2195, "loss": 0.5068, "lr": 3.892556817674328e-05, "epoch": 0.9715585893060296, "percentage": 19.45, "elapsed_time": "4:24:00", "remaining_time": "18:13:08"}
+{"current_steps": 428, "total_steps": 2195, "loss": 0.4915, "lr": 3.8915257215384485e-05, "epoch": 0.9738339021615472, "percentage": 19.5, "elapsed_time": "4:24:39", "remaining_time": "18:12:40"}
+{"current_steps": 429, "total_steps": 2195, "loss": 0.4941, "lr": 3.890489839347347e-05, "epoch": 0.9761092150170648, "percentage": 19.54, "elapsed_time": "4:25:18", "remaining_time": "18:12:08"}
+{"current_steps": 430, "total_steps": 2195, "loss": 0.4752, "lr": 3.889449173722077e-05, "epoch": 0.9783845278725825, "percentage": 19.59, "elapsed_time": "4:25:53", "remaining_time": "18:11:24"}
+{"current_steps": 431, "total_steps": 2195, "loss": 0.5074, "lr": 3.8884037272957936e-05, "epoch": 0.9806598407281001, "percentage": 19.64, "elapsed_time": "4:26:34", "remaining_time": "18:11:03"}
+{"current_steps": 432, "total_steps": 2195, "loss": 0.5254, "lr": 3.887353502713752e-05, "epoch": 0.9829351535836177, "percentage": 19.68, "elapsed_time": "4:27:14", "remaining_time": "18:10:36"}
+{"current_steps": 433, "total_steps": 2195, "loss": 0.4802, "lr": 3.886298502633294e-05, "epoch": 0.9852104664391353, "percentage": 19.73, "elapsed_time": "4:27:56", "remaining_time": "18:10:18"}
+{"current_steps": 434, "total_steps": 2195, "loss": 0.4922, "lr": 3.885238729723847e-05, "epoch": 0.987485779294653, "percentage": 19.77, "elapsed_time": "4:28:35", "remaining_time": "18:09:50"}
+{"current_steps": 435, "total_steps": 2195, "loss": 0.5277, "lr": 3.8841741866669126e-05, "epoch": 0.9897610921501706, "percentage": 19.82, "elapsed_time": "4:29:14", "remaining_time": "18:09:19"}
+{"current_steps": 436, "total_steps": 2195, "loss": 0.5036, "lr": 3.883104876156064e-05, "epoch": 0.9920364050056882, "percentage": 19.86, "elapsed_time": "4:29:56", "remaining_time": "18:09:04"}
+{"current_steps": 437, "total_steps": 2195, "loss": 0.4993, "lr": 3.882030800896937e-05, "epoch": 0.9943117178612059, "percentage": 19.91, "elapsed_time": "4:30:38", "remaining_time": "18:08:44"}
+{"current_steps": 438, "total_steps": 2195, "loss": 0.5484, "lr": 3.880951963607222e-05, "epoch": 0.9965870307167235, "percentage": 19.95, "elapsed_time": "4:31:13", "remaining_time": "18:07:58"}
+{"current_steps": 439, "total_steps": 2195, "loss": 0.4769, "lr": 3.879868367016662e-05, "epoch": 0.9988623435722411, "percentage": 20.0, "elapsed_time": "4:31:50", "remaining_time": "18:07:21"}
+{"current_steps": 440, "total_steps": 2195, "loss": 0.722, "lr": 3.878780013867038e-05, "epoch": 1.0011376564277588, "percentage": 20.05, "elapsed_time": "4:33:16", "remaining_time": "18:09:57"}
+{"current_steps": 441, "total_steps": 2195, "loss": 0.3795, "lr": 3.877686906912168e-05, "epoch": 1.0034129692832765, "percentage": 20.09, "elapsed_time": "4:33:52", "remaining_time": "18:09:18"}
+{"current_steps": 442, "total_steps": 2195, "loss": 0.3871, "lr": 3.876589048917901e-05, "epoch": 1.005688282138794, "percentage": 20.14, "elapsed_time": "4:34:34", "remaining_time": "18:08:58"}
+{"current_steps": 443, "total_steps": 2195, "loss": 0.5004, "lr": 3.8754864426621045e-05, "epoch": 1.0079635949943118, "percentage": 20.18, "elapsed_time": "4:35:16", "remaining_time": "18:08:40"}
+{"current_steps": 444, "total_steps": 2195, "loss": 0.448, "lr": 3.874379090934659e-05, "epoch": 1.0102389078498293, "percentage": 20.23, "elapsed_time": "4:35:47", "remaining_time": "18:07:36"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da3737da36dd818440e9b627b251e463e8701ec839399d4d4f13022c99422f95
+size 7160