Add files using upload-large-folder tool

.gitattributes

@@ -40,3 +40,4 @@ sft_search/qwen7b_sft_871_checkpoint-78/tokenizer.json filter=lfs diff=lfs merge
 sft_search/JOB:28315\#LR:1e-5\#BASE:DeepSeek-R1-Distill-Qwen-7B\#TOKEN:DeepSeek-R1-Distill-Qwen-7B\#BSZ:2\#ACC:4_no_error_data_871/checkpoint-68/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 sft_search/JOB:30936\#LR:1e-5\#BASE:Qwen2.5-7B-Instruct\#TOKEN:Qwen2.5-7B-Instruct\#BSZ:2\#ACC:4_math_qwq_4524_selected_add_prompt_871/checkpoint-68/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 sft_search/JOB:30936\#LR:1e-5\#BASE:Qwen2.5-7B-Instruct\#TOKEN:Qwen2.5-7B-Instruct\#BSZ:2\#ACC:4_math_qwq_4524_selected_add_prompt_871/checkpoint-55/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+sft_search/JOB:24815\#LR:1e-5\#BASE:Qwen2.5-7B-Instruct\#TOKEN:Qwen2.5-7B-Instruct\#BSZ:2\#ACC:4_ablation_resp_difficulty_1678_ablation/checkpoint-156/tokenizer.json filter=lfs diff=lfs merge=lfs -text

sft_search/JOB:#LR:1e-5#BASE:QwQ-32B#TOKEN:QwQ-32B#BSZ:2#ACC:4_merged_syn_long_359_sft_1533/checkpoint-528/added_tokens.json

@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|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
+}

sft_search/JOB:#LR:1e-5#BASE:QwQ-32B#TOKEN:QwQ-32B#BSZ:2#ACC:4_merged_syn_long_359_sft_1533/checkpoint-528/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "/capacity/userdata/models/QwQ-32B",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 27648,
+  "max_position_embeddings": 131072,
+  "max_window_layers": 64,
+  "model_type": "qwen2",
+  "num_attention_heads": 40,
+  "num_hidden_layers": 64,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.0",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}

sft_search/JOB:#LR:1e-5#BASE:QwQ-32B#TOKEN:QwQ-32B#BSZ:2#ACC:4_merged_syn_long_359_sft_1533/checkpoint-528/generation_config.json

@@ -0,0 +1,13 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "temperature": 0.6,
+  "top_k": 40,
+  "top_p": 0.95,
+  "transformers_version": "4.46.0"
+}

sft_search/JOB:#LR:1e-5#BASE:QwQ-32B#TOKEN:QwQ-32B#BSZ:2#ACC:4_merged_syn_long_359_sft_1533/checkpoint-528/model.safetensors.index.json

@@ -0,0 +1,778 @@
+{
+  "metadata": {
+    "total_size": 65527752704
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00014-of-00014.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00004-of-00014.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00005-of-00014.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.28.self_attn.k_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.28.self_attn.q_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.28.self_attn.v_proj.bias": "model-00006-of-00014.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.29.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.29.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.29.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00014.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.30.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.30.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.30.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.31.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.31.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.31.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.32.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.32.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.32.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.33.self_attn.k_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.33.self_attn.q_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.33.self_attn.v_proj.bias": "model-00007-of-00014.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.34.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.34.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.34.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.35.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.35.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.35.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.36.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.36.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.36.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.36.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.36.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.37.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.37.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.37.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.37.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.38.mlp.gate_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.38.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.38.self_attn.k_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.38.self_attn.q_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.38.self_attn.v_proj.bias": "model-00008-of-00014.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.39.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.39.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.39.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.39.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.39.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.40.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.40.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.40.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.40.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.40.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.40.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.41.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.41.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.41.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.41.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.41.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.41.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.42.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.42.mlp.down_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.42.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.42.mlp.up_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.42.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.42.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.42.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.42.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.42.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.42.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.43.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.43.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.43.mlp.gate_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.43.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.43.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.43.self_attn.k_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.43.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.43.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.43.self_attn.q_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.43.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.43.self_attn.v_proj.bias": "model-00009-of-00014.safetensors",
+    "model.layers.43.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.44.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.44.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.44.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.44.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.44.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.44.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.44.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.44.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.44.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.44.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.44.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.44.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.45.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.45.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.45.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.45.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.45.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.45.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.45.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.45.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.45.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.45.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.45.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.45.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.46.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.46.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.46.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.46.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.46.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.46.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.46.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.46.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.46.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.46.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.46.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.46.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.47.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.47.mlp.down_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.47.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.47.mlp.up_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.47.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.47.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.47.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.47.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.47.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.47.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.47.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.47.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.48.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.48.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.48.mlp.gate_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.48.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.48.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.48.self_attn.k_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.48.self_attn.k_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.48.self_attn.o_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.48.self_attn.q_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.48.self_attn.q_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.48.self_attn.v_proj.bias": "model-00010-of-00014.safetensors",
+    "model.layers.48.self_attn.v_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.49.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.49.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.49.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.49.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.49.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.49.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.49.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.49.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.49.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.49.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.49.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.49.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.50.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.50.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.50.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.50.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.50.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.50.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.50.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.50.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.50.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.50.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.50.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.50.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.51.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.51.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.51.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.51.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.51.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.51.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.51.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.51.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.51.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.51.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.51.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.51.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.52.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.52.mlp.down_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.52.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.52.mlp.up_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.52.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.52.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.52.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.52.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.52.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.52.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.52.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.52.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.53.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.53.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.53.mlp.gate_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.53.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.53.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.53.self_attn.k_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.53.self_attn.k_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.53.self_attn.o_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.53.self_attn.q_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.53.self_attn.q_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.53.self_attn.v_proj.bias": "model-00011-of-00014.safetensors",
+    "model.layers.53.self_attn.v_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.54.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.54.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.54.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.54.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.54.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.54.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.54.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.54.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.54.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.54.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.54.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.54.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.55.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.55.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.55.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.55.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.55.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.55.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.55.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.55.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.55.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.55.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.55.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.55.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.56.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.56.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.56.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.56.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.56.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.56.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.56.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.56.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.56.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.56.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.56.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.56.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.57.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.57.mlp.down_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.57.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.57.mlp.up_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.57.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.57.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.57.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.57.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.57.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.57.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.57.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.57.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.58.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.58.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.58.mlp.gate_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.58.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.58.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.58.self_attn.k_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.58.self_attn.k_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.58.self_attn.o_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.58.self_attn.q_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.58.self_attn.q_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.58.self_attn.v_proj.bias": "model-00012-of-00014.safetensors",
+    "model.layers.58.self_attn.v_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.59.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.59.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.59.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.59.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.59.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.59.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.59.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.59.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.59.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.59.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.59.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.59.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.60.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.60.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.60.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.60.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.60.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.60.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.60.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.60.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.60.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.60.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.60.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.60.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.61.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.61.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.61.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.61.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.61.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.61.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.61.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.61.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.61.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.61.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.61.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.61.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.62.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.62.mlp.down_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.62.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.62.mlp.up_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.62.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.62.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.62.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.62.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.62.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.62.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.62.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.62.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.63.input_layernorm.weight": "model-00014-of-00014.safetensors",
+    "model.layers.63.mlp.down_proj.weight": "model-00014-of-00014.safetensors",
+    "model.layers.63.mlp.gate_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.63.mlp.up_proj.weight": "model-00014-of-00014.safetensors",
+    "model.layers.63.post_attention_layernorm.weight": "model-00014-of-00014.safetensors",
+    "model.layers.63.self_attn.k_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.63.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.63.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.63.self_attn.q_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.63.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.63.self_attn.v_proj.bias": "model-00013-of-00014.safetensors",
+    "model.layers.63.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00002-of-00014.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00003-of-00014.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
+    "model.norm.weight": "model-00014-of-00014.safetensors"
+  }
+}

sft_search/JOB:#LR:1e-5#BASE:QwQ-32B#TOKEN:QwQ-32B#BSZ:2#ACC:4_merged_syn_long_359_sft_1533/checkpoint-528/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": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

sft_search/JOB:#LR:1e-5#BASE:QwQ-32B#TOKEN:QwQ-32B#BSZ:2#ACC:4_merged_syn_long_359_sft_1533/checkpoint-528/tokenizer_config.json

@@ -0,0 +1,239 @@
+{
+  "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
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "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        {{- '' }}\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  {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" and not message.tool_calls %}\n        {%- set content = message.content.split('</think>')[-1].lstrip('\\n') %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set content = message.content.split('</think>')[-1].lstrip('\\n') %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + 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<think>\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 30000,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

sft_search/JOB:#LR:1e-5#BASE:QwQ-32B#TOKEN:QwQ-32B#BSZ:2#ACC:4_merged_syn_long_359_sft_1533/checkpoint-528/trainer_state.json

@@ -0,0 +1,3729 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 22.0,
+  "eval_steps": 500,
+  "global_step": 528,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.041666666666666664,
+      "grad_norm": 3.0420663356781006,
+      "learning_rate": 1.388888888888889e-09,
+      "loss": 0.3736,
+      "step": 1
+    },
+    {
+      "epoch": 0.08333333333333333,
+      "grad_norm": 2.994919776916504,
+      "learning_rate": 2.777777777777778e-09,
+      "loss": 0.3795,
+      "step": 2
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 3.0447194576263428,
+      "learning_rate": 4.1666666666666676e-09,
+      "loss": 0.3603,
+      "step": 3
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 3.121523380279541,
+      "learning_rate": 5.555555555555556e-09,
+      "loss": 0.3694,
+      "step": 4
+    },
+    {
+      "epoch": 0.20833333333333334,
+      "grad_norm": 2.954749822616577,
+      "learning_rate": 6.944444444444445e-09,
+      "loss": 0.3758,
+      "step": 5
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 3.192906379699707,
+      "learning_rate": 8.333333333333335e-09,
+      "loss": 0.3558,
+      "step": 6
+    },
+    {
+      "epoch": 0.2916666666666667,
+      "grad_norm": 3.1066291332244873,
+      "learning_rate": 9.722222222222223e-09,
+      "loss": 0.3572,
+      "step": 7
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 3.0212528705596924,
+      "learning_rate": 1.1111111111111112e-08,
+      "loss": 0.335,
+      "step": 8
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 2.9887983798980713,
+      "learning_rate": 1.2500000000000001e-08,
+      "loss": 0.3746,
+      "step": 9
+    },
+    {
+      "epoch": 0.4166666666666667,
+      "grad_norm": 3.715954303741455,
+      "learning_rate": 1.388888888888889e-08,
+      "loss": 0.3545,
+      "step": 10
+    },
+    {
+      "epoch": 0.4583333333333333,
+      "grad_norm": 3.0021579265594482,
+      "learning_rate": 1.527777777777778e-08,
+      "loss": 0.3475,
+      "step": 11
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 3.043217897415161,
+      "learning_rate": 1.666666666666667e-08,
+      "loss": 0.359,
+      "step": 12
+    },
+    {
+      "epoch": 0.5416666666666666,
+      "grad_norm": 3.113858222961426,
+      "learning_rate": 1.8055555555555556e-08,
+      "loss": 0.3443,
+      "step": 13
+    },
+    {
+      "epoch": 0.5833333333333334,
+      "grad_norm": 2.868614435195923,
+      "learning_rate": 1.9444444444444445e-08,
+      "loss": 0.3726,
+      "step": 14
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 2.9660096168518066,
+      "learning_rate": 2.0833333333333335e-08,
+      "loss": 0.3837,
+      "step": 15
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 2.9872426986694336,
+      "learning_rate": 2.2222222222222224e-08,
+      "loss": 0.3538,
+      "step": 16
+    },
+    {
+      "epoch": 0.7083333333333334,
+      "grad_norm": 2.89996600151062,
+      "learning_rate": 2.3611111111111113e-08,
+      "loss": 0.3786,
+      "step": 17
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 3.2942259311676025,
+      "learning_rate": 2.5000000000000002e-08,
+      "loss": 0.3484,
+      "step": 18
+    },
+    {
+      "epoch": 0.7916666666666666,
+      "grad_norm": 3.223271369934082,
+      "learning_rate": 2.638888888888889e-08,
+      "loss": 0.3308,
+      "step": 19
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 3.1012966632843018,
+      "learning_rate": 2.777777777777778e-08,
+      "loss": 0.3444,
+      "step": 20
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 2.9651427268981934,
+      "learning_rate": 2.916666666666667e-08,
+      "loss": 0.3582,
+      "step": 21
+    },
+    {
+      "epoch": 0.9166666666666666,
+      "grad_norm": 3.128702163696289,
+      "learning_rate": 3.055555555555556e-08,
+      "loss": 0.3476,
+      "step": 22
+    },
+    {
+      "epoch": 0.9583333333333334,
+      "grad_norm": 2.8862669467926025,
+      "learning_rate": 3.194444444444445e-08,
+      "loss": 0.353,
+      "step": 23
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 3.1079020500183105,
+      "learning_rate": 3.333333333333334e-08,
+      "loss": 0.3489,
+      "step": 24
+    },
+    {
+      "epoch": 1.0416666666666667,
+      "grad_norm": 3.021883487701416,
+      "learning_rate": 3.472222222222222e-08,
+      "loss": 0.3563,
+      "step": 25
+    },
+    {
+      "epoch": 1.0833333333333333,
+      "grad_norm": 3.103222370147705,
+      "learning_rate": 3.611111111111111e-08,
+      "loss": 0.3556,
+      "step": 26
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 3.0339605808258057,
+      "learning_rate": 3.7500000000000005e-08,
+      "loss": 0.3516,
+      "step": 27
+    },
+    {
+      "epoch": 1.1666666666666667,
+      "grad_norm": 3.15885329246521,
+      "learning_rate": 3.888888888888889e-08,
+      "loss": 0.3519,
+      "step": 28
+    },
+    {
+      "epoch": 1.2083333333333333,
+      "grad_norm": 3.1420021057128906,
+      "learning_rate": 4.027777777777778e-08,
+      "loss": 0.3433,
+      "step": 29
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 3.124666929244995,
+      "learning_rate": 4.166666666666667e-08,
+      "loss": 0.3637,
+      "step": 30
+    },
+    {
+      "epoch": 1.2916666666666667,
+      "grad_norm": 2.8410370349884033,
+      "learning_rate": 4.305555555555556e-08,
+      "loss": 0.3734,
+      "step": 31
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 3.0200390815734863,
+      "learning_rate": 4.444444444444445e-08,
+      "loss": 0.3503,
+      "step": 32
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 3.1030161380767822,
+      "learning_rate": 4.583333333333334e-08,
+      "loss": 0.3548,
+      "step": 33
+    },
+    {
+      "epoch": 1.4166666666666667,
+      "grad_norm": 2.9274024963378906,
+      "learning_rate": 4.7222222222222226e-08,
+      "loss": 0.3756,
+      "step": 34
+    },
+    {
+      "epoch": 1.4583333333333333,
+      "grad_norm": 2.7859292030334473,
+      "learning_rate": 4.861111111111112e-08,
+      "loss": 0.3762,
+      "step": 35
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 3.281543254852295,
+      "learning_rate": 5.0000000000000004e-08,
+      "loss": 0.3551,
+      "step": 36
+    },
+    {
+      "epoch": 1.5416666666666665,
+      "grad_norm": 2.8933916091918945,
+      "learning_rate": 5.13888888888889e-08,
+      "loss": 0.3646,
+      "step": 37
+    },
+    {
+      "epoch": 1.5833333333333335,
+      "grad_norm": 3.041745662689209,
+      "learning_rate": 5.277777777777778e-08,
+      "loss": 0.349,
+      "step": 38
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 3.3109254837036133,
+      "learning_rate": 5.4166666666666675e-08,
+      "loss": 0.3488,
+      "step": 39
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 3.1408932209014893,
+      "learning_rate": 5.555555555555556e-08,
+      "loss": 0.3492,
+      "step": 40
+    },
+    {
+      "epoch": 1.7083333333333335,
+      "grad_norm": 3.164673089981079,
+      "learning_rate": 5.6944444444444454e-08,
+      "loss": 0.3479,
+      "step": 41
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 2.921013355255127,
+      "learning_rate": 5.833333333333334e-08,
+      "loss": 0.3856,
+      "step": 42
+    },
+    {
+      "epoch": 1.7916666666666665,
+      "grad_norm": 3.0510287284851074,
+      "learning_rate": 5.972222222222223e-08,
+      "loss": 0.3345,
+      "step": 43
+    },
+    {
+      "epoch": 1.8333333333333335,
+      "grad_norm": 2.9552886486053467,
+      "learning_rate": 6.111111111111112e-08,
+      "loss": 0.3754,
+      "step": 44
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 2.8241865634918213,
+      "learning_rate": 6.250000000000001e-08,
+      "loss": 0.346,
+      "step": 45
+    },
+    {
+      "epoch": 1.9166666666666665,
+      "grad_norm": 3.1247987747192383,
+      "learning_rate": 6.38888888888889e-08,
+      "loss": 0.3488,
+      "step": 46
+    },
+    {
+      "epoch": 1.9583333333333335,
+      "grad_norm": 3.0169198513031006,
+      "learning_rate": 6.527777777777778e-08,
+      "loss": 0.3407,
+      "step": 47
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.9057247638702393,
+      "learning_rate": 6.666666666666668e-08,
+      "loss": 0.3602,
+      "step": 48
+    },
+    {
+      "epoch": 2.0416666666666665,
+      "grad_norm": 3.0435783863067627,
+      "learning_rate": 6.805555555555555e-08,
+      "loss": 0.3353,
+      "step": 49
+    },
+    {
+      "epoch": 2.0833333333333335,
+      "grad_norm": 2.9771666526794434,
+      "learning_rate": 6.944444444444444e-08,
+      "loss": 0.3558,
+      "step": 50
+    },
+    {
+      "epoch": 2.125,
+      "grad_norm": 2.9318928718566895,
+      "learning_rate": 7.083333333333334e-08,
+      "loss": 0.3569,
+      "step": 51
+    },
+    {
+      "epoch": 2.1666666666666665,
+      "grad_norm": 2.918715715408325,
+      "learning_rate": 7.222222222222222e-08,
+      "loss": 0.3612,
+      "step": 52
+    },
+    {
+      "epoch": 2.2083333333333335,
+      "grad_norm": 3.207113742828369,
+      "learning_rate": 7.361111111111111e-08,
+      "loss": 0.3689,
+      "step": 53
+    },
+    {
+      "epoch": 2.25,
+      "grad_norm": 3.0282905101776123,
+      "learning_rate": 7.500000000000001e-08,
+      "loss": 0.3644,
+      "step": 54
+    },
+    {
+      "epoch": 2.2916666666666665,
+      "grad_norm": 3.0210957527160645,
+      "learning_rate": 7.63888888888889e-08,
+      "loss": 0.3457,
+      "step": 55
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 2.9547502994537354,
+      "learning_rate": 7.777777777777778e-08,
+      "loss": 0.3801,
+      "step": 56
+    },
+    {
+      "epoch": 2.375,
+      "grad_norm": 2.9681663513183594,
+      "learning_rate": 7.916666666666668e-08,
+      "loss": 0.3414,
+      "step": 57
+    },
+    {
+      "epoch": 2.4166666666666665,
+      "grad_norm": 3.0110018253326416,
+      "learning_rate": 8.055555555555557e-08,
+      "loss": 0.339,
+      "step": 58
+    },
+    {
+      "epoch": 2.4583333333333335,
+      "grad_norm": 2.954514265060425,
+      "learning_rate": 8.194444444444445e-08,
+      "loss": 0.3439,
+      "step": 59
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 2.874364137649536,
+      "learning_rate": 8.333333333333334e-08,
+      "loss": 0.3488,
+      "step": 60
+    },
+    {
+      "epoch": 2.5416666666666665,
+      "grad_norm": 2.9033541679382324,
+      "learning_rate": 8.472222222222224e-08,
+      "loss": 0.3771,
+      "step": 61
+    },
+    {
+      "epoch": 2.5833333333333335,
+      "grad_norm": 2.967888116836548,
+      "learning_rate": 8.611111111111112e-08,
+      "loss": 0.3573,
+      "step": 62
+    },
+    {
+      "epoch": 2.625,
+      "grad_norm": 2.920008897781372,
+      "learning_rate": 8.750000000000001e-08,
+      "loss": 0.3415,
+      "step": 63
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 2.958160400390625,
+      "learning_rate": 8.88888888888889e-08,
+      "loss": 0.3635,
+      "step": 64
+    },
+    {
+      "epoch": 2.7083333333333335,
+      "grad_norm": 2.921083450317383,
+      "learning_rate": 9.027777777777778e-08,
+      "loss": 0.3387,
+      "step": 65
+    },
+    {
+      "epoch": 2.75,
+      "grad_norm": 2.9987378120422363,
+      "learning_rate": 9.166666666666668e-08,
+      "loss": 0.3482,
+      "step": 66
+    },
+    {
+      "epoch": 2.7916666666666665,
+      "grad_norm": 2.878142833709717,
+      "learning_rate": 9.305555555555555e-08,
+      "loss": 0.3668,
+      "step": 67
+    },
+    {
+      "epoch": 2.8333333333333335,
+      "grad_norm": 2.827054500579834,
+      "learning_rate": 9.444444444444445e-08,
+      "loss": 0.3461,
+      "step": 68
+    },
+    {
+      "epoch": 2.875,
+      "grad_norm": 2.887481451034546,
+      "learning_rate": 9.583333333333334e-08,
+      "loss": 0.3721,
+      "step": 69
+    },
+    {
+      "epoch": 2.9166666666666665,
+      "grad_norm": 2.9160611629486084,
+      "learning_rate": 9.722222222222224e-08,
+      "loss": 0.3548,
+      "step": 70
+    },
+    {
+      "epoch": 2.9583333333333335,
+      "grad_norm": 2.8325109481811523,
+      "learning_rate": 9.861111111111111e-08,
+      "loss": 0.333,
+      "step": 71
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.872144937515259,
+      "learning_rate": 1.0000000000000001e-07,
+      "loss": 0.3457,
+      "step": 72
+    },
+    {
+      "epoch": 3.0416666666666665,
+      "grad_norm": 2.8224565982818604,
+      "learning_rate": 1.013888888888889e-07,
+      "loss": 0.3342,
+      "step": 73
+    },
+    {
+      "epoch": 3.0833333333333335,
+      "grad_norm": 2.94210147857666,
+      "learning_rate": 1.027777777777778e-07,
+      "loss": 0.3541,
+      "step": 74
+    },
+    {
+      "epoch": 3.125,
+      "grad_norm": 3.195175886154175,
+      "learning_rate": 1.0416666666666667e-07,
+      "loss": 0.3536,
+      "step": 75
+    },
+    {
+      "epoch": 3.1666666666666665,
+      "grad_norm": 2.9079749584198,
+      "learning_rate": 1.0555555555555557e-07,
+      "loss": 0.3516,
+      "step": 76
+    },
+    {
+      "epoch": 3.2083333333333335,
+      "grad_norm": 2.976221799850464,
+      "learning_rate": 1.0694444444444445e-07,
+      "loss": 0.362,
+      "step": 77
+    },
+    {
+      "epoch": 3.25,
+      "grad_norm": 3.0174481868743896,
+      "learning_rate": 1.0833333333333335e-07,
+      "loss": 0.3054,
+      "step": 78
+    },
+    {
+      "epoch": 3.2916666666666665,
+      "grad_norm": 2.8203630447387695,
+      "learning_rate": 1.0972222222222222e-07,
+      "loss": 0.3371,
+      "step": 79
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 2.852158546447754,
+      "learning_rate": 1.1111111111111112e-07,
+      "loss": 0.3462,
+      "step": 80
+    },
+    {
+      "epoch": 3.375,
+      "grad_norm": 2.8027091026306152,
+      "learning_rate": 1.1250000000000001e-07,
+      "loss": 0.3412,
+      "step": 81
+    },
+    {
+      "epoch": 3.4166666666666665,
+      "grad_norm": 2.7673850059509277,
+      "learning_rate": 1.1388888888888891e-07,
+      "loss": 0.3602,
+      "step": 82
+    },
+    {
+      "epoch": 3.4583333333333335,
+      "grad_norm": 2.704030990600586,
+      "learning_rate": 1.1527777777777778e-07,
+      "loss": 0.3333,
+      "step": 83
+    },
+    {
+      "epoch": 3.5,
+      "grad_norm": 2.733429193496704,
+      "learning_rate": 1.1666666666666668e-07,
+      "loss": 0.3446,
+      "step": 84
+    },
+    {
+      "epoch": 3.5416666666666665,
+      "grad_norm": 2.769155502319336,
+      "learning_rate": 1.1805555555555556e-07,
+      "loss": 0.3265,
+      "step": 85
+    },
+    {
+      "epoch": 3.5833333333333335,
+      "grad_norm": 2.684331178665161,
+      "learning_rate": 1.1944444444444445e-07,
+      "loss": 0.3204,
+      "step": 86
+    },
+    {
+      "epoch": 3.625,
+      "grad_norm": 2.5963892936706543,
+      "learning_rate": 1.2083333333333335e-07,
+      "loss": 0.3274,
+      "step": 87
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 2.704686403274536,
+      "learning_rate": 1.2222222222222225e-07,
+      "loss": 0.3841,
+      "step": 88
+    },
+    {
+      "epoch": 3.7083333333333335,
+      "grad_norm": 2.5046286582946777,
+      "learning_rate": 1.2361111111111112e-07,
+      "loss": 0.3182,
+      "step": 89
+    },
+    {
+      "epoch": 3.75,
+      "grad_norm": 2.624610185623169,
+      "learning_rate": 1.2500000000000002e-07,
+      "loss": 0.3385,
+      "step": 90
+    },
+    {
+      "epoch": 3.7916666666666665,
+      "grad_norm": 2.643967628479004,
+      "learning_rate": 1.263888888888889e-07,
+      "loss": 0.3514,
+      "step": 91
+    },
+    {
+      "epoch": 3.8333333333333335,
+      "grad_norm": 2.4959986209869385,
+      "learning_rate": 1.277777777777778e-07,
+      "loss": 0.3116,
+      "step": 92
+    },
+    {
+      "epoch": 3.875,
+      "grad_norm": 2.5002784729003906,
+      "learning_rate": 1.2916666666666667e-07,
+      "loss": 0.333,
+      "step": 93
+    },
+    {
+      "epoch": 3.9166666666666665,
+      "grad_norm": 2.6232120990753174,
+      "learning_rate": 1.3055555555555556e-07,
+      "loss": 0.3528,
+      "step": 94
+    },
+    {
+      "epoch": 3.9583333333333335,
+      "grad_norm": 2.924082040786743,
+      "learning_rate": 1.3194444444444446e-07,
+      "loss": 0.323,
+      "step": 95
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.642420530319214,
+      "learning_rate": 1.3333333333333336e-07,
+      "loss": 0.3405,
+      "step": 96
+    },
+    {
+      "epoch": 4.041666666666667,
+      "grad_norm": 2.5466082096099854,
+      "learning_rate": 1.3472222222222224e-07,
+      "loss": 0.3586,
+      "step": 97
+    },
+    {
+      "epoch": 4.083333333333333,
+      "grad_norm": 2.578134536743164,
+      "learning_rate": 1.361111111111111e-07,
+      "loss": 0.3187,
+      "step": 98
+    },
+    {
+      "epoch": 4.125,
+      "grad_norm": 2.4417221546173096,
+      "learning_rate": 1.375e-07,
+      "loss": 0.3273,
+      "step": 99
+    },
+    {
+      "epoch": 4.166666666666667,
+      "grad_norm": 2.417306900024414,
+      "learning_rate": 1.3888888888888888e-07,
+      "loss": 0.332,
+      "step": 100
+    },
+    {
+      "epoch": 4.208333333333333,
+      "grad_norm": 2.5107898712158203,
+      "learning_rate": 1.4027777777777778e-07,
+      "loss": 0.3274,
+      "step": 101
+    },
+    {
+      "epoch": 4.25,
+      "grad_norm": 2.4568309783935547,
+      "learning_rate": 1.4166666666666668e-07,
+      "loss": 0.3257,
+      "step": 102
+    },
+    {
+      "epoch": 4.291666666666667,
+      "grad_norm": 2.544175386428833,
+      "learning_rate": 1.4305555555555558e-07,
+      "loss": 0.3092,
+      "step": 103
+    },
+    {
+      "epoch": 4.333333333333333,
+      "grad_norm": 2.4726674556732178,
+      "learning_rate": 1.4444444444444445e-07,
+      "loss": 0.3349,
+      "step": 104
+    },
+    {
+      "epoch": 4.375,
+      "grad_norm": 2.43094801902771,
+      "learning_rate": 1.4583333333333335e-07,
+      "loss": 0.3332,
+      "step": 105
+    },
+    {
+      "epoch": 4.416666666666667,
+      "grad_norm": 2.3865795135498047,
+      "learning_rate": 1.4722222222222222e-07,
+      "loss": 0.3416,
+      "step": 106
+    },
+    {
+      "epoch": 4.458333333333333,
+      "grad_norm": 2.3298897743225098,
+      "learning_rate": 1.4861111111111112e-07,
+      "loss": 0.3159,
+      "step": 107
+    },
+    {
+      "epoch": 4.5,
+      "grad_norm": 2.3117189407348633,
+      "learning_rate": 1.5000000000000002e-07,
+      "loss": 0.2945,
+      "step": 108
+    },
+    {
+      "epoch": 4.541666666666667,
+      "grad_norm": 3.7656211853027344,
+      "learning_rate": 1.513888888888889e-07,
+      "loss": 0.3116,
+      "step": 109
+    },
+    {
+      "epoch": 4.583333333333333,
+      "grad_norm": 2.165956974029541,
+      "learning_rate": 1.527777777777778e-07,
+      "loss": 0.306,
+      "step": 110
+    },
+    {
+      "epoch": 4.625,
+      "grad_norm": 2.1313440799713135,
+      "learning_rate": 1.541666666666667e-07,
+      "loss": 0.296,
+      "step": 111
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 2.0461480617523193,
+      "learning_rate": 1.5555555555555556e-07,
+      "loss": 0.3117,
+      "step": 112
+    },
+    {
+      "epoch": 4.708333333333333,
+      "grad_norm": 2.2418055534362793,
+      "learning_rate": 1.5694444444444446e-07,
+      "loss": 0.2948,
+      "step": 113
+    },
+    {
+      "epoch": 4.75,
+      "grad_norm": 2.040050506591797,
+      "learning_rate": 1.5833333333333336e-07,
+      "loss": 0.3287,
+      "step": 114
+    },
+    {
+      "epoch": 4.791666666666667,
+      "grad_norm": 2.0866613388061523,
+      "learning_rate": 1.5972222222222223e-07,
+      "loss": 0.3385,
+      "step": 115
+    },
+    {
+      "epoch": 4.833333333333333,
+      "grad_norm": 1.9137110710144043,
+      "learning_rate": 1.6111111111111113e-07,
+      "loss": 0.3179,
+      "step": 116
+    },
+    {
+      "epoch": 4.875,
+      "grad_norm": 2.0457868576049805,
+      "learning_rate": 1.625e-07,
+      "loss": 0.336,
+      "step": 117
+    },
+    {
+      "epoch": 4.916666666666667,
+      "grad_norm": 2.0469157695770264,
+      "learning_rate": 1.638888888888889e-07,
+      "loss": 0.3468,
+      "step": 118
+    },
+    {
+      "epoch": 4.958333333333333,
+      "grad_norm": 1.9149034023284912,
+      "learning_rate": 1.6527777777777778e-07,
+      "loss": 0.3044,
+      "step": 119
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.9285746812820435,
+      "learning_rate": 1.6666666666666668e-07,
+      "loss": 0.3047,
+      "step": 120
+    },
+    {
+      "epoch": 5.041666666666667,
+      "grad_norm": 1.811112642288208,
+      "learning_rate": 1.6805555555555558e-07,
+      "loss": 0.3183,
+      "step": 121
+    },
+    {
+      "epoch": 5.083333333333333,
+      "grad_norm": 1.9258331060409546,
+      "learning_rate": 1.6944444444444448e-07,
+      "loss": 0.3166,
+      "step": 122
+    },
+    {
+      "epoch": 5.125,
+      "grad_norm": 1.8934910297393799,
+      "learning_rate": 1.7083333333333335e-07,
+      "loss": 0.2836,
+      "step": 123
+    },
+    {
+      "epoch": 5.166666666666667,
+      "grad_norm": 2.0665881633758545,
+      "learning_rate": 1.7222222222222225e-07,
+      "loss": 0.2818,
+      "step": 124
+    },
+    {
+      "epoch": 5.208333333333333,
+      "grad_norm": 1.805400013923645,
+      "learning_rate": 1.7361111111111115e-07,
+      "loss": 0.3076,
+      "step": 125
+    },
+    {
+      "epoch": 5.25,
+      "grad_norm": 1.8032187223434448,
+      "learning_rate": 1.7500000000000002e-07,
+      "loss": 0.3089,
+      "step": 126
+    },
+    {
+      "epoch": 5.291666666666667,
+      "grad_norm": 1.8458822965621948,
+      "learning_rate": 1.763888888888889e-07,
+      "loss": 0.2854,
+      "step": 127
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 1.8904489278793335,
+      "learning_rate": 1.777777777777778e-07,
+      "loss": 0.328,
+      "step": 128
+    },
+    {
+      "epoch": 5.375,
+      "grad_norm": 1.8797101974487305,
+      "learning_rate": 1.791666666666667e-07,
+      "loss": 0.2959,
+      "step": 129
+    },
+    {
+      "epoch": 5.416666666666667,
+      "grad_norm": 1.803653597831726,
+      "learning_rate": 1.8055555555555556e-07,
+      "loss": 0.3015,
+      "step": 130
+    },
+    {
+      "epoch": 5.458333333333333,
+      "grad_norm": 1.7006596326828003,
+      "learning_rate": 1.8194444444444446e-07,
+      "loss": 0.3023,
+      "step": 131
+    },
+    {
+      "epoch": 5.5,
+      "grad_norm": 1.7527501583099365,
+      "learning_rate": 1.8333333333333336e-07,
+      "loss": 0.2986,
+      "step": 132
+    },
+    {
+      "epoch": 5.541666666666667,
+      "grad_norm": 1.742294192314148,
+      "learning_rate": 1.8472222222222226e-07,
+      "loss": 0.3512,
+      "step": 133
+    },
+    {
+      "epoch": 5.583333333333333,
+      "grad_norm": 1.7947728633880615,
+      "learning_rate": 1.861111111111111e-07,
+      "loss": 0.3315,
+      "step": 134
+    },
+    {
+      "epoch": 5.625,
+      "grad_norm": 2.2189319133758545,
+      "learning_rate": 1.875e-07,
+      "loss": 0.3007,
+      "step": 135
+    },
+    {
+      "epoch": 5.666666666666667,
+      "grad_norm": 1.5900083780288696,
+      "learning_rate": 1.888888888888889e-07,
+      "loss": 0.3097,
+      "step": 136
+    },
+    {
+      "epoch": 5.708333333333333,
+      "grad_norm": 1.666605830192566,
+      "learning_rate": 1.902777777777778e-07,
+      "loss": 0.2978,
+      "step": 137
+    },
+    {
+      "epoch": 5.75,
+      "grad_norm": 1.9505888223648071,
+      "learning_rate": 1.9166666666666668e-07,
+      "loss": 0.2945,
+      "step": 138
+    },
+    {
+      "epoch": 5.791666666666667,
+      "grad_norm": 1.7134442329406738,
+      "learning_rate": 1.9305555555555557e-07,
+      "loss": 0.2899,
+      "step": 139
+    },
+    {
+      "epoch": 5.833333333333333,
+      "grad_norm": 1.6610468626022339,
+      "learning_rate": 1.9444444444444447e-07,
+      "loss": 0.2843,
+      "step": 140
+    },
+    {
+      "epoch": 5.875,
+      "grad_norm": 1.7936415672302246,
+      "learning_rate": 1.9583333333333337e-07,
+      "loss": 0.2912,
+      "step": 141
+    },
+    {
+      "epoch": 5.916666666666667,
+      "grad_norm": 1.7085412740707397,
+      "learning_rate": 1.9722222222222222e-07,
+      "loss": 0.2899,
+      "step": 142
+    },
+    {
+      "epoch": 5.958333333333333,
+      "grad_norm": 1.4961903095245361,
+      "learning_rate": 1.9861111111111112e-07,
+      "loss": 0.3261,
+      "step": 143
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.5674827098846436,
+      "learning_rate": 2.0000000000000002e-07,
+      "loss": 0.3114,
+      "step": 144
+    },
+    {
+      "epoch": 6.041666666666667,
+      "grad_norm": 1.7490308284759521,
+      "learning_rate": 2.0138888888888892e-07,
+      "loss": 0.2702,
+      "step": 145
+    },
+    {
+      "epoch": 6.083333333333333,
+      "grad_norm": 1.4944590330123901,
+      "learning_rate": 2.027777777777778e-07,
+      "loss": 0.3148,
+      "step": 146
+    },
+    {
+      "epoch": 6.125,
+      "grad_norm": 1.4320622682571411,
+      "learning_rate": 2.041666666666667e-07,
+      "loss": 0.3125,
+      "step": 147
+    },
+    {
+      "epoch": 6.166666666666667,
+      "grad_norm": 1.5699678659439087,
+      "learning_rate": 2.055555555555556e-07,
+      "loss": 0.3001,
+      "step": 148
+    },
+    {
+      "epoch": 6.208333333333333,
+      "grad_norm": 1.5706146955490112,
+      "learning_rate": 2.0694444444444449e-07,
+      "loss": 0.2786,
+      "step": 149
+    },
+    {
+      "epoch": 6.25,
+      "grad_norm": 1.5522855520248413,
+      "learning_rate": 2.0833333333333333e-07,
+      "loss": 0.3057,
+      "step": 150
+    },
+    {
+      "epoch": 6.291666666666667,
+      "grad_norm": 1.583873987197876,
+      "learning_rate": 2.0972222222222223e-07,
+      "loss": 0.2849,
+      "step": 151
+    },
+    {
+      "epoch": 6.333333333333333,
+      "grad_norm": 1.480340838432312,
+      "learning_rate": 2.1111111111111113e-07,
+      "loss": 0.323,
+      "step": 152
+    },
+    {
+      "epoch": 6.375,
+      "grad_norm": 1.3989101648330688,
+      "learning_rate": 2.1250000000000003e-07,
+      "loss": 0.3083,
+      "step": 153
+    },
+    {
+      "epoch": 6.416666666666667,
+      "grad_norm": 1.440209984779358,
+      "learning_rate": 2.138888888888889e-07,
+      "loss": 0.3132,
+      "step": 154
+    },
+    {
+      "epoch": 6.458333333333333,
+      "grad_norm": 1.4770876169204712,
+      "learning_rate": 2.152777777777778e-07,
+      "loss": 0.3102,
+      "step": 155
+    },
+    {
+      "epoch": 6.5,
+      "grad_norm": 1.4692769050598145,
+      "learning_rate": 2.166666666666667e-07,
+      "loss": 0.2842,
+      "step": 156
+    },
+    {
+      "epoch": 6.541666666666667,
+      "grad_norm": 1.5429232120513916,
+      "learning_rate": 2.180555555555556e-07,
+      "loss": 0.2806,
+      "step": 157
+    },
+    {
+      "epoch": 6.583333333333333,
+      "grad_norm": 1.5067572593688965,
+      "learning_rate": 2.1944444444444445e-07,
+      "loss": 0.2594,
+      "step": 158
+    },
+    {
+      "epoch": 6.625,
+      "grad_norm": 1.4937934875488281,
+      "learning_rate": 2.2083333333333335e-07,
+      "loss": 0.2918,
+      "step": 159
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.5953493118286133,
+      "learning_rate": 2.2222222222222224e-07,
+      "loss": 0.2716,
+      "step": 160
+    },
+    {
+      "epoch": 6.708333333333333,
+      "grad_norm": 1.7118210792541504,
+      "learning_rate": 2.2361111111111112e-07,
+      "loss": 0.3177,
+      "step": 161
+    },
+    {
+      "epoch": 6.75,
+      "grad_norm": 1.7111620903015137,
+      "learning_rate": 2.2500000000000002e-07,
+      "loss": 0.2741,
+      "step": 162
+    },
+    {
+      "epoch": 6.791666666666667,
+      "grad_norm": 1.466434359550476,
+      "learning_rate": 2.2638888888888892e-07,
+      "loss": 0.2956,
+      "step": 163
+    },
+    {
+      "epoch": 6.833333333333333,
+      "grad_norm": 1.4168503284454346,
+      "learning_rate": 2.2777777777777781e-07,
+      "loss": 0.2887,
+      "step": 164
+    },
+    {
+      "epoch": 6.875,
+      "grad_norm": 1.866154670715332,
+      "learning_rate": 2.2916666666666666e-07,
+      "loss": 0.2902,
+      "step": 165
+    },
+    {
+      "epoch": 6.916666666666667,
+      "grad_norm": 1.5482758283615112,
+      "learning_rate": 2.3055555555555556e-07,
+      "loss": 0.2783,
+      "step": 166
+    },
+    {
+      "epoch": 6.958333333333333,
+      "grad_norm": 1.5047681331634521,
+      "learning_rate": 2.3194444444444446e-07,
+      "loss": 0.2832,
+      "step": 167
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.5048089027404785,
+      "learning_rate": 2.3333333333333336e-07,
+      "loss": 0.2847,
+      "step": 168
+    },
+    {
+      "epoch": 7.041666666666667,
+      "grad_norm": 1.4647302627563477,
+      "learning_rate": 2.3472222222222223e-07,
+      "loss": 0.2804,
+      "step": 169
+    },
+    {
+      "epoch": 7.083333333333333,
+      "grad_norm": 1.4436026811599731,
+      "learning_rate": 2.3611111111111113e-07,
+      "loss": 0.2836,
+      "step": 170
+    },
+    {
+      "epoch": 7.125,
+      "grad_norm": 1.2146880626678467,
+      "learning_rate": 2.3750000000000003e-07,
+      "loss": 0.2983,
+      "step": 171
+    },
+    {
+      "epoch": 7.166666666666667,
+      "grad_norm": 1.2718520164489746,
+      "learning_rate": 2.388888888888889e-07,
+      "loss": 0.3131,
+      "step": 172
+    },
+    {
+      "epoch": 7.208333333333333,
+      "grad_norm": 1.4090962409973145,
+      "learning_rate": 2.402777777777778e-07,
+      "loss": 0.2617,
+      "step": 173
+    },
+    {
+      "epoch": 7.25,
+      "grad_norm": 1.345558762550354,
+      "learning_rate": 2.416666666666667e-07,
+      "loss": 0.2922,
+      "step": 174
+    },
+    {
+      "epoch": 7.291666666666667,
+      "grad_norm": 1.5957825183868408,
+      "learning_rate": 2.430555555555556e-07,
+      "loss": 0.2656,
+      "step": 175
+    },
+    {
+      "epoch": 7.333333333333333,
+      "grad_norm": 1.1209359169006348,
+      "learning_rate": 2.444444444444445e-07,
+      "loss": 0.3198,
+      "step": 176
+    },
+    {
+      "epoch": 7.375,
+      "grad_norm": 1.459965705871582,
+      "learning_rate": 2.4583333333333334e-07,
+      "loss": 0.2834,
+      "step": 177
+    },
+    {
+      "epoch": 7.416666666666667,
+      "grad_norm": 1.3747780323028564,
+      "learning_rate": 2.4722222222222224e-07,
+      "loss": 0.2681,
+      "step": 178
+    },
+    {
+      "epoch": 7.458333333333333,
+      "grad_norm": 1.4269925355911255,
+      "learning_rate": 2.4861111111111114e-07,
+      "loss": 0.2791,
+      "step": 179
+    },
+    {
+      "epoch": 7.5,
+      "grad_norm": 1.5597667694091797,
+      "learning_rate": 2.5000000000000004e-07,
+      "loss": 0.2599,
+      "step": 180
+    },
+    {
+      "epoch": 7.541666666666667,
+      "grad_norm": 1.275232195854187,
+      "learning_rate": 2.513888888888889e-07,
+      "loss": 0.2778,
+      "step": 181
+    },
+    {
+      "epoch": 7.583333333333333,
+      "grad_norm": 1.294310450553894,
+      "learning_rate": 2.527777777777778e-07,
+      "loss": 0.2751,
+      "step": 182
+    },
+    {
+      "epoch": 7.625,
+      "grad_norm": 1.4105271100997925,
+      "learning_rate": 2.541666666666667e-07,
+      "loss": 0.2684,
+      "step": 183
+    },
+    {
+      "epoch": 7.666666666666667,
+      "grad_norm": 1.30532968044281,
+      "learning_rate": 2.555555555555556e-07,
+      "loss": 0.2576,
+      "step": 184
+    },
+    {
+      "epoch": 7.708333333333333,
+      "grad_norm": 1.2970374822616577,
+      "learning_rate": 2.5694444444444443e-07,
+      "loss": 0.2865,
+      "step": 185
+    },
+    {
+      "epoch": 7.75,
+      "grad_norm": 1.2033214569091797,
+      "learning_rate": 2.5833333333333333e-07,
+      "loss": 0.2995,
+      "step": 186
+    },
+    {
+      "epoch": 7.791666666666667,
+      "grad_norm": 1.2993123531341553,
+      "learning_rate": 2.5972222222222223e-07,
+      "loss": 0.2742,
+      "step": 187
+    },
+    {
+      "epoch": 7.833333333333333,
+      "grad_norm": 1.2954715490341187,
+      "learning_rate": 2.6111111111111113e-07,
+      "loss": 0.2857,
+      "step": 188
+    },
+    {
+      "epoch": 7.875,
+      "grad_norm": 1.240033507347107,
+      "learning_rate": 2.6250000000000003e-07,
+      "loss": 0.2919,
+      "step": 189
+    },
+    {
+      "epoch": 7.916666666666667,
+      "grad_norm": 1.4194754362106323,
+      "learning_rate": 2.6388888888888893e-07,
+      "loss": 0.255,
+      "step": 190
+    },
+    {
+      "epoch": 7.958333333333333,
+      "grad_norm": 1.2441984415054321,
+      "learning_rate": 2.652777777777778e-07,
+      "loss": 0.2835,
+      "step": 191
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.2889862060546875,
+      "learning_rate": 2.666666666666667e-07,
+      "loss": 0.2888,
+      "step": 192
+    },
+    {
+      "epoch": 8.041666666666666,
+      "grad_norm": 1.3920190334320068,
+      "learning_rate": 2.6805555555555557e-07,
+      "loss": 0.2566,
+      "step": 193
+    },
+    {
+      "epoch": 8.083333333333334,
+      "grad_norm": 1.6845390796661377,
+      "learning_rate": 2.6944444444444447e-07,
+      "loss": 0.3016,
+      "step": 194
+    },
+    {
+      "epoch": 8.125,
+      "grad_norm": 1.2597659826278687,
+      "learning_rate": 2.7083333333333337e-07,
+      "loss": 0.267,
+      "step": 195
+    },
+    {
+      "epoch": 8.166666666666666,
+      "grad_norm": 1.3180322647094727,
+      "learning_rate": 2.722222222222222e-07,
+      "loss": 0.2509,
+      "step": 196
+    },
+    {
+      "epoch": 8.208333333333334,
+      "grad_norm": 1.2117564678192139,
+      "learning_rate": 2.736111111111111e-07,
+      "loss": 0.3001,
+      "step": 197
+    },
+    {
+      "epoch": 8.25,
+      "grad_norm": 1.388692855834961,
+      "learning_rate": 2.75e-07,
+      "loss": 0.2575,
+      "step": 198
+    },
+    {
+      "epoch": 8.291666666666666,
+      "grad_norm": 1.2022684812545776,
+      "learning_rate": 2.763888888888889e-07,
+      "loss": 0.2909,
+      "step": 199
+    },
+    {
+      "epoch": 8.333333333333334,
+      "grad_norm": 1.3304301500320435,
+      "learning_rate": 2.7777777777777776e-07,
+      "loss": 0.2651,
+      "step": 200
+    },
+    {
+      "epoch": 8.375,
+      "grad_norm": 1.2710360288619995,
+      "learning_rate": 2.7916666666666666e-07,
+      "loss": 0.2743,
+      "step": 201
+    },
+    {
+      "epoch": 8.416666666666666,
+      "grad_norm": 1.2704249620437622,
+      "learning_rate": 2.8055555555555556e-07,
+      "loss": 0.2821,
+      "step": 202
+    },
+    {
+      "epoch": 8.458333333333334,
+      "grad_norm": 1.227536916732788,
+      "learning_rate": 2.8194444444444446e-07,
+      "loss": 0.2665,
+      "step": 203
+    },
+    {
+      "epoch": 8.5,
+      "grad_norm": 1.2702319622039795,
+      "learning_rate": 2.8333333333333336e-07,
+      "loss": 0.2725,
+      "step": 204
+    },
+    {
+      "epoch": 8.541666666666666,
+      "grad_norm": 1.2817600965499878,
+      "learning_rate": 2.8472222222222225e-07,
+      "loss": 0.2649,
+      "step": 205
+    },
+    {
+      "epoch": 8.583333333333334,
+      "grad_norm": 1.2167285680770874,
+      "learning_rate": 2.8611111111111115e-07,
+      "loss": 0.2659,
+      "step": 206
+    },
+    {
+      "epoch": 8.625,
+      "grad_norm": 1.2920770645141602,
+      "learning_rate": 2.8750000000000005e-07,
+      "loss": 0.2504,
+      "step": 207
+    },
+    {
+      "epoch": 8.666666666666666,
+      "grad_norm": 1.5629119873046875,
+      "learning_rate": 2.888888888888889e-07,
+      "loss": 0.2682,
+      "step": 208
+    },
+    {
+      "epoch": 8.708333333333334,
+      "grad_norm": 1.179979920387268,
+      "learning_rate": 2.902777777777778e-07,
+      "loss": 0.279,
+      "step": 209
+    },
+    {
+      "epoch": 8.75,
+      "grad_norm": 1.2344728708267212,
+      "learning_rate": 2.916666666666667e-07,
+      "loss": 0.2797,
+      "step": 210
+    },
+    {
+      "epoch": 8.791666666666666,
+      "grad_norm": 1.1335073709487915,
+      "learning_rate": 2.930555555555556e-07,
+      "loss": 0.3035,
+      "step": 211
+    },
+    {
+      "epoch": 8.833333333333334,
+      "grad_norm": 1.2874823808670044,
+      "learning_rate": 2.9444444444444444e-07,
+      "loss": 0.2482,
+      "step": 212
+    },
+    {
+      "epoch": 8.875,
+      "grad_norm": 1.2381609678268433,
+      "learning_rate": 2.9583333333333334e-07,
+      "loss": 0.2589,
+      "step": 213
+    },
+    {
+      "epoch": 8.916666666666666,
+      "grad_norm": 1.1577295064926147,
+      "learning_rate": 2.9722222222222224e-07,
+      "loss": 0.2802,
+      "step": 214
+    },
+    {
+      "epoch": 8.958333333333334,
+      "grad_norm": 1.2495061159133911,
+      "learning_rate": 2.9861111111111114e-07,
+      "loss": 0.2745,
+      "step": 215
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.2668520212173462,
+      "learning_rate": 3.0000000000000004e-07,
+      "loss": 0.2465,
+      "step": 216
+    },
+    {
+      "epoch": 9.041666666666666,
+      "grad_norm": 1.2854626178741455,
+      "learning_rate": 3.013888888888889e-07,
+      "loss": 0.2463,
+      "step": 217
+    },
+    {
+      "epoch": 9.083333333333334,
+      "grad_norm": 1.2555216550827026,
+      "learning_rate": 3.027777777777778e-07,
+      "loss": 0.2449,
+      "step": 218
+    },
+    {
+      "epoch": 9.125,
+      "grad_norm": 1.280135989189148,
+      "learning_rate": 3.041666666666667e-07,
+      "loss": 0.2418,
+      "step": 219
+    },
+    {
+      "epoch": 9.166666666666666,
+      "grad_norm": 1.204350233078003,
+      "learning_rate": 3.055555555555556e-07,
+      "loss": 0.2632,
+      "step": 220
+    },
+    {
+      "epoch": 9.208333333333334,
+      "grad_norm": 1.1124131679534912,
+      "learning_rate": 3.069444444444445e-07,
+      "loss": 0.2812,
+      "step": 221
+    },
+    {
+      "epoch": 9.25,
+      "grad_norm": 1.152017593383789,
+      "learning_rate": 3.083333333333334e-07,
+      "loss": 0.2539,
+      "step": 222
+    },
+    {
+      "epoch": 9.291666666666666,
+      "grad_norm": 1.220454454421997,
+      "learning_rate": 3.097222222222223e-07,
+      "loss": 0.258,
+      "step": 223
+    },
+    {
+      "epoch": 9.333333333333334,
+      "grad_norm": 1.173936128616333,
+      "learning_rate": 3.111111111111111e-07,
+      "loss": 0.261,
+      "step": 224
+    },
+    {
+      "epoch": 9.375,
+      "grad_norm": 1.088989496231079,
+      "learning_rate": 3.125e-07,
+      "loss": 0.3008,
+      "step": 225
+    },
+    {
+      "epoch": 9.416666666666666,
+      "grad_norm": 1.1486353874206543,
+      "learning_rate": 3.138888888888889e-07,
+      "loss": 0.279,
+      "step": 226
+    },
+    {
+      "epoch": 9.458333333333334,
+      "grad_norm": 1.3387539386749268,
+      "learning_rate": 3.152777777777778e-07,
+      "loss": 0.2492,
+      "step": 227
+    },
+    {
+      "epoch": 9.5,
+      "grad_norm": 1.1536083221435547,
+      "learning_rate": 3.166666666666667e-07,
+      "loss": 0.2604,
+      "step": 228
+    },
+    {
+      "epoch": 9.541666666666666,
+      "grad_norm": 1.3351128101348877,
+      "learning_rate": 3.1805555555555557e-07,
+      "loss": 0.2448,
+      "step": 229
+    },
+    {
+      "epoch": 9.583333333333334,
+      "grad_norm": 1.1133586168289185,
+      "learning_rate": 3.1944444444444447e-07,
+      "loss": 0.2646,
+      "step": 230
+    },
+    {
+      "epoch": 9.625,
+      "grad_norm": 1.1756969690322876,
+      "learning_rate": 3.2083333333333337e-07,
+      "loss": 0.2601,
+      "step": 231
+    },
+    {
+      "epoch": 9.666666666666666,
+      "grad_norm": 1.092045783996582,
+      "learning_rate": 3.2222222222222227e-07,
+      "loss": 0.2747,
+      "step": 232
+    },
+    {
+      "epoch": 9.708333333333334,
+      "grad_norm": 1.0904310941696167,
+      "learning_rate": 3.2361111111111117e-07,
+      "loss": 0.2629,
+      "step": 233
+    },
+    {
+      "epoch": 9.75,
+      "grad_norm": 1.1380146741867065,
+      "learning_rate": 3.25e-07,
+      "loss": 0.2616,
+      "step": 234
+    },
+    {
+      "epoch": 9.791666666666666,
+      "grad_norm": 1.4132006168365479,
+      "learning_rate": 3.263888888888889e-07,
+      "loss": 0.2243,
+      "step": 235
+    },
+    {
+      "epoch": 9.833333333333334,
+      "grad_norm": 1.2515556812286377,
+      "learning_rate": 3.277777777777778e-07,
+      "loss": 0.2581,
+      "step": 236
+    },
+    {
+      "epoch": 9.875,
+      "grad_norm": 1.1147801876068115,
+      "learning_rate": 3.2916666666666666e-07,
+      "loss": 0.2733,
+      "step": 237
+    },
+    {
+      "epoch": 9.916666666666666,
+      "grad_norm": 1.2185537815093994,
+      "learning_rate": 3.3055555555555556e-07,
+      "loss": 0.2525,
+      "step": 238
+    },
+    {
+      "epoch": 9.958333333333334,
+      "grad_norm": 1.2508174180984497,
+      "learning_rate": 3.3194444444444445e-07,
+      "loss": 0.2491,
+      "step": 239
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.1197481155395508,
+      "learning_rate": 3.3333333333333335e-07,
+      "loss": 0.274,
+      "step": 240
+    },
+    {
+      "epoch": 10.041666666666666,
+      "grad_norm": 1.1348915100097656,
+      "learning_rate": 3.3472222222222225e-07,
+      "loss": 0.2534,
+      "step": 241
+    },
+    {
+      "epoch": 10.083333333333334,
+      "grad_norm": 1.062381625175476,
+      "learning_rate": 3.3611111111111115e-07,
+      "loss": 0.2616,
+      "step": 242
+    },
+    {
+      "epoch": 10.125,
+      "grad_norm": 1.2434192895889282,
+      "learning_rate": 3.3750000000000005e-07,
+      "loss": 0.2281,
+      "step": 243
+    },
+    {
+      "epoch": 10.166666666666666,
+      "grad_norm": 1.2125924825668335,
+      "learning_rate": 3.3888888888888895e-07,
+      "loss": 0.2286,
+      "step": 244
+    },
+    {
+      "epoch": 10.208333333333334,
+      "grad_norm": 1.1648585796356201,
+      "learning_rate": 3.402777777777778e-07,
+      "loss": 0.2435,
+      "step": 245
+    },
+    {
+      "epoch": 10.25,
+      "grad_norm": 1.1663062572479248,
+      "learning_rate": 3.416666666666667e-07,
+      "loss": 0.2459,
+      "step": 246
+    },
+    {
+      "epoch": 10.291666666666666,
+      "grad_norm": 1.1701024770736694,
+      "learning_rate": 3.430555555555556e-07,
+      "loss": 0.2427,
+      "step": 247
+    },
+    {
+      "epoch": 10.333333333333334,
+      "grad_norm": 1.3048993349075317,
+      "learning_rate": 3.444444444444445e-07,
+      "loss": 0.2352,
+      "step": 248
+    },
+    {
+      "epoch": 10.375,
+      "grad_norm": 1.172383427619934,
+      "learning_rate": 3.458333333333334e-07,
+      "loss": 0.2287,
+      "step": 249
+    },
+    {
+      "epoch": 10.416666666666666,
+      "grad_norm": 1.178991675376892,
+      "learning_rate": 3.472222222222223e-07,
+      "loss": 0.2625,
+      "step": 250
+    },
+    {
+      "epoch": 10.458333333333334,
+      "grad_norm": 1.0647447109222412,
+      "learning_rate": 3.4861111111111114e-07,
+      "loss": 0.2552,
+      "step": 251
+    },
+    {
+      "epoch": 10.5,
+      "grad_norm": 1.1307735443115234,
+      "learning_rate": 3.5000000000000004e-07,
+      "loss": 0.2445,
+      "step": 252
+    },
+    {
+      "epoch": 10.541666666666666,
+      "grad_norm": 1.06700599193573,
+      "learning_rate": 3.513888888888889e-07,
+      "loss": 0.274,
+      "step": 253
+    },
+    {
+      "epoch": 10.583333333333334,
+      "grad_norm": 1.1176058053970337,
+      "learning_rate": 3.527777777777778e-07,
+      "loss": 0.2633,
+      "step": 254
+    },
+    {
+      "epoch": 10.625,
+      "grad_norm": 1.190774917602539,
+      "learning_rate": 3.541666666666667e-07,
+      "loss": 0.2472,
+      "step": 255
+    },
+    {
+      "epoch": 10.666666666666666,
+      "grad_norm": 1.3855525255203247,
+      "learning_rate": 3.555555555555556e-07,
+      "loss": 0.2198,
+      "step": 256
+    },
+    {
+      "epoch": 10.708333333333334,
+      "grad_norm": 1.1426787376403809,
+      "learning_rate": 3.569444444444445e-07,
+      "loss": 0.2665,
+      "step": 257
+    },
+    {
+      "epoch": 10.75,
+      "grad_norm": 1.0948460102081299,
+      "learning_rate": 3.583333333333334e-07,
+      "loss": 0.2574,
+      "step": 258
+    },
+    {
+      "epoch": 10.791666666666666,
+      "grad_norm": 1.1022791862487793,
+      "learning_rate": 3.597222222222223e-07,
+      "loss": 0.2357,
+      "step": 259
+    },
+    {
+      "epoch": 10.833333333333334,
+      "grad_norm": 1.103832483291626,
+      "learning_rate": 3.611111111111111e-07,
+      "loss": 0.2636,
+      "step": 260
+    },
+    {
+      "epoch": 10.875,
+      "grad_norm": 1.1584415435791016,
+      "learning_rate": 3.625e-07,
+      "loss": 0.2447,
+      "step": 261
+    },
+    {
+      "epoch": 10.916666666666666,
+      "grad_norm": 1.1745778322219849,
+      "learning_rate": 3.638888888888889e-07,
+      "loss": 0.245,
+      "step": 262
+    },
+    {
+      "epoch": 10.958333333333334,
+      "grad_norm": 1.1118824481964111,
+      "learning_rate": 3.652777777777778e-07,
+      "loss": 0.252,
+      "step": 263
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.1371052265167236,
+      "learning_rate": 3.666666666666667e-07,
+      "loss": 0.2473,
+      "step": 264
+    },
+    {
+      "epoch": 11.041666666666666,
+      "grad_norm": 1.0493159294128418,
+      "learning_rate": 3.680555555555556e-07,
+      "loss": 0.2445,
+      "step": 265
+    },
+    {
+      "epoch": 11.083333333333334,
+      "grad_norm": 1.0633114576339722,
+      "learning_rate": 3.694444444444445e-07,
+      "loss": 0.2493,
+      "step": 266
+    },
+    {
+      "epoch": 11.125,
+      "grad_norm": 1.1716960668563843,
+      "learning_rate": 3.708333333333334e-07,
+      "loss": 0.228,
+      "step": 267
+    },
+    {
+      "epoch": 11.166666666666666,
+      "grad_norm": 0.946899950504303,
+      "learning_rate": 3.722222222222222e-07,
+      "loss": 0.2517,
+      "step": 268
+    },
+    {
+      "epoch": 11.208333333333334,
+      "grad_norm": 1.290504813194275,
+      "learning_rate": 3.736111111111111e-07,
+      "loss": 0.2079,
+      "step": 269
+    },
+    {
+      "epoch": 11.25,
+      "grad_norm": 1.0798704624176025,
+      "learning_rate": 3.75e-07,
+      "loss": 0.2327,
+      "step": 270
+    },
+    {
+      "epoch": 11.291666666666666,
+      "grad_norm": 1.0250619649887085,
+      "learning_rate": 3.763888888888889e-07,
+      "loss": 0.2363,
+      "step": 271
+    },
+    {
+      "epoch": 11.333333333333334,
+      "grad_norm": 1.0896459817886353,
+      "learning_rate": 3.777777777777778e-07,
+      "loss": 0.2165,
+      "step": 272
+    },
+    {
+      "epoch": 11.375,
+      "grad_norm": 0.9856969118118286,
+      "learning_rate": 3.791666666666667e-07,
+      "loss": 0.249,
+      "step": 273
+    },
+    {
+      "epoch": 11.416666666666666,
+      "grad_norm": 1.1642574071884155,
+      "learning_rate": 3.805555555555556e-07,
+      "loss": 0.2341,
+      "step": 274
+    },
+    {
+      "epoch": 11.458333333333334,
+      "grad_norm": 1.18202805519104,
+      "learning_rate": 3.819444444444445e-07,
+      "loss": 0.2282,
+      "step": 275
+    },
+    {
+      "epoch": 11.5,
+      "grad_norm": 1.0939830541610718,
+      "learning_rate": 3.8333333333333335e-07,
+      "loss": 0.2338,
+      "step": 276
+    },
+    {
+      "epoch": 11.541666666666666,
+      "grad_norm": 1.2663522958755493,
+      "learning_rate": 3.8472222222222225e-07,
+      "loss": 0.2124,
+      "step": 277
+    },
+    {
+      "epoch": 11.583333333333334,
+      "grad_norm": 1.044758915901184,
+      "learning_rate": 3.8611111111111115e-07,
+      "loss": 0.2633,
+      "step": 278
+    },
+    {
+      "epoch": 11.625,
+      "grad_norm": 1.1343523263931274,
+      "learning_rate": 3.8750000000000005e-07,
+      "loss": 0.2326,
+      "step": 279
+    },
+    {
+      "epoch": 11.666666666666666,
+      "grad_norm": 1.118573784828186,
+      "learning_rate": 3.8888888888888895e-07,
+      "loss": 0.2239,
+      "step": 280
+    },
+    {
+      "epoch": 11.708333333333334,
+      "grad_norm": 1.0948375463485718,
+      "learning_rate": 3.9027777777777785e-07,
+      "loss": 0.2348,
+      "step": 281
+    },
+    {
+      "epoch": 11.75,
+      "grad_norm": 1.2570106983184814,
+      "learning_rate": 3.9166666666666675e-07,
+      "loss": 0.2156,
+      "step": 282
+    },
+    {
+      "epoch": 11.791666666666666,
+      "grad_norm": 1.1027518510818481,
+      "learning_rate": 3.9305555555555565e-07,
+      "loss": 0.2299,
+      "step": 283
+    },
+    {
+      "epoch": 11.833333333333334,
+      "grad_norm": 1.1455954313278198,
+      "learning_rate": 3.9444444444444444e-07,
+      "loss": 0.2221,
+      "step": 284
+    },
+    {
+      "epoch": 11.875,
+      "grad_norm": 1.2163667678833008,
+      "learning_rate": 3.9583333333333334e-07,
+      "loss": 0.2296,
+      "step": 285
+    },
+    {
+      "epoch": 11.916666666666666,
+      "grad_norm": 1.256474494934082,
+      "learning_rate": 3.9722222222222224e-07,
+      "loss": 0.2466,
+      "step": 286
+    },
+    {
+      "epoch": 11.958333333333334,
+      "grad_norm": 1.1235169172286987,
+      "learning_rate": 3.9861111111111114e-07,
+      "loss": 0.2366,
+      "step": 287
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.0120381116867065,
+      "learning_rate": 4.0000000000000003e-07,
+      "loss": 0.2371,
+      "step": 288
+    },
+    {
+      "epoch": 12.041666666666666,
+      "grad_norm": 1.3755816221237183,
+      "learning_rate": 4.0138888888888893e-07,
+      "loss": 0.2154,
+      "step": 289
+    },
+    {
+      "epoch": 12.083333333333334,
+      "grad_norm": 0.9584789872169495,
+      "learning_rate": 4.0277777777777783e-07,
+      "loss": 0.2558,
+      "step": 290
+    },
+    {
+      "epoch": 12.125,
+      "grad_norm": 1.0496126413345337,
+      "learning_rate": 4.041666666666667e-07,
+      "loss": 0.2369,
+      "step": 291
+    },
+    {
+      "epoch": 12.166666666666666,
+      "grad_norm": 1.0518829822540283,
+      "learning_rate": 4.055555555555556e-07,
+      "loss": 0.2237,
+      "step": 292
+    },
+    {
+      "epoch": 12.208333333333334,
+      "grad_norm": 1.0739285945892334,
+      "learning_rate": 4.069444444444445e-07,
+      "loss": 0.2022,
+      "step": 293
+    },
+    {
+      "epoch": 12.25,
+      "grad_norm": 1.0657017230987549,
+      "learning_rate": 4.083333333333334e-07,
+      "loss": 0.2204,
+      "step": 294
+    },
+    {
+      "epoch": 12.291666666666666,
+      "grad_norm": 0.9074841737747192,
+      "learning_rate": 4.097222222222223e-07,
+      "loss": 0.2504,
+      "step": 295
+    },
+    {
+      "epoch": 12.333333333333334,
+      "grad_norm": 1.385315179824829,
+      "learning_rate": 4.111111111111112e-07,
+      "loss": 0.2247,
+      "step": 296
+    },
+    {
+      "epoch": 12.375,
+      "grad_norm": 1.1710454225540161,
+      "learning_rate": 4.125000000000001e-07,
+      "loss": 0.2183,
+      "step": 297
+    },
+    {
+      "epoch": 12.416666666666666,
+      "grad_norm": 1.3828837871551514,
+      "learning_rate": 4.1388888888888897e-07,
+      "loss": 0.2176,
+      "step": 298
+    },
+    {
+      "epoch": 12.458333333333334,
+      "grad_norm": 1.058730959892273,
+      "learning_rate": 4.1527777777777777e-07,
+      "loss": 0.2343,
+      "step": 299
+    },
+    {
+      "epoch": 12.5,
+      "grad_norm": 0.9766915440559387,
+      "learning_rate": 4.1666666666666667e-07,
+      "loss": 0.2457,
+      "step": 300
+    },
+    {
+      "epoch": 12.541666666666666,
+      "grad_norm": 1.1404693126678467,
+      "learning_rate": 4.1805555555555556e-07,
+      "loss": 0.1953,
+      "step": 301
+    },
+    {
+      "epoch": 12.583333333333334,
+      "grad_norm": 1.1021593809127808,
+      "learning_rate": 4.1944444444444446e-07,
+      "loss": 0.1904,
+      "step": 302
+    },
+    {
+      "epoch": 12.625,
+      "grad_norm": 1.2560713291168213,
+      "learning_rate": 4.2083333333333336e-07,
+      "loss": 0.207,
+      "step": 303
+    },
+    {
+      "epoch": 12.666666666666666,
+      "grad_norm": 1.1121106147766113,
+      "learning_rate": 4.2222222222222226e-07,
+      "loss": 0.1944,
+      "step": 304
+    },
+    {
+      "epoch": 12.708333333333334,
+      "grad_norm": 1.1668109893798828,
+      "learning_rate": 4.2361111111111116e-07,
+      "loss": 0.1974,
+      "step": 305
+    },
+    {
+      "epoch": 12.75,
+      "grad_norm": 1.1621122360229492,
+      "learning_rate": 4.2500000000000006e-07,
+      "loss": 0.1972,
+      "step": 306
+    },
+    {
+      "epoch": 12.791666666666666,
+      "grad_norm": 1.0159415006637573,
+      "learning_rate": 4.263888888888889e-07,
+      "loss": 0.2218,
+      "step": 307
+    },
+    {
+      "epoch": 12.833333333333334,
+      "grad_norm": 1.2106002569198608,
+      "learning_rate": 4.277777777777778e-07,
+      "loss": 0.2199,
+      "step": 308
+    },
+    {
+      "epoch": 12.875,
+      "grad_norm": 1.0972415208816528,
+      "learning_rate": 4.291666666666667e-07,
+      "loss": 0.223,
+      "step": 309
+    },
+    {
+      "epoch": 12.916666666666666,
+      "grad_norm": 1.1199716329574585,
+      "learning_rate": 4.305555555555556e-07,
+      "loss": 0.2112,
+      "step": 310
+    },
+    {
+      "epoch": 12.958333333333334,
+      "grad_norm": 1.1161723136901855,
+      "learning_rate": 4.319444444444445e-07,
+      "loss": 0.2118,
+      "step": 311
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.1183927059173584,
+      "learning_rate": 4.333333333333334e-07,
+      "loss": 0.2141,
+      "step": 312
+    },
+    {
+      "epoch": 13.041666666666666,
+      "grad_norm": 1.0980181694030762,
+      "learning_rate": 4.347222222222223e-07,
+      "loss": 0.207,
+      "step": 313
+    },
+    {
+      "epoch": 13.083333333333334,
+      "grad_norm": 1.114234447479248,
+      "learning_rate": 4.361111111111112e-07,
+      "loss": 0.1935,
+      "step": 314
+    },
+    {
+      "epoch": 13.125,
+      "grad_norm": 1.0884588956832886,
+      "learning_rate": 4.375e-07,
+      "loss": 0.1912,
+      "step": 315
+    },
+    {
+      "epoch": 13.166666666666666,
+      "grad_norm": 1.2149794101715088,
+      "learning_rate": 4.388888888888889e-07,
+      "loss": 0.2052,
+      "step": 316
+    },
+    {
+      "epoch": 13.208333333333334,
+      "grad_norm": 1.0066922903060913,
+      "learning_rate": 4.402777777777778e-07,
+      "loss": 0.2085,
+      "step": 317
+    },
+    {
+      "epoch": 13.25,
+      "grad_norm": 1.297834873199463,
+      "learning_rate": 4.416666666666667e-07,
+      "loss": 0.1785,
+      "step": 318
+    },
+    {
+      "epoch": 13.291666666666666,
+      "grad_norm": 1.0707037448883057,
+      "learning_rate": 4.430555555555556e-07,
+      "loss": 0.1867,
+      "step": 319
+    },
+    {
+      "epoch": 13.333333333333334,
+      "grad_norm": 0.994250476360321,
+      "learning_rate": 4.444444444444445e-07,
+      "loss": 0.2001,
+      "step": 320
+    },
+    {
+      "epoch": 13.375,
+      "grad_norm": 0.9893432259559631,
+      "learning_rate": 4.458333333333334e-07,
+      "loss": 0.2034,
+      "step": 321
+    },
+    {
+      "epoch": 13.416666666666666,
+      "grad_norm": 1.1544864177703857,
+      "learning_rate": 4.4722222222222223e-07,
+      "loss": 0.229,
+      "step": 322
+    },
+    {
+      "epoch": 13.458333333333334,
+      "grad_norm": 0.9520488381385803,
+      "learning_rate": 4.4861111111111113e-07,
+      "loss": 0.2176,
+      "step": 323
+    },
+    {
+      "epoch": 13.5,
+      "grad_norm": 1.2035622596740723,
+      "learning_rate": 4.5000000000000003e-07,
+      "loss": 0.192,
+      "step": 324
+    },
+    {
+      "epoch": 13.541666666666666,
+      "grad_norm": 1.0155614614486694,
+      "learning_rate": 4.5138888888888893e-07,
+      "loss": 0.2317,
+      "step": 325
+    },
+    {
+      "epoch": 13.583333333333334,
+      "grad_norm": 1.1402587890625,
+      "learning_rate": 4.5277777777777783e-07,
+      "loss": 0.1964,
+      "step": 326
+    },
+    {
+      "epoch": 13.625,
+      "grad_norm": 1.0315797328948975,
+      "learning_rate": 4.5416666666666673e-07,
+      "loss": 0.1976,
+      "step": 327
+    },
+    {
+      "epoch": 13.666666666666666,
+      "grad_norm": 0.949512779712677,
+      "learning_rate": 4.5555555555555563e-07,
+      "loss": 0.1981,
+      "step": 328
+    },
+    {
+      "epoch": 13.708333333333334,
+      "grad_norm": 1.00216543674469,
+      "learning_rate": 4.5694444444444453e-07,
+      "loss": 0.2125,
+      "step": 329
+    },
+    {
+      "epoch": 13.75,
+      "grad_norm": 0.9637646675109863,
+      "learning_rate": 4.583333333333333e-07,
+      "loss": 0.1978,
+      "step": 330
+    },
+    {
+      "epoch": 13.791666666666666,
+      "grad_norm": 1.009885549545288,
+      "learning_rate": 4.597222222222222e-07,
+      "loss": 0.1988,
+      "step": 331
+    },
+    {
+      "epoch": 13.833333333333334,
+      "grad_norm": 0.9766483306884766,
+      "learning_rate": 4.611111111111111e-07,
+      "loss": 0.232,
+      "step": 332
+    },
+    {
+      "epoch": 13.875,
+      "grad_norm": 0.9823787808418274,
+      "learning_rate": 4.625e-07,
+      "loss": 0.2032,
+      "step": 333
+    },
+    {
+      "epoch": 13.916666666666666,
+      "grad_norm": 1.3104783296585083,
+      "learning_rate": 4.638888888888889e-07,
+      "loss": 0.1751,
+      "step": 334
+    },
+    {
+      "epoch": 13.958333333333334,
+      "grad_norm": 1.0261071920394897,
+      "learning_rate": 4.652777777777778e-07,
+      "loss": 0.2084,
+      "step": 335
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.0240532159805298,
+      "learning_rate": 4.666666666666667e-07,
+      "loss": 0.2093,
+      "step": 336
+    },
+    {
+      "epoch": 14.041666666666666,
+      "grad_norm": 0.9477185010910034,
+      "learning_rate": 4.680555555555556e-07,
+      "loss": 0.2109,
+      "step": 337
+    },
+    {
+      "epoch": 14.083333333333334,
+      "grad_norm": 0.9221709370613098,
+      "learning_rate": 4.6944444444444446e-07,
+      "loss": 0.2134,
+      "step": 338
+    },
+    {
+      "epoch": 14.125,
+      "grad_norm": 0.9197400212287903,
+      "learning_rate": 4.7083333333333336e-07,
+      "loss": 0.2093,
+      "step": 339
+    },
+    {
+      "epoch": 14.166666666666666,
+      "grad_norm": 0.9760138392448425,
+      "learning_rate": 4.7222222222222226e-07,
+      "loss": 0.204,
+      "step": 340
+    },
+    {
+      "epoch": 14.208333333333334,
+      "grad_norm": 0.9252342581748962,
+      "learning_rate": 4.7361111111111116e-07,
+      "loss": 0.208,
+      "step": 341
+    },
+    {
+      "epoch": 14.25,
+      "grad_norm": 1.112654447555542,
+      "learning_rate": 4.7500000000000006e-07,
+      "loss": 0.1693,
+      "step": 342
+    },
+    {
+      "epoch": 14.291666666666666,
+      "grad_norm": 0.9246732592582703,
+      "learning_rate": 4.7638888888888896e-07,
+      "loss": 0.2012,
+      "step": 343
+    },
+    {
+      "epoch": 14.333333333333334,
+      "grad_norm": 0.9642884731292725,
+      "learning_rate": 4.777777777777778e-07,
+      "loss": 0.2147,
+      "step": 344
+    },
+    {
+      "epoch": 14.375,
+      "grad_norm": 0.9451230764389038,
+      "learning_rate": 4.791666666666667e-07,
+      "loss": 0.1753,
+      "step": 345
+    },
+    {
+      "epoch": 14.416666666666666,
+      "grad_norm": 1.1428673267364502,
+      "learning_rate": 4.805555555555556e-07,
+      "loss": 0.1697,
+      "step": 346
+    },
+    {
+      "epoch": 14.458333333333334,
+      "grad_norm": 0.9287844300270081,
+      "learning_rate": 4.819444444444445e-07,
+      "loss": 0.2142,
+      "step": 347
+    },
+    {
+      "epoch": 14.5,
+      "grad_norm": 1.2430188655853271,
+      "learning_rate": 4.833333333333334e-07,
+      "loss": 0.1492,
+      "step": 348
+    },
+    {
+      "epoch": 14.541666666666666,
+      "grad_norm": 1.0802409648895264,
+      "learning_rate": 4.847222222222223e-07,
+      "loss": 0.1885,
+      "step": 349
+    },
+    {
+      "epoch": 14.583333333333334,
+      "grad_norm": 1.0437086820602417,
+      "learning_rate": 4.861111111111112e-07,
+      "loss": 0.1878,
+      "step": 350
+    },
+    {
+      "epoch": 14.625,
+      "grad_norm": 1.1413215398788452,
+      "learning_rate": 4.875000000000001e-07,
+      "loss": 0.1616,
+      "step": 351
+    },
+    {
+      "epoch": 14.666666666666666,
+      "grad_norm": 1.067886471748352,
+      "learning_rate": 4.88888888888889e-07,
+      "loss": 0.1827,
+      "step": 352
+    },
+    {
+      "epoch": 14.708333333333334,
+      "grad_norm": 1.0765631198883057,
+      "learning_rate": 4.902777777777779e-07,
+      "loss": 0.1581,
+      "step": 353
+    },
+    {
+      "epoch": 14.75,
+      "grad_norm": 1.1207886934280396,
+      "learning_rate": 4.916666666666667e-07,
+      "loss": 0.1741,
+      "step": 354
+    },
+    {
+      "epoch": 14.791666666666666,
+      "grad_norm": 0.9442469477653503,
+      "learning_rate": 4.930555555555556e-07,
+      "loss": 0.2154,
+      "step": 355
+    },
+    {
+      "epoch": 14.833333333333334,
+      "grad_norm": 1.0349509716033936,
+      "learning_rate": 4.944444444444445e-07,
+      "loss": 0.1747,
+      "step": 356
+    },
+    {
+      "epoch": 14.875,
+      "grad_norm": 1.06031334400177,
+      "learning_rate": 4.958333333333334e-07,
+      "loss": 0.1713,
+      "step": 357
+    },
+    {
+      "epoch": 14.916666666666666,
+      "grad_norm": 1.0730364322662354,
+      "learning_rate": 4.972222222222223e-07,
+      "loss": 0.1748,
+      "step": 358
+    },
+    {
+      "epoch": 14.958333333333334,
+      "grad_norm": 1.0952244997024536,
+      "learning_rate": 4.986111111111112e-07,
+      "loss": 0.1967,
+      "step": 359
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.0405199527740479,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 0.1798,
+      "step": 360
+    },
+    {
+      "epoch": 15.041666666666666,
+      "grad_norm": 0.9302091598510742,
+      "learning_rate": 5.013888888888889e-07,
+      "loss": 0.1822,
+      "step": 361
+    },
+    {
+      "epoch": 15.083333333333334,
+      "grad_norm": 0.9061443209648132,
+      "learning_rate": 5.027777777777778e-07,
+      "loss": 0.1979,
+      "step": 362
+    },
+    {
+      "epoch": 15.125,
+      "grad_norm": 1.0499804019927979,
+      "learning_rate": 5.041666666666667e-07,
+      "loss": 0.1784,
+      "step": 363
+    },
+    {
+      "epoch": 15.166666666666666,
+      "grad_norm": 1.130347728729248,
+      "learning_rate": 5.055555555555556e-07,
+      "loss": 0.1226,
+      "step": 364
+    },
+    {
+      "epoch": 15.208333333333334,
+      "grad_norm": 1.0812504291534424,
+      "learning_rate": 5.069444444444445e-07,
+      "loss": 0.1512,
+      "step": 365
+    },
+    {
+      "epoch": 15.25,
+      "grad_norm": 1.0023159980773926,
+      "learning_rate": 5.083333333333334e-07,
+      "loss": 0.1741,
+      "step": 366
+    },
+    {
+      "epoch": 15.291666666666666,
+      "grad_norm": 1.1736576557159424,
+      "learning_rate": 5.097222222222223e-07,
+      "loss": 0.1464,
+      "step": 367
+    },
+    {
+      "epoch": 15.333333333333334,
+      "grad_norm": 1.1097235679626465,
+      "learning_rate": 5.111111111111112e-07,
+      "loss": 0.1592,
+      "step": 368
+    },
+    {
+      "epoch": 15.375,
+      "grad_norm": 1.0511345863342285,
+      "learning_rate": 5.125e-07,
+      "loss": 0.1677,
+      "step": 369
+    },
+    {
+      "epoch": 15.416666666666666,
+      "grad_norm": 1.037803053855896,
+      "learning_rate": 5.138888888888889e-07,
+      "loss": 0.1686,
+      "step": 370
+    },
+    {
+      "epoch": 15.458333333333334,
+      "grad_norm": 0.9850645661354065,
+      "learning_rate": 5.152777777777778e-07,
+      "loss": 0.1805,
+      "step": 371
+    },
+    {
+      "epoch": 15.5,
+      "grad_norm": 1.0566681623458862,
+      "learning_rate": 5.166666666666667e-07,
+      "loss": 0.1497,
+      "step": 372
+    },
+    {
+      "epoch": 15.541666666666666,
+      "grad_norm": 0.883122980594635,
+      "learning_rate": 5.180555555555556e-07,
+      "loss": 0.1918,
+      "step": 373
+    },
+    {
+      "epoch": 15.583333333333334,
+      "grad_norm": 1.0410674810409546,
+      "learning_rate": 5.194444444444445e-07,
+      "loss": 0.1802,
+      "step": 374
+    },
+    {
+      "epoch": 15.625,
+      "grad_norm": 1.7478364706039429,
+      "learning_rate": 5.208333333333334e-07,
+      "loss": 0.1892,
+      "step": 375
+    },
+    {
+      "epoch": 15.666666666666666,
+      "grad_norm": 1.07477867603302,
+      "learning_rate": 5.222222222222223e-07,
+      "loss": 0.1587,
+      "step": 376
+    },
+    {
+      "epoch": 15.708333333333334,
+      "grad_norm": 1.0230439901351929,
+      "learning_rate": 5.236111111111112e-07,
+      "loss": 0.1649,
+      "step": 377
+    },
+    {
+      "epoch": 15.75,
+      "grad_norm": 1.0138881206512451,
+      "learning_rate": 5.250000000000001e-07,
+      "loss": 0.1505,
+      "step": 378
+    },
+    {
+      "epoch": 15.791666666666666,
+      "grad_norm": 0.9828330278396606,
+      "learning_rate": 5.26388888888889e-07,
+      "loss": 0.1496,
+      "step": 379
+    },
+    {
+      "epoch": 15.833333333333334,
+      "grad_norm": 0.9980834126472473,
+      "learning_rate": 5.277777777777779e-07,
+      "loss": 0.1653,
+      "step": 380
+    },
+    {
+      "epoch": 15.875,
+      "grad_norm": 0.9464366436004639,
+      "learning_rate": 5.291666666666668e-07,
+      "loss": 0.2077,
+      "step": 381
+    },
+    {
+      "epoch": 15.916666666666666,
+      "grad_norm": 0.9358134865760803,
+      "learning_rate": 5.305555555555557e-07,
+      "loss": 0.1702,
+      "step": 382
+    },
+    {
+      "epoch": 15.958333333333334,
+      "grad_norm": 0.9587668180465698,
+      "learning_rate": 5.319444444444446e-07,
+      "loss": 0.2003,
+      "step": 383
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.039010763168335,
+      "learning_rate": 5.333333333333335e-07,
+      "loss": 0.1737,
+      "step": 384
+    },
+    {
+      "epoch": 16.041666666666668,
+      "grad_norm": 1.0073297023773193,
+      "learning_rate": 5.347222222222222e-07,
+      "loss": 0.1417,
+      "step": 385
+    },
+    {
+      "epoch": 16.083333333333332,
+      "grad_norm": 0.9441164135932922,
+      "learning_rate": 5.361111111111111e-07,
+      "loss": 0.1643,
+      "step": 386
+    },
+    {
+      "epoch": 16.125,
+      "grad_norm": 0.9800100922584534,
+      "learning_rate": 5.375e-07,
+      "loss": 0.1701,
+      "step": 387
+    },
+    {
+      "epoch": 16.166666666666668,
+      "grad_norm": 0.9619889855384827,
+      "learning_rate": 5.388888888888889e-07,
+      "loss": 0.1726,
+      "step": 388
+    },
+    {
+      "epoch": 16.208333333333332,
+      "grad_norm": 0.9608114957809448,
+      "learning_rate": 5.402777777777778e-07,
+      "loss": 0.1671,
+      "step": 389
+    },
+    {
+      "epoch": 16.25,
+      "grad_norm": 0.975229799747467,
+      "learning_rate": 5.416666666666667e-07,
+      "loss": 0.1334,
+      "step": 390
+    },
+    {
+      "epoch": 16.291666666666668,
+      "grad_norm": 0.9500177502632141,
+      "learning_rate": 5.430555555555556e-07,
+      "loss": 0.1641,
+      "step": 391
+    },
+    {
+      "epoch": 16.333333333333332,
+      "grad_norm": 1.0363067388534546,
+      "learning_rate": 5.444444444444444e-07,
+      "loss": 0.133,
+      "step": 392
+    },
+    {
+      "epoch": 16.375,
+      "grad_norm": 1.384044885635376,
+      "learning_rate": 5.458333333333333e-07,
+      "loss": 0.1691,
+      "step": 393
+    },
+    {
+      "epoch": 16.416666666666668,
+      "grad_norm": 0.9603424668312073,
+      "learning_rate": 5.472222222222222e-07,
+      "loss": 0.161,
+      "step": 394
+    },
+    {
+      "epoch": 16.458333333333332,
+      "grad_norm": 0.9141762852668762,
+      "learning_rate": 5.486111111111111e-07,
+      "loss": 0.1725,
+      "step": 395
+    },
+    {
+      "epoch": 16.5,
+      "grad_norm": 0.9520753026008606,
+      "learning_rate": 5.5e-07,
+      "loss": 0.1519,
+      "step": 396
+    },
+    {
+      "epoch": 16.541666666666668,
+      "grad_norm": 0.9954721927642822,
+      "learning_rate": 5.513888888888889e-07,
+      "loss": 0.1682,
+      "step": 397
+    },
+    {
+      "epoch": 16.583333333333332,
+      "grad_norm": 1.077360987663269,
+      "learning_rate": 5.527777777777778e-07,
+      "loss": 0.1625,
+      "step": 398
+    },
+    {
+      "epoch": 16.625,
+      "grad_norm": 0.9265204668045044,
+      "learning_rate": 5.541666666666667e-07,
+      "loss": 0.1698,
+      "step": 399
+    },
+    {
+      "epoch": 16.666666666666668,
+      "grad_norm": 1.196536660194397,
+      "learning_rate": 5.555555555555555e-07,
+      "loss": 0.1406,
+      "step": 400
+    },
+    {
+      "epoch": 16.708333333333332,
+      "grad_norm": 0.9234820604324341,
+      "learning_rate": 5.569444444444444e-07,
+      "loss": 0.172,
+      "step": 401
+    },
+    {
+      "epoch": 16.75,
+      "grad_norm": 0.9750126600265503,
+      "learning_rate": 5.583333333333333e-07,
+      "loss": 0.158,
+      "step": 402
+    },
+    {
+      "epoch": 16.791666666666668,
+      "grad_norm": 0.9417174458503723,
+      "learning_rate": 5.597222222222222e-07,
+      "loss": 0.1549,
+      "step": 403
+    },
+    {
+      "epoch": 16.833333333333332,
+      "grad_norm": 1.0074892044067383,
+      "learning_rate": 5.611111111111111e-07,
+      "loss": 0.117,
+      "step": 404
+    },
+    {
+      "epoch": 16.875,
+      "grad_norm": 0.961407482624054,
+      "learning_rate": 5.625e-07,
+      "loss": 0.1219,
+      "step": 405
+    },
+    {
+      "epoch": 16.916666666666668,
+      "grad_norm": 0.9712786078453064,
+      "learning_rate": 5.638888888888889e-07,
+      "loss": 0.1462,
+      "step": 406
+    },
+    {
+      "epoch": 16.958333333333332,
+      "grad_norm": 0.9871298670768738,
+      "learning_rate": 5.652777777777778e-07,
+      "loss": 0.1485,
+      "step": 407
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.4290050268173218,
+      "learning_rate": 5.666666666666667e-07,
+      "loss": 0.1392,
+      "step": 408
+    },
+    {
+      "epoch": 17.041666666666668,
+      "grad_norm": 1.0211368799209595,
+      "learning_rate": 5.680555555555556e-07,
+      "loss": 0.1252,
+      "step": 409
+    },
+    {
+      "epoch": 17.083333333333332,
+      "grad_norm": 0.9359925389289856,
+      "learning_rate": 5.694444444444445e-07,
+      "loss": 0.1486,
+      "step": 410
+    },
+    {
+      "epoch": 17.125,
+      "grad_norm": 1.0732674598693848,
+      "learning_rate": 5.708333333333334e-07,
+      "loss": 0.1167,
+      "step": 411
+    },
+    {
+      "epoch": 17.166666666666668,
+      "grad_norm": 0.8616218566894531,
+      "learning_rate": 5.722222222222223e-07,
+      "loss": 0.1408,
+      "step": 412
+    },
+    {
+      "epoch": 17.208333333333332,
+      "grad_norm": 0.8566951155662537,
+      "learning_rate": 5.736111111111112e-07,
+      "loss": 0.1629,
+      "step": 413
+    },
+    {
+      "epoch": 17.25,
+      "grad_norm": 0.9342655539512634,
+      "learning_rate": 5.750000000000001e-07,
+      "loss": 0.1344,
+      "step": 414
+    },
+    {
+      "epoch": 17.291666666666668,
+      "grad_norm": 0.9016939401626587,
+      "learning_rate": 5.76388888888889e-07,
+      "loss": 0.1396,
+      "step": 415
+    },
+    {
+      "epoch": 17.333333333333332,
+      "grad_norm": 0.992168664932251,
+      "learning_rate": 5.777777777777778e-07,
+      "loss": 0.1672,
+      "step": 416
+    },
+    {
+      "epoch": 17.375,
+      "grad_norm": 0.9596002697944641,
+      "learning_rate": 5.791666666666667e-07,
+      "loss": 0.1147,
+      "step": 417
+    },
+    {
+      "epoch": 17.416666666666668,
+      "grad_norm": 1.0081878900527954,
+      "learning_rate": 5.805555555555556e-07,
+      "loss": 0.1248,
+      "step": 418
+    },
+    {
+      "epoch": 17.458333333333332,
+      "grad_norm": 1.3084962368011475,
+      "learning_rate": 5.819444444444445e-07,
+      "loss": 0.1684,
+      "step": 419
+    },
+    {
+      "epoch": 17.5,
+      "grad_norm": 1.0510737895965576,
+      "learning_rate": 5.833333333333334e-07,
+      "loss": 0.1092,
+      "step": 420
+    },
+    {
+      "epoch": 17.541666666666668,
+      "grad_norm": 0.9097137451171875,
+      "learning_rate": 5.847222222222223e-07,
+      "loss": 0.1454,
+      "step": 421
+    },
+    {
+      "epoch": 17.583333333333332,
+      "grad_norm": 1.0495959520339966,
+      "learning_rate": 5.861111111111112e-07,
+      "loss": 0.122,
+      "step": 422
+    },
+    {
+      "epoch": 17.625,
+      "grad_norm": 0.9778475165367126,
+      "learning_rate": 5.875e-07,
+      "loss": 0.1614,
+      "step": 423
+    },
+    {
+      "epoch": 17.666666666666668,
+      "grad_norm": 0.9457103610038757,
+      "learning_rate": 5.888888888888889e-07,
+      "loss": 0.1541,
+      "step": 424
+    },
+    {
+      "epoch": 17.708333333333332,
+      "grad_norm": 1.0476627349853516,
+      "learning_rate": 5.902777777777778e-07,
+      "loss": 0.1245,
+      "step": 425
+    },
+    {
+      "epoch": 17.75,
+      "grad_norm": 0.9881095290184021,
+      "learning_rate": 5.916666666666667e-07,
+      "loss": 0.129,
+      "step": 426
+    },
+    {
+      "epoch": 17.791666666666668,
+      "grad_norm": 1.0222713947296143,
+      "learning_rate": 5.930555555555556e-07,
+      "loss": 0.1197,
+      "step": 427
+    },
+    {
+      "epoch": 17.833333333333332,
+      "grad_norm": 0.917041003704071,
+      "learning_rate": 5.944444444444445e-07,
+      "loss": 0.1451,
+      "step": 428
+    },
+    {
+      "epoch": 17.875,
+      "grad_norm": 1.03107750415802,
+      "learning_rate": 5.958333333333334e-07,
+      "loss": 0.142,
+      "step": 429
+    },
+    {
+      "epoch": 17.916666666666668,
+      "grad_norm": 0.9509056210517883,
+      "learning_rate": 5.972222222222223e-07,
+      "loss": 0.1589,
+      "step": 430
+    },
+    {
+      "epoch": 17.958333333333332,
+      "grad_norm": 1.0196317434310913,
+      "learning_rate": 5.986111111111112e-07,
+      "loss": 0.1127,
+      "step": 431
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.0310330390930176,
+      "learning_rate": 6.000000000000001e-07,
+      "loss": 0.1392,
+      "step": 432
+    },
+    {
+      "epoch": 18.041666666666668,
+      "grad_norm": 0.8874520659446716,
+      "learning_rate": 6.013888888888889e-07,
+      "loss": 0.1151,
+      "step": 433
+    },
+    {
+      "epoch": 18.083333333333332,
+      "grad_norm": 1.087056040763855,
+      "learning_rate": 6.027777777777778e-07,
+      "loss": 0.1121,
+      "step": 434
+    },
+    {
+      "epoch": 18.125,
+      "grad_norm": 0.9367361068725586,
+      "learning_rate": 6.041666666666667e-07,
+      "loss": 0.1193,
+      "step": 435
+    },
+    {
+      "epoch": 18.166666666666668,
+      "grad_norm": 0.8241494297981262,
+      "learning_rate": 6.055555555555556e-07,
+      "loss": 0.1454,
+      "step": 436
+    },
+    {
+      "epoch": 18.208333333333332,
+      "grad_norm": 0.9220920205116272,
+      "learning_rate": 6.069444444444445e-07,
+      "loss": 0.1387,
+      "step": 437
+    },
+    {
+      "epoch": 18.25,
+      "grad_norm": 1.3029752969741821,
+      "learning_rate": 6.083333333333334e-07,
+      "loss": 0.1158,
+      "step": 438
+    },
+    {
+      "epoch": 18.291666666666668,
+      "grad_norm": 0.8959658145904541,
+      "learning_rate": 6.097222222222223e-07,
+      "loss": 0.1001,
+      "step": 439
+    },
+    {
+      "epoch": 18.333333333333332,
+      "grad_norm": 1.0864644050598145,
+      "learning_rate": 6.111111111111112e-07,
+      "loss": 0.1427,
+      "step": 440
+    },
+    {
+      "epoch": 18.375,
+      "grad_norm": 0.9895029664039612,
+      "learning_rate": 6.125000000000001e-07,
+      "loss": 0.0934,
+      "step": 441
+    },
+    {
+      "epoch": 18.416666666666668,
+      "grad_norm": 0.914701521396637,
+      "learning_rate": 6.13888888888889e-07,
+      "loss": 0.1438,
+      "step": 442
+    },
+    {
+      "epoch": 18.458333333333332,
+      "grad_norm": 1.0225493907928467,
+      "learning_rate": 6.152777777777779e-07,
+      "loss": 0.1061,
+      "step": 443
+    },
+    {
+      "epoch": 18.5,
+      "grad_norm": 0.9732456207275391,
+      "learning_rate": 6.166666666666668e-07,
+      "loss": 0.1383,
+      "step": 444
+    },
+    {
+      "epoch": 18.541666666666668,
+      "grad_norm": 1.1919742822647095,
+      "learning_rate": 6.180555555555557e-07,
+      "loss": 0.1053,
+      "step": 445
+    },
+    {
+      "epoch": 18.583333333333332,
+      "grad_norm": 0.9139710068702698,
+      "learning_rate": 6.194444444444446e-07,
+      "loss": 0.1431,
+      "step": 446
+    },
+    {
+      "epoch": 18.625,
+      "grad_norm": 1.4889836311340332,
+      "learning_rate": 6.208333333333334e-07,
+      "loss": 0.1151,
+      "step": 447
+    },
+    {
+      "epoch": 18.666666666666668,
+      "grad_norm": 1.1063458919525146,
+      "learning_rate": 6.222222222222223e-07,
+      "loss": 0.1398,
+      "step": 448
+    },
+    {
+      "epoch": 18.708333333333332,
+      "grad_norm": 0.8999032974243164,
+      "learning_rate": 6.236111111111112e-07,
+      "loss": 0.1312,
+      "step": 449
+    },
+    {
+      "epoch": 18.75,
+      "grad_norm": 1.087420105934143,
+      "learning_rate": 6.25e-07,
+      "loss": 0.1372,
+      "step": 450
+    },
+    {
+      "epoch": 18.791666666666668,
+      "grad_norm": 0.9422047734260559,
+      "learning_rate": 6.263888888888888e-07,
+      "loss": 0.1113,
+      "step": 451
+    },
+    {
+      "epoch": 18.833333333333332,
+      "grad_norm": 0.9118720889091492,
+      "learning_rate": 6.277777777777778e-07,
+      "loss": 0.1371,
+      "step": 452
+    },
+    {
+      "epoch": 18.875,
+      "grad_norm": 0.9406179785728455,
+      "learning_rate": 6.291666666666666e-07,
+      "loss": 0.1033,
+      "step": 453
+    },
+    {
+      "epoch": 18.916666666666668,
+      "grad_norm": 1.0062140226364136,
+      "learning_rate": 6.305555555555556e-07,
+      "loss": 0.1045,
+      "step": 454
+    },
+    {
+      "epoch": 18.958333333333332,
+      "grad_norm": 1.0125685930252075,
+      "learning_rate": 6.319444444444444e-07,
+      "loss": 0.1011,
+      "step": 455
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 1.1490026712417603,
+      "learning_rate": 6.333333333333334e-07,
+      "loss": 0.1053,
+      "step": 456
+    },
+    {
+      "epoch": 19.041666666666668,
+      "grad_norm": 0.9452362060546875,
+      "learning_rate": 6.347222222222222e-07,
+      "loss": 0.1048,
+      "step": 457
+    },
+    {
+      "epoch": 19.083333333333332,
+      "grad_norm": 0.9303408265113831,
+      "learning_rate": 6.361111111111111e-07,
+      "loss": 0.1311,
+      "step": 458
+    },
+    {
+      "epoch": 19.125,
+      "grad_norm": 1.7544077634811401,
+      "learning_rate": 6.375e-07,
+      "loss": 0.083,
+      "step": 459
+    },
+    {
+      "epoch": 19.166666666666668,
+      "grad_norm": 0.9402689933776855,
+      "learning_rate": 6.388888888888889e-07,
+      "loss": 0.0935,
+      "step": 460
+    },
+    {
+      "epoch": 19.208333333333332,
+      "grad_norm": 1.0125726461410522,
+      "learning_rate": 6.402777777777778e-07,
+      "loss": 0.0889,
+      "step": 461
+    },
+    {
+      "epoch": 19.25,
+      "grad_norm": 1.292421817779541,
+      "learning_rate": 6.416666666666667e-07,
+      "loss": 0.0945,
+      "step": 462
+    },
+    {
+      "epoch": 19.291666666666668,
+      "grad_norm": 0.9764171838760376,
+      "learning_rate": 6.430555555555556e-07,
+      "loss": 0.1283,
+      "step": 463
+    },
+    {
+      "epoch": 19.333333333333332,
+      "grad_norm": 1.0043569803237915,
+      "learning_rate": 6.444444444444445e-07,
+      "loss": 0.1499,
+      "step": 464
+    },
+    {
+      "epoch": 19.375,
+      "grad_norm": 1.2067495584487915,
+      "learning_rate": 6.458333333333334e-07,
+      "loss": 0.0669,
+      "step": 465
+    },
+    {
+      "epoch": 19.416666666666668,
+      "grad_norm": 0.949088990688324,
+      "learning_rate": 6.472222222222223e-07,
+      "loss": 0.1311,
+      "step": 466
+    },
+    {
+      "epoch": 19.458333333333332,
+      "grad_norm": 0.9593291878700256,
+      "learning_rate": 6.486111111111111e-07,
+      "loss": 0.0984,
+      "step": 467
+    },
+    {
+      "epoch": 19.5,
+      "grad_norm": 1.4252955913543701,
+      "learning_rate": 6.5e-07,
+      "loss": 0.1416,
+      "step": 468
+    },
+    {
+      "epoch": 19.541666666666668,
+      "grad_norm": 1.1030102968215942,
+      "learning_rate": 6.513888888888889e-07,
+      "loss": 0.1049,
+      "step": 469
+    },
+    {
+      "epoch": 19.583333333333332,
+      "grad_norm": 1.0402379035949707,
+      "learning_rate": 6.527777777777778e-07,
+      "loss": 0.0731,
+      "step": 470
+    },
+    {
+      "epoch": 19.625,
+      "grad_norm": 0.9812272191047668,
+      "learning_rate": 6.541666666666667e-07,
+      "loss": 0.1191,
+      "step": 471
+    },
+    {
+      "epoch": 19.666666666666668,
+      "grad_norm": 0.9214704036712646,
+      "learning_rate": 6.555555555555556e-07,
+      "loss": 0.1176,
+      "step": 472
+    },
+    {
+      "epoch": 19.708333333333332,
+      "grad_norm": 0.9828305244445801,
+      "learning_rate": 6.569444444444445e-07,
+      "loss": 0.0962,
+      "step": 473
+    },
+    {
+      "epoch": 19.75,
+      "grad_norm": 1.0520708560943604,
+      "learning_rate": 6.583333333333333e-07,
+      "loss": 0.1113,
+      "step": 474
+    },
+    {
+      "epoch": 19.791666666666668,
+      "grad_norm": 0.9608191251754761,
+      "learning_rate": 6.597222222222223e-07,
+      "loss": 0.1283,
+      "step": 475
+    },
+    {
+      "epoch": 19.833333333333332,
+      "grad_norm": 0.8865001797676086,
+      "learning_rate": 6.611111111111111e-07,
+      "loss": 0.0908,
+      "step": 476
+    },
+    {
+      "epoch": 19.875,
+      "grad_norm": 0.958601176738739,
+      "learning_rate": 6.625000000000001e-07,
+      "loss": 0.1098,
+      "step": 477
+    },
+    {
+      "epoch": 19.916666666666668,
+      "grad_norm": 0.9882670044898987,
+      "learning_rate": 6.638888888888889e-07,
+      "loss": 0.0902,
+      "step": 478
+    },
+    {
+      "epoch": 19.958333333333332,
+      "grad_norm": 1.687676191329956,
+      "learning_rate": 6.652777777777779e-07,
+      "loss": 0.1102,
+      "step": 479
+    },
+    {
+      "epoch": 20.0,
+      "grad_norm": 0.9980921149253845,
+      "learning_rate": 6.666666666666667e-07,
+      "loss": 0.0964,
+      "step": 480
+    },
+    {
+      "epoch": 20.041666666666668,
+      "grad_norm": 0.8867719769477844,
+      "learning_rate": 6.680555555555557e-07,
+      "loss": 0.0864,
+      "step": 481
+    },
+    {
+      "epoch": 20.083333333333332,
+      "grad_norm": 0.7900561690330505,
+      "learning_rate": 6.694444444444445e-07,
+      "loss": 0.0907,
+      "step": 482
+    },
+    {
+      "epoch": 20.125,
+      "grad_norm": 0.8275968432426453,
+      "learning_rate": 6.708333333333333e-07,
+      "loss": 0.0828,
+      "step": 483
+    },
+    {
+      "epoch": 20.166666666666668,
+      "grad_norm": 0.9496970772743225,
+      "learning_rate": 6.722222222222223e-07,
+      "loss": 0.0953,
+      "step": 484
+    },
+    {
+      "epoch": 20.208333333333332,
+      "grad_norm": 1.5215312242507935,
+      "learning_rate": 6.736111111111111e-07,
+      "loss": 0.0767,
+      "step": 485
+    },
+    {
+      "epoch": 20.25,
+      "grad_norm": 1.0067987442016602,
+      "learning_rate": 6.750000000000001e-07,
+      "loss": 0.1058,
+      "step": 486
+    },
+    {
+      "epoch": 20.291666666666668,
+      "grad_norm": 0.9322746992111206,
+      "learning_rate": 6.763888888888889e-07,
+      "loss": 0.0787,
+      "step": 487
+    },
+    {
+      "epoch": 20.333333333333332,
+      "grad_norm": 0.9669644236564636,
+      "learning_rate": 6.777777777777779e-07,
+      "loss": 0.0836,
+      "step": 488
+    },
+    {
+      "epoch": 20.375,
+      "grad_norm": 0.9791039824485779,
+      "learning_rate": 6.791666666666667e-07,
+      "loss": 0.0929,
+      "step": 489
+    },
+    {
+      "epoch": 20.416666666666668,
+      "grad_norm": 0.9835072755813599,
+      "learning_rate": 6.805555555555556e-07,
+      "loss": 0.087,
+      "step": 490
+    },
+    {
+      "epoch": 20.458333333333332,
+      "grad_norm": 0.8946750164031982,
+      "learning_rate": 6.819444444444445e-07,
+      "loss": 0.0744,
+      "step": 491
+    },
+    {
+      "epoch": 20.5,
+      "grad_norm": 1.2975590229034424,
+      "learning_rate": 6.833333333333334e-07,
+      "loss": 0.0818,
+      "step": 492
+    },
+    {
+      "epoch": 20.541666666666668,
+      "grad_norm": 1.3321139812469482,
+      "learning_rate": 6.847222222222223e-07,
+      "loss": 0.1223,
+      "step": 493
+    },
+    {
+      "epoch": 20.583333333333332,
+      "grad_norm": 0.960700273513794,
+      "learning_rate": 6.861111111111112e-07,
+      "loss": 0.089,
+      "step": 494
+    },
+    {
+      "epoch": 20.625,
+      "grad_norm": 0.9136600494384766,
+      "learning_rate": 6.875000000000001e-07,
+      "loss": 0.1148,
+      "step": 495
+    },
+    {
+      "epoch": 20.666666666666668,
+      "grad_norm": 0.9647133946418762,
+      "learning_rate": 6.88888888888889e-07,
+      "loss": 0.1426,
+      "step": 496
+    },
+    {
+      "epoch": 20.708333333333332,
+      "grad_norm": 1.2264877557754517,
+      "learning_rate": 6.902777777777778e-07,
+      "loss": 0.1122,
+      "step": 497
+    },
+    {
+      "epoch": 20.75,
+      "grad_norm": 1.1542943716049194,
+      "learning_rate": 6.916666666666668e-07,
+      "loss": 0.0984,
+      "step": 498
+    },
+    {
+      "epoch": 20.791666666666668,
+      "grad_norm": 0.9621667265892029,
+      "learning_rate": 6.930555555555556e-07,
+      "loss": 0.0954,
+      "step": 499
+    },
+    {
+      "epoch": 20.833333333333332,
+      "grad_norm": 0.9104325175285339,
+      "learning_rate": 6.944444444444446e-07,
+      "loss": 0.0685,
+      "step": 500
+    },
+    {
+      "epoch": 20.875,
+      "grad_norm": 0.8492055535316467,
+      "learning_rate": 6.958333333333334e-07,
+      "loss": 0.1292,
+      "step": 501
+    },
+    {
+      "epoch": 20.916666666666668,
+      "grad_norm": 1.0698487758636475,
+      "learning_rate": 6.972222222222223e-07,
+      "loss": 0.0752,
+      "step": 502
+    },
+    {
+      "epoch": 20.958333333333332,
+      "grad_norm": 1.2914539575576782,
+      "learning_rate": 6.986111111111112e-07,
+      "loss": 0.0822,
+      "step": 503
+    },
+    {
+      "epoch": 21.0,
+      "grad_norm": 1.169684648513794,
+      "learning_rate": 7.000000000000001e-07,
+      "loss": 0.0836,
+      "step": 504
+    },
+    {
+      "epoch": 21.041666666666668,
+      "grad_norm": 0.9427048563957214,
+      "learning_rate": 7.01388888888889e-07,
+      "loss": 0.0832,
+      "step": 505
+    },
+    {
+      "epoch": 21.083333333333332,
+      "grad_norm": 0.8189440369606018,
+      "learning_rate": 7.027777777777778e-07,
+      "loss": 0.0573,
+      "step": 506
+    },
+    {
+      "epoch": 21.125,
+      "grad_norm": 0.9091562628746033,
+      "learning_rate": 7.041666666666668e-07,
+      "loss": 0.0668,
+      "step": 507
+    },
+    {
+      "epoch": 21.166666666666668,
+      "grad_norm": 0.8724696040153503,
+      "learning_rate": 7.055555555555556e-07,
+      "loss": 0.0617,
+      "step": 508
+    },
+    {
+      "epoch": 21.208333333333332,
+      "grad_norm": 0.8674677610397339,
+      "learning_rate": 7.069444444444446e-07,
+      "loss": 0.1071,
+      "step": 509
+    },
+    {
+      "epoch": 21.25,
+      "grad_norm": 1.266666293144226,
+      "learning_rate": 7.083333333333334e-07,
+      "loss": 0.0655,
+      "step": 510
+    },
+    {
+      "epoch": 21.291666666666668,
+      "grad_norm": 1.0958201885223389,
+      "learning_rate": 7.097222222222224e-07,
+      "loss": 0.0854,
+      "step": 511
+    },
+    {
+      "epoch": 21.333333333333332,
+      "grad_norm": 1.1242408752441406,
+      "learning_rate": 7.111111111111112e-07,
+      "loss": 0.092,
+      "step": 512
+    },
+    {
+      "epoch": 21.375,
+      "grad_norm": 1.0377740859985352,
+      "learning_rate": 7.125e-07,
+      "loss": 0.1144,
+      "step": 513
+    },
+    {
+      "epoch": 21.416666666666668,
+      "grad_norm": 1.004826307296753,
+      "learning_rate": 7.13888888888889e-07,
+      "loss": 0.0644,
+      "step": 514
+    },
+    {
+      "epoch": 21.458333333333332,
+      "grad_norm": 1.0715738534927368,
+      "learning_rate": 7.152777777777778e-07,
+      "loss": 0.0823,
+      "step": 515
+    },
+    {
+      "epoch": 21.5,
+      "grad_norm": 1.0161361694335938,
+      "learning_rate": 7.166666666666668e-07,
+      "loss": 0.0742,
+      "step": 516
+    },
+    {
+      "epoch": 21.541666666666668,
+      "grad_norm": 0.9937750101089478,
+      "learning_rate": 7.180555555555556e-07,
+      "loss": 0.1172,
+      "step": 517
+    },
+    {
+      "epoch": 21.583333333333332,
+      "grad_norm": 0.9221785664558411,
+      "learning_rate": 7.194444444444446e-07,
+      "loss": 0.0613,
+      "step": 518
+    },
+    {
+      "epoch": 21.625,
+      "grad_norm": 1.4999557733535767,
+      "learning_rate": 7.208333333333333e-07,
+      "loss": 0.1102,
+      "step": 519
+    },
+    {
+      "epoch": 21.666666666666668,
+      "grad_norm": 1.3033219575881958,
+      "learning_rate": 7.222222222222222e-07,
+      "loss": 0.0741,
+      "step": 520
+    },
+    {
+      "epoch": 21.708333333333332,
+      "grad_norm": 1.052722692489624,
+      "learning_rate": 7.236111111111111e-07,
+      "loss": 0.0928,
+      "step": 521
+    },
+    {
+      "epoch": 21.75,
+      "grad_norm": 1.1567740440368652,
+      "learning_rate": 7.25e-07,
+      "loss": 0.0774,
+      "step": 522
+    },
+    {
+      "epoch": 21.791666666666668,
+      "grad_norm": 1.0079861879348755,
+      "learning_rate": 7.263888888888889e-07,
+      "loss": 0.1046,
+      "step": 523
+    },
+    {
+      "epoch": 21.833333333333332,
+      "grad_norm": 1.5346205234527588,
+      "learning_rate": 7.277777777777778e-07,
+      "loss": 0.0668,
+      "step": 524
+    },
+    {
+      "epoch": 21.875,
+      "grad_norm": 1.2241863012313843,
+      "learning_rate": 7.291666666666667e-07,
+      "loss": 0.0746,
+      "step": 525
+    },
+    {
+      "epoch": 21.916666666666668,
+      "grad_norm": 1.198899507522583,
+      "learning_rate": 7.305555555555556e-07,
+      "loss": 0.0706,
+      "step": 526
+    },
+    {
+      "epoch": 21.958333333333332,
+      "grad_norm": 1.3769057989120483,
+      "learning_rate": 7.319444444444445e-07,
+      "loss": 0.0684,
+      "step": 527
+    },
+    {
+      "epoch": 22.0,
+      "grad_norm": 1.162265419960022,
+      "learning_rate": 7.333333333333334e-07,
+      "loss": 0.0636,
+      "step": 528
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 240000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10000,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 403103519604736.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

sft_search/JOB:24815#LR:1e-5#BASE:Qwen2.5-7B-Instruct#TOKEN:Qwen2.5-7B-Instruct#BSZ:2#ACC:4_ablation_resp_difficulty_1678_ablation/checkpoint-156/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

sft_search/JOB:24815#LR:1e-5#BASE:Qwen2.5-7B-Instruct#TOKEN:Qwen2.5-7B-Instruct#BSZ:2#ACC:4_ablation_resp_difficulty_1678_ablation/checkpoint-156/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a1c0c92457443985d374d3350c01e0cb0703d39efbb9043cb24af6fd3d5e8d7
+size 7544

sft_search/JOB:30702#LR:1e-5#BASE:Qwen2.5-7B-Instruct#TOKEN:Qwen2.5-7B-Instruct#BSZ:2#ACC:4_ablation_resp_format_1064_random_sample_871_ablation/checkpoint-27/trainer_state.json

@@ -0,0 +1,222 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.9636363636363636,
+  "eval_steps": 500,
+  "global_step": 27,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07272727272727272,
+      "grad_norm": 17.320636749267578,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 1.5279,
+      "step": 1
+    },
+    {
+      "epoch": 0.14545454545454545,
+      "grad_norm": 19.119434356689453,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 1.5704,
+      "step": 2
+    },
+    {
+      "epoch": 0.21818181818181817,
+      "grad_norm": 15.651047706604004,
+      "learning_rate": 1e-05,
+      "loss": 1.4433,
+      "step": 3
+    },
+    {
+      "epoch": 0.2909090909090909,
+      "grad_norm": 8.04721450805664,
+      "learning_rate": 9.995614150494293e-06,
+      "loss": 1.1664,
+      "step": 4
+    },
+    {
+      "epoch": 0.36363636363636365,
+      "grad_norm": 7.035794258117676,
+      "learning_rate": 9.982464296247523e-06,
+      "loss": 1.0968,
+      "step": 5
+    },
+    {
+      "epoch": 0.43636363636363634,
+      "grad_norm": 5.759443759918213,
+      "learning_rate": 9.960573506572391e-06,
+      "loss": 1.0498,
+      "step": 6
+    },
+    {
+      "epoch": 0.509090909090909,
+      "grad_norm": 6.248385429382324,
+      "learning_rate": 9.929980185352525e-06,
+      "loss": 0.9743,
+      "step": 7
+    },
+    {
+      "epoch": 0.5818181818181818,
+      "grad_norm": 4.9278883934021,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.9035,
+      "step": 8
+    },
+    {
+      "epoch": 0.6545454545454545,
+      "grad_norm": 3.3208532333374023,
+      "learning_rate": 9.842915805643156e-06,
+      "loss": 0.8818,
+      "step": 9
+    },
+    {
+      "epoch": 0.7272727272727273,
+      "grad_norm": 2.926849126815796,
+      "learning_rate": 9.786597487660336e-06,
+      "loss": 0.8444,
+      "step": 10
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 3.0539627075195312,
+      "learning_rate": 9.721881851187406e-06,
+      "loss": 0.8301,
+      "step": 11
+    },
+    {
+      "epoch": 0.8727272727272727,
+      "grad_norm": 2.6069295406341553,
+      "learning_rate": 9.648882429441258e-06,
+      "loss": 0.836,
+      "step": 12
+    },
+    {
+      "epoch": 0.9454545454545454,
+      "grad_norm": 2.059221029281616,
+      "learning_rate": 9.567727288213005e-06,
+      "loss": 0.7984,
+      "step": 13
+    },
+    {
+      "epoch": 1.018181818181818,
+      "grad_norm": 3.098663330078125,
+      "learning_rate": 9.478558801197065e-06,
+      "loss": 0.9665,
+      "step": 14
+    },
+    {
+      "epoch": 1.0909090909090908,
+      "grad_norm": 2.1216795444488525,
+      "learning_rate": 9.381533400219319e-06,
+      "loss": 0.6988,
+      "step": 15
+    },
+    {
+      "epoch": 1.1636363636363636,
+      "grad_norm": 2.0530576705932617,
+      "learning_rate": 9.276821300802535e-06,
+      "loss": 0.6336,
+      "step": 16
+    },
+    {
+      "epoch": 1.2363636363636363,
+      "grad_norm": 1.8729101419448853,
+      "learning_rate": 9.164606203550498e-06,
+      "loss": 0.6305,
+      "step": 17
+    },
+    {
+      "epoch": 1.309090909090909,
+      "grad_norm": 1.659616231918335,
+      "learning_rate": 9.045084971874738e-06,
+      "loss": 0.5688,
+      "step": 18
+    },
+    {
+      "epoch": 1.3818181818181818,
+      "grad_norm": 2.068031072616577,
+      "learning_rate": 8.9184672866292e-06,
+      "loss": 0.6353,
+      "step": 19
+    },
+    {
+      "epoch": 1.4545454545454546,
+      "grad_norm": 1.9470592737197876,
+      "learning_rate": 8.784975278258783e-06,
+      "loss": 0.5986,
+      "step": 20
+    },
+    {
+      "epoch": 1.5272727272727273,
+      "grad_norm": 1.8197174072265625,
+      "learning_rate": 8.644843137107058e-06,
+      "loss": 0.5576,
+      "step": 21
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 1.7431132793426514,
+      "learning_rate": 8.498316702566828e-06,
+      "loss": 0.6405,
+      "step": 22
+    },
+    {
+      "epoch": 1.6727272727272728,
+      "grad_norm": 1.7749464511871338,
+      "learning_rate": 8.345653031794292e-06,
+      "loss": 0.5903,
+      "step": 23
+    },
+    {
+      "epoch": 1.7454545454545456,
+      "grad_norm": 1.7153894901275635,
+      "learning_rate": 8.18711994874345e-06,
+      "loss": 0.5586,
+      "step": 24
+    },
+    {
+      "epoch": 1.8181818181818183,
+      "grad_norm": 1.6373153924942017,
+      "learning_rate": 8.022995574311876e-06,
+      "loss": 0.5614,
+      "step": 25
+    },
+    {
+      "epoch": 1.8909090909090909,
+      "grad_norm": 1.7297462224960327,
+      "learning_rate": 7.85356783842216e-06,
+      "loss": 0.6556,
+      "step": 26
+    },
+    {
+      "epoch": 1.9636363636363636,
+      "grad_norm": 1.8113256692886353,
+      "learning_rate": 7.679133974894984e-06,
+      "loss": 0.6212,
+      "step": 27
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 78,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5074619187200.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

sft_search/JOB:30936#LR:1e-5#BASE:Qwen2.5-7B-Instruct#TOKEN:Qwen2.5-7B-Instruct#BSZ:2#ACC:4_math_qwq_4524_selected_add_prompt_871/checkpoint-55/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7b947c6cbeab0aa9634740df35f4305dc8d5d62a8f89603f0e69b5189665fa1
+size 4877660776

sft_search/JOB:30936#LR:1e-5#BASE:Qwen2.5-7B-Instruct#TOKEN:Qwen2.5-7B-Instruct#BSZ:2#ACC:4_math_qwq_4524_selected_add_prompt_871/checkpoint-55/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d25586f37f4fbf2b84517627cd3adce6fd040be41b8dadeb83b7796ec9a24f4f
+size 4330865200

sft_search/JOB:30936#LR:1e-5#BASE:Qwen2.5-7B-Instruct#TOKEN:Qwen2.5-7B-Instruct#BSZ:2#ACC:4_math_qwq_4524_selected_add_prompt_871/checkpoint-55/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1678a1416dbedd1fd4a7265769ae7d979882b7d2a97e36674c8ac181d34b2c2e
+size 1089994880

sft_search/JOB:8557#LR:1e-5#BASE:Qwen2.5-7B-Instruct#TOKEN:Qwen2.5-7B-Instruct#BSZ:2#ACC:4_ablation_resp_reason_1099_ablation/checkpoint-102/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": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}