Upload folder using huggingface_hub

.gitattributes

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

added_tokens.json

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

config.json

@@ -0,0 +1,32 @@
+{
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "eos_token_id": 151645,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2560,
+  "initializer_range": 0.02,
+  "intermediate_size": 9728,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 36,
+  "model_type": "qwen3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 8,
+  "pad_token_id": 151654,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "unsloth_fixed": true,
+  "unsloth_version": "2025.4.4",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "max_length": 32768,
+  "max_new_tokens": 2048,
+  "pad_token_id": 151654,
+  "transformers_version": "4.51.3"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ddf0742586b99bf90c8652959c07ee0f8696346d18d69391b1f7532cf5f846f
+size 4967215360

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96ff0b559ba89c449e5e40340f600212c706643a1b9329f4267ba5329320b666
+size 3077766632

model.safetensors.index.json

@@ -0,0 +1,405 @@
+{
+  "metadata": {
+    "total_size": 8044936192
+  },
+  "weight_map": {
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c800b778fa7e115e4c34de8529902de8b61c9a1b4bab3eb8295d06dafff030e
+size 14645

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b92649eec197d2aba10aaedf8d2fd3bd1a27a232f2e1e418c29e7530005e7bf4
+size 1465

special_tokens_map.json

@@ -0,0 +1,16 @@
+{
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|vision_pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

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

tokenizer_config.json

@@ -0,0 +1,189 @@
+{
+  "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
+    }
+  },
+  "bos_token": null,
+  "chat_template": "{% if 'role' in messages[0] %}{% for message in messages %}{% if message['role'] == 'user' %}{{'<|im_start|>user\n' + message['content'] + '<|im_end|>\n'}}{% elif message['role'] == 'assistant' %}{{'<|im_start|>assistant\n' + message['content'] + '<|im_end|>\n' }}{% else %}{{ '<|im_start|>system\n' + message['content'] + '<|im_end|>\n' }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}{% else %}{% for message in messages %}{% if message['from'] == 'human' %}{{'<|im_start|>user\n' + message['value'] + '<|im_end|>\n'}}{% elif message['from'] == 'gpt' %}{{'<|im_start|>assistant\n' + message['value'] + '<|im_end|>\n' }}{% else %}{{ '<|im_start|>system\n' + message['value'] + '<|im_end|>\n' }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "extra_special_tokens": {},
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|vision_pad|>",
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,2218 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.99904,
+  "eval_steps": 500,
+  "global_step": 3124,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0064,
+      "grad_norm": 1.6328125,
+      "learning_rate": 2.7000000000000002e-05,
+      "loss": 1.239,
+      "step": 10
+    },
+    {
+      "epoch": 0.0128,
+      "grad_norm": 2.328125,
+      "learning_rate": 2.99993816906306e-05,
+      "loss": 1.1046,
+      "step": 20
+    },
+    {
+      "epoch": 0.0192,
+      "grad_norm": 1.78125,
+      "learning_rate": 2.999724439035231e-05,
+      "loss": 1.1791,
+      "step": 30
+    },
+    {
+      "epoch": 0.0256,
+      "grad_norm": 1.78125,
+      "learning_rate": 2.9993580683196947e-05,
+      "loss": 1.213,
+      "step": 40
+    },
+    {
+      "epoch": 0.032,
+      "grad_norm": 1.875,
+      "learning_rate": 2.9988390942053532e-05,
+      "loss": 1.2061,
+      "step": 50
+    },
+    {
+      "epoch": 0.0384,
+      "grad_norm": 2.125,
+      "learning_rate": 2.998167569512954e-05,
+      "loss": 1.1253,
+      "step": 60
+    },
+    {
+      "epoch": 0.0448,
+      "grad_norm": 2.53125,
+      "learning_rate": 2.997343562589714e-05,
+      "loss": 1.1766,
+      "step": 70
+    },
+    {
+      "epoch": 0.0512,
+      "grad_norm": 2.0,
+      "learning_rate": 2.9963671573023613e-05,
+      "loss": 1.1442,
+      "step": 80
+    },
+    {
+      "epoch": 0.0576,
+      "grad_norm": 3.171875,
+      "learning_rate": 2.9952384530286028e-05,
+      "loss": 1.0799,
+      "step": 90
+    },
+    {
+      "epoch": 0.064,
+      "grad_norm": 2.09375,
+      "learning_rate": 2.9939575646470072e-05,
+      "loss": 1.108,
+      "step": 100
+    },
+    {
+      "epoch": 0.0704,
+      "grad_norm": 2.65625,
+      "learning_rate": 2.9925246225253124e-05,
+      "loss": 1.1166,
+      "step": 110
+    },
+    {
+      "epoch": 0.0768,
+      "grad_norm": 1.9375,
+      "learning_rate": 2.9909397725071577e-05,
+      "loss": 1.1037,
+      "step": 120
+    },
+    {
+      "epoch": 0.0832,
+      "grad_norm": 2.359375,
+      "learning_rate": 2.9892031758972416e-05,
+      "loss": 1.1854,
+      "step": 130
+    },
+    {
+      "epoch": 0.0896,
+      "grad_norm": 1.6171875,
+      "learning_rate": 2.9873150094449012e-05,
+      "loss": 1.1581,
+      "step": 140
+    },
+    {
+      "epoch": 0.096,
+      "grad_norm": 3.859375,
+      "learning_rate": 2.9852754653261236e-05,
+      "loss": 1.1411,
+      "step": 150
+    },
+    {
+      "epoch": 0.1024,
+      "grad_norm": 2.796875,
+      "learning_rate": 2.9830847511239886e-05,
+      "loss": 1.2032,
+      "step": 160
+    },
+    {
+      "epoch": 0.1088,
+      "grad_norm": 1.75,
+      "learning_rate": 2.980743089807539e-05,
+      "loss": 1.1435,
+      "step": 170
+    },
+    {
+      "epoch": 0.1152,
+      "grad_norm": 2.203125,
+      "learning_rate": 2.9782507197090873e-05,
+      "loss": 1.105,
+      "step": 180
+    },
+    {
+      "epoch": 0.1216,
+      "grad_norm": 2.78125,
+      "learning_rate": 2.9756078944999583e-05,
+      "loss": 1.1362,
+      "step": 190
+    },
+    {
+      "epoch": 0.128,
+      "grad_norm": 1.9453125,
+      "learning_rate": 2.972814883164671e-05,
+      "loss": 1.0945,
+      "step": 200
+    },
+    {
+      "epoch": 0.1344,
+      "grad_norm": 1.9609375,
+      "learning_rate": 2.9698719699735635e-05,
+      "loss": 1.138,
+      "step": 210
+    },
+    {
+      "epoch": 0.1408,
+      "grad_norm": 2.609375,
+      "learning_rate": 2.9667794544538547e-05,
+      "loss": 1.1012,
+      "step": 220
+    },
+    {
+      "epoch": 0.1472,
+      "grad_norm": 2.21875,
+      "learning_rate": 2.963537651359166e-05,
+      "loss": 1.1816,
+      "step": 230
+    },
+    {
+      "epoch": 0.1536,
+      "grad_norm": 2.078125,
+      "learning_rate": 2.9601468906374795e-05,
+      "loss": 1.1537,
+      "step": 240
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 1.921875,
+      "learning_rate": 2.9566075173975597e-05,
+      "loss": 1.1143,
+      "step": 250
+    },
+    {
+      "epoch": 0.1664,
+      "grad_norm": 1.703125,
+      "learning_rate": 2.9529198918738284e-05,
+      "loss": 1.1162,
+      "step": 260
+    },
+    {
+      "epoch": 0.1728,
+      "grad_norm": 2.296875,
+      "learning_rate": 2.949084389389699e-05,
+      "loss": 1.1628,
+      "step": 270
+    },
+    {
+      "epoch": 0.1792,
+      "grad_norm": 2.15625,
+      "learning_rate": 2.9451014003193773e-05,
+      "loss": 1.1051,
+      "step": 280
+    },
+    {
+      "epoch": 0.1856,
+      "grad_norm": 2.4375,
+      "learning_rate": 2.9409713300481292e-05,
+      "loss": 1.117,
+      "step": 290
+    },
+    {
+      "epoch": 0.192,
+      "grad_norm": 2.484375,
+      "learning_rate": 2.9366945989310225e-05,
+      "loss": 1.1111,
+      "step": 300
+    },
+    {
+      "epoch": 0.1984,
+      "grad_norm": 2.234375,
+      "learning_rate": 2.9322716422501393e-05,
+      "loss": 1.1349,
+      "step": 310
+    },
+    {
+      "epoch": 0.2048,
+      "grad_norm": 1.921875,
+      "learning_rate": 2.9277029101702786e-05,
+      "loss": 1.0544,
+      "step": 320
+    },
+    {
+      "epoch": 0.2112,
+      "grad_norm": 1.8046875,
+      "learning_rate": 2.9229888676931357e-05,
+      "loss": 1.1074,
+      "step": 330
+    },
+    {
+      "epoch": 0.2176,
+      "grad_norm": 1.828125,
+      "learning_rate": 2.9181299946099754e-05,
+      "loss": 1.1687,
+      "step": 340
+    },
+    {
+      "epoch": 0.224,
+      "grad_norm": 1.71875,
+      "learning_rate": 2.9131267854527993e-05,
+      "loss": 1.1373,
+      "step": 350
+    },
+    {
+      "epoch": 0.2304,
+      "grad_norm": 2.0625,
+      "learning_rate": 2.9079797494440136e-05,
+      "loss": 1.1449,
+      "step": 360
+    },
+    {
+      "epoch": 0.2368,
+      "grad_norm": 2.703125,
+      "learning_rate": 2.9026894104445985e-05,
+      "loss": 1.0586,
+      "step": 370
+    },
+    {
+      "epoch": 0.2432,
+      "grad_norm": 2.21875,
+      "learning_rate": 2.8972563069007933e-05,
+      "loss": 1.1321,
+      "step": 380
+    },
+    {
+      "epoch": 0.2496,
+      "grad_norm": 2.234375,
+      "learning_rate": 2.891680991789291e-05,
+      "loss": 1.0835,
+      "step": 390
+    },
+    {
+      "epoch": 0.256,
+      "grad_norm": 2.03125,
+      "learning_rate": 2.8859640325609582e-05,
+      "loss": 1.0869,
+      "step": 400
+    },
+    {
+      "epoch": 0.2624,
+      "grad_norm": 1.78125,
+      "learning_rate": 2.8801060110830807e-05,
+      "loss": 1.0757,
+      "step": 410
+    },
+    {
+      "epoch": 0.2688,
+      "grad_norm": 2.046875,
+      "learning_rate": 2.8741075235801406e-05,
+      "loss": 1.077,
+      "step": 420
+    },
+    {
+      "epoch": 0.2752,
+      "grad_norm": 2.0625,
+      "learning_rate": 2.8679691805731334e-05,
+      "loss": 1.1234,
+      "step": 430
+    },
+    {
+      "epoch": 0.2816,
+      "grad_norm": 1.9296875,
+      "learning_rate": 2.8616916068174298e-05,
+      "loss": 1.1914,
+      "step": 440
+    },
+    {
+      "epoch": 0.288,
+      "grad_norm": 2.265625,
+      "learning_rate": 2.8552754412391885e-05,
+      "loss": 1.175,
+      "step": 450
+    },
+    {
+      "epoch": 0.2944,
+      "grad_norm": 2.75,
+      "learning_rate": 2.8487213368703274e-05,
+      "loss": 1.1524,
+      "step": 460
+    },
+    {
+      "epoch": 0.3008,
+      "grad_norm": 1.9375,
+      "learning_rate": 2.8420299607820574e-05,
+      "loss": 1.0483,
+      "step": 470
+    },
+    {
+      "epoch": 0.3072,
+      "grad_norm": 2.28125,
+      "learning_rate": 2.8352019940169903e-05,
+      "loss": 1.1059,
+      "step": 480
+    },
+    {
+      "epoch": 0.3136,
+      "grad_norm": 2.359375,
+      "learning_rate": 2.8282381315198205e-05,
+      "loss": 1.2019,
+      "step": 490
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 1.8125,
+      "learning_rate": 2.8211390820665962e-05,
+      "loss": 1.1448,
+      "step": 500
+    },
+    {
+      "epoch": 0.3264,
+      "grad_norm": 1.9375,
+      "learning_rate": 2.8139055681925785e-05,
+      "loss": 1.166,
+      "step": 510
+    },
+    {
+      "epoch": 0.3328,
+      "grad_norm": 1.9375,
+      "learning_rate": 2.8065383261187055e-05,
+      "loss": 1.1071,
+      "step": 520
+    },
+    {
+      "epoch": 0.3392,
+      "grad_norm": 1.8125,
+      "learning_rate": 2.7990381056766583e-05,
+      "loss": 1.0663,
+      "step": 530
+    },
+    {
+      "epoch": 0.3456,
+      "grad_norm": 2.359375,
+      "learning_rate": 2.7914056702325426e-05,
+      "loss": 1.0808,
+      "step": 540
+    },
+    {
+      "epoch": 0.352,
+      "grad_norm": 1.9609375,
+      "learning_rate": 2.783641796609197e-05,
+      "loss": 1.0989,
+      "step": 550
+    },
+    {
+      "epoch": 0.3584,
+      "grad_norm": 1.78125,
+      "learning_rate": 2.7757472750071263e-05,
+      "loss": 1.1476,
+      "step": 560
+    },
+    {
+      "epoch": 0.3648,
+      "grad_norm": 2.5625,
+      "learning_rate": 2.7677229089240768e-05,
+      "loss": 1.1165,
+      "step": 570
+    },
+    {
+      "epoch": 0.3712,
+      "grad_norm": 2.03125,
+      "learning_rate": 2.7595695150732566e-05,
+      "loss": 1.1196,
+      "step": 580
+    },
+    {
+      "epoch": 0.3776,
+      "grad_norm": 2.015625,
+      "learning_rate": 2.75128792330021e-05,
+      "loss": 1.1399,
+      "step": 590
+    },
+    {
+      "epoch": 0.384,
+      "grad_norm": 2.328125,
+      "learning_rate": 2.7428789764983587e-05,
+      "loss": 1.1192,
+      "step": 600
+    },
+    {
+      "epoch": 0.3904,
+      "grad_norm": 2.21875,
+      "learning_rate": 2.7343435305232116e-05,
+      "loss": 1.1318,
+      "step": 610
+    },
+    {
+      "epoch": 0.3968,
+      "grad_norm": 1.796875,
+      "learning_rate": 2.7256824541052566e-05,
+      "loss": 1.2052,
+      "step": 620
+    },
+    {
+      "epoch": 0.4032,
+      "grad_norm": 1.984375,
+      "learning_rate": 2.7168966287615415e-05,
+      "loss": 1.1994,
+      "step": 630
+    },
+    {
+      "epoch": 0.4096,
+      "grad_norm": 2.0,
+      "learning_rate": 2.7079869487059554e-05,
+      "loss": 1.0983,
+      "step": 640
+    },
+    {
+      "epoch": 0.416,
+      "grad_norm": 2.359375,
+      "learning_rate": 2.6989543207582145e-05,
+      "loss": 1.0591,
+      "step": 650
+    },
+    {
+      "epoch": 0.4224,
+      "grad_norm": 2.09375,
+      "learning_rate": 2.6897996642515685e-05,
+      "loss": 1.1056,
+      "step": 660
+    },
+    {
+      "epoch": 0.4288,
+      "grad_norm": 2.28125,
+      "learning_rate": 2.6805239109392308e-05,
+      "loss": 1.1212,
+      "step": 670
+    },
+    {
+      "epoch": 0.4352,
+      "grad_norm": 1.84375,
+      "learning_rate": 2.671128004899545e-05,
+      "loss": 1.1195,
+      "step": 680
+    },
+    {
+      "epoch": 0.4416,
+      "grad_norm": 2.078125,
+      "learning_rate": 2.661612902439898e-05,
+      "loss": 1.139,
+      "step": 690
+    },
+    {
+      "epoch": 0.448,
+      "grad_norm": 2.0625,
+      "learning_rate": 2.651979571999388e-05,
+      "loss": 1.1123,
+      "step": 700
+    },
+    {
+      "epoch": 0.4544,
+      "grad_norm": 1.8515625,
+      "learning_rate": 2.6422289940502575e-05,
+      "loss": 1.1357,
+      "step": 710
+    },
+    {
+      "epoch": 0.4608,
+      "grad_norm": 2.375,
+      "learning_rate": 2.6323621609981003e-05,
+      "loss": 1.136,
+      "step": 720
+    },
+    {
+      "epoch": 0.4672,
+      "grad_norm": 1.8671875,
+      "learning_rate": 2.6223800770808576e-05,
+      "loss": 1.1279,
+      "step": 730
+    },
+    {
+      "epoch": 0.4736,
+      "grad_norm": 2.265625,
+      "learning_rate": 2.6122837582666065e-05,
+      "loss": 1.1178,
+      "step": 740
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 1.96875,
+      "learning_rate": 2.6020742321501548e-05,
+      "loss": 1.1197,
+      "step": 750
+    },
+    {
+      "epoch": 0.4864,
+      "grad_norm": 2.140625,
+      "learning_rate": 2.5917525378484542e-05,
+      "loss": 1.1162,
+      "step": 760
+    },
+    {
+      "epoch": 0.4928,
+      "grad_norm": 1.828125,
+      "learning_rate": 2.581319725894839e-05,
+      "loss": 1.089,
+      "step": 770
+    },
+    {
+      "epoch": 0.4992,
+      "grad_norm": 1.9765625,
+      "learning_rate": 2.5707768581321067e-05,
+      "loss": 1.1866,
+      "step": 780
+    },
+    {
+      "epoch": 0.5056,
+      "grad_norm": 2.109375,
+      "learning_rate": 2.560125007604439e-05,
+      "loss": 1.152,
+      "step": 790
+    },
+    {
+      "epoch": 0.512,
+      "grad_norm": 2.984375,
+      "learning_rate": 2.5493652584481944e-05,
+      "loss": 1.1716,
+      "step": 800
+    },
+    {
+      "epoch": 0.5184,
+      "grad_norm": 2.3125,
+      "learning_rate": 2.53849870578156e-05,
+      "loss": 1.0954,
+      "step": 810
+    },
+    {
+      "epoch": 0.5248,
+      "grad_norm": 2.578125,
+      "learning_rate": 2.5275264555930956e-05,
+      "loss": 1.0561,
+      "step": 820
+    },
+    {
+      "epoch": 0.5312,
+      "grad_norm": 1.96875,
+      "learning_rate": 2.5164496246291653e-05,
+      "loss": 1.1518,
+      "step": 830
+    },
+    {
+      "epoch": 0.5376,
+      "grad_norm": 2.609375,
+      "learning_rate": 2.505269340280277e-05,
+      "loss": 1.1361,
+      "step": 840
+    },
+    {
+      "epoch": 0.544,
+      "grad_norm": 2.53125,
+      "learning_rate": 2.4939867404663354e-05,
+      "loss": 1.1038,
+      "step": 850
+    },
+    {
+      "epoch": 0.5504,
+      "grad_norm": 2.546875,
+      "learning_rate": 2.4826029735208273e-05,
+      "loss": 1.2142,
+      "step": 860
+    },
+    {
+      "epoch": 0.5568,
+      "grad_norm": 1.7421875,
+      "learning_rate": 2.471119198073946e-05,
+      "loss": 1.139,
+      "step": 870
+    },
+    {
+      "epoch": 0.5632,
+      "grad_norm": 2.5625,
+      "learning_rate": 2.4595365829346638e-05,
+      "loss": 1.0655,
+      "step": 880
+    },
+    {
+      "epoch": 0.5696,
+      "grad_norm": 2.65625,
+      "learning_rate": 2.447856306971775e-05,
+      "loss": 1.0564,
+      "step": 890
+    },
+    {
+      "epoch": 0.576,
+      "grad_norm": 2.9375,
+      "learning_rate": 2.436079558993909e-05,
+      "loss": 1.1195,
+      "step": 900
+    },
+    {
+      "epoch": 0.5824,
+      "grad_norm": 2.921875,
+      "learning_rate": 2.424207537628537e-05,
+      "loss": 1.0753,
+      "step": 910
+    },
+    {
+      "epoch": 0.5888,
+      "grad_norm": 1.7421875,
+      "learning_rate": 2.4122414511999727e-05,
+      "loss": 1.1333,
+      "step": 920
+    },
+    {
+      "epoch": 0.5952,
+      "grad_norm": 2.46875,
+      "learning_rate": 2.4001825176063943e-05,
+      "loss": 1.0526,
+      "step": 930
+    },
+    {
+      "epoch": 0.6016,
+      "grad_norm": 2.15625,
+      "learning_rate": 2.3880319641958862e-05,
+      "loss": 1.0355,
+      "step": 940
+    },
+    {
+      "epoch": 0.608,
+      "grad_norm": 2.171875,
+      "learning_rate": 2.375791027641521e-05,
+      "loss": 1.1634,
+      "step": 950
+    },
+    {
+      "epoch": 0.6144,
+      "grad_norm": 2.1875,
+      "learning_rate": 2.3634609538154882e-05,
+      "loss": 1.0915,
+      "step": 960
+    },
+    {
+      "epoch": 0.6208,
+      "grad_norm": 2.203125,
+      "learning_rate": 2.3510429976622975e-05,
+      "loss": 1.0805,
+      "step": 970
+    },
+    {
+      "epoch": 0.6272,
+      "grad_norm": 1.9375,
+      "learning_rate": 2.338538423071045e-05,
+      "loss": 1.0966,
+      "step": 980
+    },
+    {
+      "epoch": 0.6336,
+      "grad_norm": 1.765625,
+      "learning_rate": 2.32594850274678e-05,
+      "loss": 1.0045,
+      "step": 990
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 2.890625,
+      "learning_rate": 2.3132745180809675e-05,
+      "loss": 1.0359,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6464,
+      "grad_norm": 1.8984375,
+      "learning_rate": 2.3005177590210713e-05,
+      "loss": 1.1221,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6528,
+      "grad_norm": 1.828125,
+      "learning_rate": 2.287679523939261e-05,
+      "loss": 1.1291,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6592,
+      "grad_norm": 2.046875,
+      "learning_rate": 2.2747611195002687e-05,
+      "loss": 1.0904,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6656,
+      "grad_norm": 1.890625,
+      "learning_rate": 2.2617638605283966e-05,
+      "loss": 1.1363,
+      "step": 1040
+    },
+    {
+      "epoch": 0.672,
+      "grad_norm": 2.046875,
+      "learning_rate": 2.248689069873693e-05,
+      "loss": 1.1025,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6784,
+      "grad_norm": 2.75,
+      "learning_rate": 2.235538078277317e-05,
+      "loss": 1.096,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6848,
+      "grad_norm": 1.8515625,
+      "learning_rate": 2.222312224236093e-05,
+      "loss": 1.084,
+      "step": 1070
+    },
+    {
+      "epoch": 0.6912,
+      "grad_norm": 2.296875,
+      "learning_rate": 2.2090128538662844e-05,
+      "loss": 1.1673,
+      "step": 1080
+    },
+    {
+      "epoch": 0.6976,
+      "grad_norm": 2.328125,
+      "learning_rate": 2.1956413207665835e-05,
+      "loss": 1.1464,
+      "step": 1090
+    },
+    {
+      "epoch": 0.704,
+      "grad_norm": 2.34375,
+      "learning_rate": 2.1821989858803438e-05,
+      "loss": 1.0856,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7104,
+      "grad_norm": 1.59375,
+      "learning_rate": 2.1686872173570655e-05,
+      "loss": 1.1303,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7168,
+      "grad_norm": 1.9296875,
+      "learning_rate": 2.1551073904131452e-05,
+      "loss": 1.1097,
+      "step": 1120
+    },
+    {
+      "epoch": 0.7232,
+      "grad_norm": 2.59375,
+      "learning_rate": 2.141460887191909e-05,
+      "loss": 1.1261,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7296,
+      "grad_norm": 1.9921875,
+      "learning_rate": 2.1277490966229375e-05,
+      "loss": 1.1666,
+      "step": 1140
+    },
+    {
+      "epoch": 0.736,
+      "grad_norm": 2.078125,
+      "learning_rate": 2.1139734142807027e-05,
+      "loss": 1.1462,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7424,
+      "grad_norm": 2.234375,
+      "learning_rate": 2.100135242242528e-05,
+      "loss": 1.0577,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7488,
+      "grad_norm": 1.9765625,
+      "learning_rate": 2.0862359889458835e-05,
+      "loss": 1.1435,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7552,
+      "grad_norm": 1.640625,
+      "learning_rate": 2.07227706904504e-05,
+      "loss": 1.101,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7616,
+      "grad_norm": 2.59375,
+      "learning_rate": 2.0582599032670846e-05,
+      "loss": 1.0775,
+      "step": 1190
+    },
+    {
+      "epoch": 0.768,
+      "grad_norm": 1.828125,
+      "learning_rate": 2.0441859182673195e-05,
+      "loss": 1.0275,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7744,
+      "grad_norm": 1.9296875,
+      "learning_rate": 2.0300565464840613e-05,
+      "loss": 1.1269,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7808,
+      "grad_norm": 1.7734375,
+      "learning_rate": 2.0158732259928453e-05,
+      "loss": 0.993,
+      "step": 1220
+    },
+    {
+      "epoch": 0.7872,
+      "grad_norm": 2.015625,
+      "learning_rate": 2.0016374003600604e-05,
+      "loss": 1.126,
+      "step": 1230
+    },
+    {
+      "epoch": 0.7936,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.9873505184960255e-05,
+      "loss": 1.0847,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 2.125,
+      "learning_rate": 1.9730140345075187e-05,
+      "loss": 1.0819,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8064,
+      "grad_norm": 1.9375,
+      "learning_rate": 1.9586294075497825e-05,
+      "loss": 1.0683,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8128,
+      "grad_norm": 1.8046875,
+      "learning_rate": 1.9441981016780074e-05,
+      "loss": 1.1951,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8192,
+      "grad_norm": 1.7578125,
+      "learning_rate": 1.929721585698326e-05,
+      "loss": 1.0735,
+      "step": 1280
+    },
+    {
+      "epoch": 0.8256,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.9152013330183175e-05,
+      "loss": 1.0937,
+      "step": 1290
+    },
+    {
+      "epoch": 0.832,
+      "grad_norm": 1.78125,
+      "learning_rate": 1.9006388214970442e-05,
+      "loss": 1.1384,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8384,
+      "grad_norm": 1.84375,
+      "learning_rate": 1.8860355332946405e-05,
+      "loss": 1.0922,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8448,
+      "grad_norm": 1.890625,
+      "learning_rate": 1.871392954721454e-05,
+      "loss": 1.1578,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8512,
+      "grad_norm": 1.84375,
+      "learning_rate": 1.856712576086774e-05,
+      "loss": 1.1425,
+      "step": 1330
+    },
+    {
+      "epoch": 0.8576,
+      "grad_norm": 2.21875,
+      "learning_rate": 1.8419958915471483e-05,
+      "loss": 1.0089,
+      "step": 1340
+    },
+    {
+      "epoch": 0.864,
+      "grad_norm": 1.8828125,
+      "learning_rate": 1.8272443989543102e-05,
+      "loss": 1.1178,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8704,
+      "grad_norm": 1.7890625,
+      "learning_rate": 1.8124595997027255e-05,
+      "loss": 1.1146,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8768,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.7976429985767842e-05,
+      "loss": 1.1538,
+      "step": 1370
+    },
+    {
+      "epoch": 0.8832,
+      "grad_norm": 1.9140625,
+      "learning_rate": 1.782796103597645e-05,
+      "loss": 1.1488,
+      "step": 1380
+    },
+    {
+      "epoch": 0.8896,
+      "grad_norm": 2.1875,
+      "learning_rate": 1.7679204258697482e-05,
+      "loss": 1.0986,
+      "step": 1390
+    },
+    {
+      "epoch": 0.896,
+      "grad_norm": 2.328125,
+      "learning_rate": 1.7530174794270187e-05,
+      "loss": 1.1119,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9024,
+      "grad_norm": 2.5625,
+      "learning_rate": 1.738088781078768e-05,
+      "loss": 1.0861,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9088,
+      "grad_norm": 1.84375,
+      "learning_rate": 1.723135850255315e-05,
+      "loss": 1.0417,
+      "step": 1420
+    },
+    {
+      "epoch": 0.9152,
+      "grad_norm": 1.78125,
+      "learning_rate": 1.7081602088533396e-05,
+      "loss": 1.0948,
+      "step": 1430
+    },
+    {
+      "epoch": 0.9216,
+      "grad_norm": 1.984375,
+      "learning_rate": 1.693163381080984e-05,
+      "loss": 1.1347,
+      "step": 1440
+    },
+    {
+      "epoch": 0.928,
+      "grad_norm": 1.6796875,
+      "learning_rate": 1.6781468933027233e-05,
+      "loss": 1.0932,
+      "step": 1450
+    },
+    {
+      "epoch": 0.9344,
+      "grad_norm": 1.96875,
+      "learning_rate": 1.6631122738840098e-05,
+      "loss": 1.1218,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9408,
+      "grad_norm": 2.671875,
+      "learning_rate": 1.6480610530357198e-05,
+      "loss": 1.0991,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9472,
+      "grad_norm": 1.984375,
+      "learning_rate": 1.6329947626584088e-05,
+      "loss": 1.0969,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9536,
+      "grad_norm": 1.75,
+      "learning_rate": 1.617914936186395e-05,
+      "loss": 1.0271,
+      "step": 1490
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 1.890625,
+      "learning_rate": 1.6028231084316914e-05,
+      "loss": 1.0578,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9664,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.5877208154277893e-05,
+      "loss": 1.033,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9728,
+      "grad_norm": 1.8515625,
+      "learning_rate": 1.5726095942733252e-05,
+      "loss": 1.0599,
+      "step": 1520
+    },
+    {
+      "epoch": 0.9792,
+      "grad_norm": 3.4375,
+      "learning_rate": 1.5574909829756355e-05,
+      "loss": 1.1695,
+      "step": 1530
+    },
+    {
+      "epoch": 0.9856,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.5423665202942176e-05,
+      "loss": 1.0135,
+      "step": 1540
+    },
+    {
+      "epoch": 0.992,
+      "grad_norm": 2.03125,
+      "learning_rate": 1.5272377455841203e-05,
+      "loss": 1.0969,
+      "step": 1550
+    },
+    {
+      "epoch": 0.9984,
+      "grad_norm": 2.09375,
+      "learning_rate": 1.5121061986392642e-05,
+      "loss": 1.1522,
+      "step": 1560
+    },
+    {
+      "epoch": 1.00448,
+      "grad_norm": 2.6875,
+      "learning_rate": 1.4969734195357282e-05,
+      "loss": 1.0479,
+      "step": 1570
+    },
+    {
+      "epoch": 1.01088,
+      "grad_norm": 1.78125,
+      "learning_rate": 1.481840948474997e-05,
+      "loss": 1.01,
+      "step": 1580
+    },
+    {
+      "epoch": 1.01728,
+      "grad_norm": 1.7109375,
+      "learning_rate": 1.466710325627205e-05,
+      "loss": 1.0202,
+      "step": 1590
+    },
+    {
+      "epoch": 1.02368,
+      "grad_norm": 1.984375,
+      "learning_rate": 1.4515830909743756e-05,
+      "loss": 1.0148,
+      "step": 1600
+    },
+    {
+      "epoch": 1.03008,
+      "grad_norm": 1.8125,
+      "learning_rate": 1.436460784153685e-05,
+      "loss": 0.8966,
+      "step": 1610
+    },
+    {
+      "epoch": 1.03648,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.4213449443007595e-05,
+      "loss": 0.9647,
+      "step": 1620
+    },
+    {
+      "epoch": 1.04288,
+      "grad_norm": 2.140625,
+      "learning_rate": 1.4062371098930216e-05,
+      "loss": 0.9537,
+      "step": 1630
+    },
+    {
+      "epoch": 1.04928,
+      "grad_norm": 1.890625,
+      "learning_rate": 1.3911388185931075e-05,
+      "loss": 0.95,
+      "step": 1640
+    },
+    {
+      "epoch": 1.05568,
+      "grad_norm": 1.8828125,
+      "learning_rate": 1.3760516070923629e-05,
+      "loss": 1.0116,
+      "step": 1650
+    },
+    {
+      "epoch": 1.06208,
+      "grad_norm": 2.296875,
+      "learning_rate": 1.360977010954442e-05,
+      "loss": 1.01,
+      "step": 1660
+    },
+    {
+      "epoch": 1.06848,
+      "grad_norm": 1.8203125,
+      "learning_rate": 1.3459165644590173e-05,
+      "loss": 1.0077,
+      "step": 1670
+    },
+    {
+      "epoch": 1.07488,
+      "grad_norm": 1.5546875,
+      "learning_rate": 1.3308718004456223e-05,
+      "loss": 1.0607,
+      "step": 1680
+    },
+    {
+      "epoch": 1.08128,
+      "grad_norm": 1.84375,
+      "learning_rate": 1.3158442501576419e-05,
+      "loss": 0.9339,
+      "step": 1690
+    },
+    {
+      "epoch": 1.08768,
+      "grad_norm": 2.296875,
+      "learning_rate": 1.3008354430864615e-05,
+      "loss": 0.9697,
+      "step": 1700
+    },
+    {
+      "epoch": 1.09408,
+      "grad_norm": 2.5,
+      "learning_rate": 1.2858469068157982e-05,
+      "loss": 0.9857,
+      "step": 1710
+    },
+    {
+      "epoch": 1.10048,
+      "grad_norm": 1.671875,
+      "learning_rate": 1.270880166866224e-05,
+      "loss": 1.0362,
+      "step": 1720
+    },
+    {
+      "epoch": 1.10688,
+      "grad_norm": 2.125,
+      "learning_rate": 1.2559367465398994e-05,
+      "loss": 1.0572,
+      "step": 1730
+    },
+    {
+      "epoch": 1.11328,
+      "grad_norm": 2.265625,
+      "learning_rate": 1.2410181667655342e-05,
+      "loss": 1.0424,
+      "step": 1740
+    },
+    {
+      "epoch": 1.11968,
+      "grad_norm": 1.921875,
+      "learning_rate": 1.2261259459435866e-05,
+      "loss": 0.9826,
+      "step": 1750
+    },
+    {
+      "epoch": 1.12608,
+      "grad_norm": 1.7578125,
+      "learning_rate": 1.2112615997917243e-05,
+      "loss": 1.06,
+      "step": 1760
+    },
+    {
+      "epoch": 1.13248,
+      "grad_norm": 1.6953125,
+      "learning_rate": 1.1964266411905538e-05,
+      "loss": 0.9296,
+      "step": 1770
+    },
+    {
+      "epoch": 1.13888,
+      "grad_norm": 2.609375,
+      "learning_rate": 1.1816225800296402e-05,
+      "loss": 0.9885,
+      "step": 1780
+    },
+    {
+      "epoch": 1.14528,
+      "grad_norm": 2.0625,
+      "learning_rate": 1.166850923053836e-05,
+      "loss": 1.0868,
+      "step": 1790
+    },
+    {
+      "epoch": 1.15168,
+      "grad_norm": 1.7265625,
+      "learning_rate": 1.1521131737099206e-05,
+      "loss": 1.0495,
+      "step": 1800
+    },
+    {
+      "epoch": 1.15808,
+      "grad_norm": 1.8828125,
+      "learning_rate": 1.1374108319935852e-05,
+      "loss": 0.9584,
+      "step": 1810
+    },
+    {
+      "epoch": 1.16448,
+      "grad_norm": 2.0625,
+      "learning_rate": 1.1227453942967615e-05,
+      "loss": 0.9654,
+      "step": 1820
+    },
+    {
+      "epoch": 1.17088,
+      "grad_norm": 1.6796875,
+      "learning_rate": 1.1081183532553197e-05,
+      "loss": 1.039,
+      "step": 1830
+    },
+    {
+      "epoch": 1.17728,
+      "grad_norm": 1.75,
+      "learning_rate": 1.0935311975971545e-05,
+      "loss": 1.0529,
+      "step": 1840
+    },
+    {
+      "epoch": 1.18368,
+      "grad_norm": 1.59375,
+      "learning_rate": 1.0789854119906566e-05,
+      "loss": 0.9898,
+      "step": 1850
+    },
+    {
+      "epoch": 1.19008,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.06448247689361e-05,
+      "loss": 0.9998,
+      "step": 1860
+    },
+    {
+      "epoch": 1.19648,
+      "grad_norm": 1.9453125,
+      "learning_rate": 1.050023868402509e-05,
+      "loss": 0.9979,
+      "step": 1870
+    },
+    {
+      "epoch": 1.20288,
+      "grad_norm": 2.078125,
+      "learning_rate": 1.035611058102322e-05,
+      "loss": 0.9595,
+      "step": 1880
+    },
+    {
+      "epoch": 1.20928,
+      "grad_norm": 1.6953125,
+      "learning_rate": 1.0212455129167196e-05,
+      "loss": 1.0216,
+      "step": 1890
+    },
+    {
+      "epoch": 1.21568,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.0069286949587653e-05,
+      "loss": 0.9082,
+      "step": 1900
+    },
+    {
+      "epoch": 1.22208,
+      "grad_norm": 2.0625,
+      "learning_rate": 9.926620613821094e-06,
+      "loss": 0.9402,
+      "step": 1910
+    },
+    {
+      "epoch": 1.22848,
+      "grad_norm": 3.015625,
+      "learning_rate": 9.784470642326767e-06,
+      "loss": 1.0484,
+      "step": 1920
+    },
+    {
+      "epoch": 1.23488,
+      "grad_norm": 2.359375,
+      "learning_rate": 9.642851503008788e-06,
+      "loss": 1.0093,
+      "step": 1930
+    },
+    {
+      "epoch": 1.24128,
+      "grad_norm": 1.6640625,
+      "learning_rate": 9.50177760974366e-06,
+      "loss": 0.9979,
+      "step": 1940
+    },
+    {
+      "epoch": 1.24768,
+      "grad_norm": 2.265625,
+      "learning_rate": 9.361263320913159e-06,
+      "loss": 0.957,
+      "step": 1950
+    },
+    {
+      "epoch": 1.25408,
+      "grad_norm": 2.140625,
+      "learning_rate": 9.22132293794303e-06,
+      "loss": 1.0589,
+      "step": 1960
+    },
+    {
+      "epoch": 1.26048,
+      "grad_norm": 1.96875,
+      "learning_rate": 9.081970703847344e-06,
+      "loss": 1.0065,
+      "step": 1970
+    },
+    {
+      "epoch": 1.26688,
+      "grad_norm": 2.421875,
+      "learning_rate": 8.943220801778871e-06,
+      "loss": 0.9914,
+      "step": 1980
+    },
+    {
+      "epoch": 1.27328,
+      "grad_norm": 1.8984375,
+      "learning_rate": 8.805087353585561e-06,
+      "loss": 0.9532,
+      "step": 1990
+    },
+    {
+      "epoch": 1.27968,
+      "grad_norm": 2.1875,
+      "learning_rate": 8.667584418373177e-06,
+      "loss": 0.961,
+      "step": 2000
+    },
+    {
+      "epoch": 1.2860800000000001,
+      "grad_norm": 1.984375,
+      "learning_rate": 8.530725991074432e-06,
+      "loss": 0.9762,
+      "step": 2010
+    },
+    {
+      "epoch": 1.29248,
+      "grad_norm": 2.171875,
+      "learning_rate": 8.39452600102455e-06,
+      "loss": 1.045,
+      "step": 2020
+    },
+    {
+      "epoch": 1.29888,
+      "grad_norm": 2.15625,
+      "learning_rate": 8.25899831054357e-06,
+      "loss": 0.9369,
+      "step": 2030
+    },
+    {
+      "epoch": 1.30528,
+      "grad_norm": 2.953125,
+      "learning_rate": 8.124156713525451e-06,
+      "loss": 0.9506,
+      "step": 2040
+    },
+    {
+      "epoch": 1.31168,
+      "grad_norm": 2.625,
+      "learning_rate": 7.990014934034132e-06,
+      "loss": 0.9693,
+      "step": 2050
+    },
+    {
+      "epoch": 1.31808,
+      "grad_norm": 1.703125,
+      "learning_rate": 7.856586624906731e-06,
+      "loss": 0.9546,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3244799999999999,
+      "grad_norm": 1.625,
+      "learning_rate": 7.723885366363937e-06,
+      "loss": 0.9469,
+      "step": 2070
+    },
+    {
+      "epoch": 1.33088,
+      "grad_norm": 2.390625,
+      "learning_rate": 7.591924664627853e-06,
+      "loss": 1.0108,
+      "step": 2080
+    },
+    {
+      "epoch": 1.33728,
+      "grad_norm": 2.203125,
+      "learning_rate": 7.460717950547333e-06,
+      "loss": 0.9485,
+      "step": 2090
+    },
+    {
+      "epoch": 1.34368,
+      "grad_norm": 2.875,
+      "learning_rate": 7.330278578230995e-06,
+      "loss": 0.9319,
+      "step": 2100
+    },
+    {
+      "epoch": 1.35008,
+      "grad_norm": 2.453125,
+      "learning_rate": 7.200619823688085e-06,
+      "loss": 0.9649,
+      "step": 2110
+    },
+    {
+      "epoch": 1.35648,
+      "grad_norm": 2.140625,
+      "learning_rate": 7.071754883477204e-06,
+      "loss": 0.8863,
+      "step": 2120
+    },
+    {
+      "epoch": 1.36288,
+      "grad_norm": 2.53125,
+      "learning_rate": 6.943696873363214e-06,
+      "loss": 0.9181,
+      "step": 2130
+    },
+    {
+      "epoch": 1.36928,
+      "grad_norm": 2.65625,
+      "learning_rate": 6.816458826982305e-06,
+      "loss": 0.9467,
+      "step": 2140
+    },
+    {
+      "epoch": 1.37568,
+      "grad_norm": 2.046875,
+      "learning_rate": 6.690053694515455e-06,
+      "loss": 0.9795,
+      "step": 2150
+    },
+    {
+      "epoch": 1.38208,
+      "grad_norm": 1.7578125,
+      "learning_rate": 6.564494341370372e-06,
+      "loss": 1.011,
+      "step": 2160
+    },
+    {
+      "epoch": 1.38848,
+      "grad_norm": 2.4375,
+      "learning_rate": 6.4397935468720524e-06,
+      "loss": 0.9757,
+      "step": 2170
+    },
+    {
+      "epoch": 1.3948800000000001,
+      "grad_norm": 1.8984375,
+      "learning_rate": 6.315964002962113e-06,
+      "loss": 1.0207,
+      "step": 2180
+    },
+    {
+      "epoch": 1.40128,
+      "grad_norm": 3.34375,
+      "learning_rate": 6.193018312907032e-06,
+      "loss": 0.9519,
+      "step": 2190
+    },
+    {
+      "epoch": 1.40768,
+      "grad_norm": 1.9609375,
+      "learning_rate": 6.070968990015372e-06,
+      "loss": 1.0672,
+      "step": 2200
+    },
+    {
+      "epoch": 1.41408,
+      "grad_norm": 2.421875,
+      "learning_rate": 5.949828456364229e-06,
+      "loss": 1.023,
+      "step": 2210
+    },
+    {
+      "epoch": 1.42048,
+      "grad_norm": 2.46875,
+      "learning_rate": 5.829609041534874e-06,
+      "loss": 1.0135,
+      "step": 2220
+    },
+    {
+      "epoch": 1.42688,
+      "grad_norm": 2.265625,
+      "learning_rate": 5.710322981357886e-06,
+      "loss": 0.9879,
+      "step": 2230
+    },
+    {
+      "epoch": 1.4332799999999999,
+      "grad_norm": 1.75,
+      "learning_rate": 5.591982416667802e-06,
+      "loss": 1.0821,
+      "step": 2240
+    },
+    {
+      "epoch": 1.43968,
+      "grad_norm": 2.3125,
+      "learning_rate": 5.474599392067413e-06,
+      "loss": 1.031,
+      "step": 2250
+    },
+    {
+      "epoch": 1.44608,
+      "grad_norm": 2.21875,
+      "learning_rate": 5.3581858547019095e-06,
+      "loss": 1.0214,
+      "step": 2260
+    },
+    {
+      "epoch": 1.45248,
+      "grad_norm": 2.609375,
+      "learning_rate": 5.242753653042872e-06,
+      "loss": 0.967,
+      "step": 2270
+    },
+    {
+      "epoch": 1.45888,
+      "grad_norm": 1.765625,
+      "learning_rate": 5.128314535682366e-06,
+      "loss": 0.9439,
+      "step": 2280
+    },
+    {
+      "epoch": 1.46528,
+      "grad_norm": 1.6640625,
+      "learning_rate": 5.014880150137175e-06,
+      "loss": 0.9451,
+      "step": 2290
+    },
+    {
+      "epoch": 1.47168,
+      "grad_norm": 2.203125,
+      "learning_rate": 4.902462041663321e-06,
+      "loss": 1.0129,
+      "step": 2300
+    },
+    {
+      "epoch": 1.47808,
+      "grad_norm": 2.234375,
+      "learning_rate": 4.7910716520810185e-06,
+      "loss": 1.0104,
+      "step": 2310
+    },
+    {
+      "epoch": 1.48448,
+      "grad_norm": 1.6953125,
+      "learning_rate": 4.680720318610107e-06,
+      "loss": 0.9509,
+      "step": 2320
+    },
+    {
+      "epoch": 1.49088,
+      "grad_norm": 2.25,
+      "learning_rate": 4.571419272716175e-06,
+      "loss": 1.0345,
+      "step": 2330
+    },
+    {
+      "epoch": 1.49728,
+      "grad_norm": 2.390625,
+      "learning_rate": 4.463179638967428e-06,
+      "loss": 1.0448,
+      "step": 2340
+    },
+    {
+      "epoch": 1.5036800000000001,
+      "grad_norm": 2.828125,
+      "learning_rate": 4.356012433902436e-06,
+      "loss": 1.0301,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5100799999999999,
+      "grad_norm": 1.671875,
+      "learning_rate": 4.249928564908896e-06,
+      "loss": 1.0551,
+      "step": 2360
+    },
+    {
+      "epoch": 1.51648,
+      "grad_norm": 2.515625,
+      "learning_rate": 4.144938829113458e-06,
+      "loss": 1.0107,
+      "step": 2370
+    },
+    {
+      "epoch": 1.52288,
+      "grad_norm": 2.5625,
+      "learning_rate": 4.0410539122828275e-06,
+      "loss": 0.9923,
+      "step": 2380
+    },
+    {
+      "epoch": 1.52928,
+      "grad_norm": 2.09375,
+      "learning_rate": 3.938284387736164e-06,
+      "loss": 0.9977,
+      "step": 2390
+    },
+    {
+      "epoch": 1.5356800000000002,
+      "grad_norm": 2.046875,
+      "learning_rate": 3.8366407152689374e-06,
+      "loss": 1.0079,
+      "step": 2400
+    },
+    {
+      "epoch": 1.54208,
+      "grad_norm": 1.7890625,
+      "learning_rate": 3.7361332400883608e-06,
+      "loss": 1.0236,
+      "step": 2410
+    },
+    {
+      "epoch": 1.54848,
+      "grad_norm": 1.796875,
+      "learning_rate": 3.636772191760427e-06,
+      "loss": 0.9698,
+      "step": 2420
+    },
+    {
+      "epoch": 1.55488,
+      "grad_norm": 1.9453125,
+      "learning_rate": 3.53856768316878e-06,
+      "loss": 1.0645,
+      "step": 2430
+    },
+    {
+      "epoch": 1.56128,
+      "grad_norm": 2.640625,
+      "learning_rate": 3.441529709485424e-06,
+      "loss": 0.9716,
+      "step": 2440
+    },
+    {
+      "epoch": 1.56768,
+      "grad_norm": 2.609375,
+      "learning_rate": 3.3456681471534207e-06,
+      "loss": 0.9717,
+      "step": 2450
+    },
+    {
+      "epoch": 1.57408,
+      "grad_norm": 1.953125,
+      "learning_rate": 3.2509927528816896e-06,
+      "loss": 1.0421,
+      "step": 2460
+    },
+    {
+      "epoch": 1.58048,
+      "grad_norm": 1.8515625,
+      "learning_rate": 3.1575131626519553e-06,
+      "loss": 1.0386,
+      "step": 2470
+    },
+    {
+      "epoch": 1.5868799999999998,
+      "grad_norm": 1.8828125,
+      "learning_rate": 3.0652388907380187e-06,
+      "loss": 0.9578,
+      "step": 2480
+    },
+    {
+      "epoch": 1.59328,
+      "grad_norm": 2.515625,
+      "learning_rate": 2.9741793287373987e-06,
+      "loss": 0.9847,
+      "step": 2490
+    },
+    {
+      "epoch": 1.59968,
+      "grad_norm": 2.78125,
+      "learning_rate": 2.8843437446154625e-06,
+      "loss": 1.0303,
+      "step": 2500
+    },
+    {
+      "epoch": 1.60608,
+      "grad_norm": 2.34375,
+      "learning_rate": 2.795741281762152e-06,
+      "loss": 0.944,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6124800000000001,
+      "grad_norm": 1.7734375,
+      "learning_rate": 2.708380958061358e-06,
+      "loss": 1.0533,
+      "step": 2520
+    },
+    {
+      "epoch": 1.6188799999999999,
+      "grad_norm": 1.7265625,
+      "learning_rate": 2.6222716649730964e-06,
+      "loss": 0.9802,
+      "step": 2530
+    },
+    {
+      "epoch": 1.62528,
+      "grad_norm": 1.75,
+      "learning_rate": 2.5374221666285474e-06,
+      "loss": 0.9677,
+      "step": 2540
+    },
+    {
+      "epoch": 1.63168,
+      "grad_norm": 1.859375,
+      "learning_rate": 2.4538410989380394e-06,
+      "loss": 1.0376,
+      "step": 2550
+    },
+    {
+      "epoch": 1.63808,
+      "grad_norm": 2.546875,
+      "learning_rate": 2.3715369687121103e-06,
+      "loss": 0.9554,
+      "step": 2560
+    },
+    {
+      "epoch": 1.6444800000000002,
+      "grad_norm": 2.609375,
+      "learning_rate": 2.2905181527956694e-06,
+      "loss": 0.9778,
+      "step": 2570
+    },
+    {
+      "epoch": 1.65088,
+      "grad_norm": 2.125,
+      "learning_rate": 2.2107928972154264e-06,
+      "loss": 1.0311,
+      "step": 2580
+    },
+    {
+      "epoch": 1.65728,
+      "grad_norm": 1.9765625,
+      "learning_rate": 2.132369316340608e-06,
+      "loss": 0.9612,
+      "step": 2590
+    },
+    {
+      "epoch": 1.66368,
+      "grad_norm": 2.15625,
+      "learning_rate": 2.0552553920570862e-06,
+      "loss": 0.9553,
+      "step": 2600
+    },
+    {
+      "epoch": 1.67008,
+      "grad_norm": 2.609375,
+      "learning_rate": 1.9794589729550022e-06,
+      "loss": 0.9915,
+      "step": 2610
+    },
+    {
+      "epoch": 1.67648,
+      "grad_norm": 1.7734375,
+      "learning_rate": 1.9049877735299203e-06,
+      "loss": 0.9905,
+      "step": 2620
+    },
+    {
+      "epoch": 1.68288,
+      "grad_norm": 1.8125,
+      "learning_rate": 1.8318493733976672e-06,
+      "loss": 0.9517,
+      "step": 2630
+    },
+    {
+      "epoch": 1.6892800000000001,
+      "grad_norm": 1.6484375,
+      "learning_rate": 1.760051216522882e-06,
+      "loss": 1.0555,
+      "step": 2640
+    },
+    {
+      "epoch": 1.6956799999999999,
+      "grad_norm": 1.828125,
+      "learning_rate": 1.6896006104613759e-06,
+      "loss": 0.9605,
+      "step": 2650
+    },
+    {
+      "epoch": 1.70208,
+      "grad_norm": 1.984375,
+      "learning_rate": 1.6205047256163813e-06,
+      "loss": 0.9612,
+      "step": 2660
+    },
+    {
+      "epoch": 1.70848,
+      "grad_norm": 2.578125,
+      "learning_rate": 1.5527705945087422e-06,
+      "loss": 1.0219,
+      "step": 2670
+    },
+    {
+      "epoch": 1.71488,
+      "grad_norm": 2.25,
+      "learning_rate": 1.4864051110611565e-06,
+      "loss": 1.019,
+      "step": 2680
+    },
+    {
+      "epoch": 1.7212800000000001,
+      "grad_norm": 1.75,
+      "learning_rate": 1.4214150298965217e-06,
+      "loss": 1.0037,
+      "step": 2690
+    },
+    {
+      "epoch": 1.7276799999999999,
+      "grad_norm": 2.078125,
+      "learning_rate": 1.357806965650447e-06,
+      "loss": 0.9484,
+      "step": 2700
+    },
+    {
+      "epoch": 1.73408,
+      "grad_norm": 1.6796875,
+      "learning_rate": 1.2955873922980355e-06,
+      "loss": 1.052,
+      "step": 2710
+    },
+    {
+      "epoch": 1.74048,
+      "grad_norm": 1.953125,
+      "learning_rate": 1.2347626424949483e-06,
+      "loss": 1.0051,
+      "step": 2720
+    },
+    {
+      "epoch": 1.74688,
+      "grad_norm": 2.609375,
+      "learning_rate": 1.1753389069328862e-06,
+      "loss": 0.9812,
+      "step": 2730
+    },
+    {
+      "epoch": 1.75328,
+      "grad_norm": 2.1875,
+      "learning_rate": 1.1173222337095074e-06,
+      "loss": 0.9297,
+      "step": 2740
+    },
+    {
+      "epoch": 1.75968,
+      "grad_norm": 2.078125,
+      "learning_rate": 1.0607185277128435e-06,
+      "loss": 0.996,
+      "step": 2750
+    },
+    {
+      "epoch": 1.76608,
+      "grad_norm": 2.078125,
+      "learning_rate": 1.0055335500203278e-06,
+      "loss": 0.9412,
+      "step": 2760
+    },
+    {
+      "epoch": 1.77248,
+      "grad_norm": 1.6484375,
+      "learning_rate": 9.517729173124095e-07,
+      "loss": 1.0159,
+      "step": 2770
+    },
+    {
+      "epoch": 1.77888,
+      "grad_norm": 1.875,
+      "learning_rate": 8.994421013009191e-07,
+      "loss": 0.9885,
+      "step": 2780
+    },
+    {
+      "epoch": 1.78528,
+      "grad_norm": 2.265625,
+      "learning_rate": 8.485464281721416e-07,
+      "loss": 1.1004,
+      "step": 2790
+    },
+    {
+      "epoch": 1.79168,
+      "grad_norm": 1.828125,
+      "learning_rate": 7.990910780447347e-07,
+      "loss": 0.9443,
+      "step": 2800
+    },
+    {
+      "epoch": 1.7980800000000001,
+      "grad_norm": 2.734375,
+      "learning_rate": 7.510810844425009e-07,
+      "loss": 0.945,
+      "step": 2810
+    },
+    {
+      "epoch": 1.8044799999999999,
+      "grad_norm": 2.015625,
+      "learning_rate": 7.045213337820633e-07,
+      "loss": 1.1081,
+      "step": 2820
+    },
+    {
+      "epoch": 1.81088,
+      "grad_norm": 2.71875,
+      "learning_rate": 6.594165648755562e-07,
+      "loss": 1.0107,
+      "step": 2830
+    },
+    {
+      "epoch": 1.81728,
+      "grad_norm": 2.421875,
+      "learning_rate": 6.157713684482896e-07,
+      "loss": 1.0491,
+      "step": 2840
+    },
+    {
+      "epoch": 1.82368,
+      "grad_norm": 1.953125,
+      "learning_rate": 5.735901866715243e-07,
+      "loss": 0.9906,
+      "step": 2850
+    },
+    {
+      "epoch": 1.8300800000000002,
+      "grad_norm": 2.234375,
+      "learning_rate": 5.328773127103403e-07,
+      "loss": 1.0141,
+      "step": 2860
+    },
+    {
+      "epoch": 1.83648,
+      "grad_norm": 1.671875,
+      "learning_rate": 4.936368902866884e-07,
+      "loss": 0.9968,
+      "step": 2870
+    },
+    {
+      "epoch": 1.84288,
+      "grad_norm": 1.7734375,
+      "learning_rate": 4.5587291325764667e-07,
+      "loss": 0.9854,
+      "step": 2880
+    },
+    {
+      "epoch": 1.84928,
+      "grad_norm": 1.6484375,
+      "learning_rate": 4.195892252089267e-07,
+      "loss": 1.0206,
+      "step": 2890
+    },
+    {
+      "epoch": 1.85568,
+      "grad_norm": 2.3125,
+      "learning_rate": 3.8478951906367744e-07,
+      "loss": 0.9919,
+      "step": 2900
+    },
+    {
+      "epoch": 1.86208,
+      "grad_norm": 2.234375,
+      "learning_rate": 3.51477336706621e-07,
+      "loss": 1.0347,
+      "step": 2910
+    },
+    {
+      "epoch": 1.86848,
+      "grad_norm": 1.7421875,
+      "learning_rate": 3.196560686235611e-07,
+      "loss": 0.9982,
+      "step": 2920
+    },
+    {
+      "epoch": 1.87488,
+      "grad_norm": 2.21875,
+      "learning_rate": 2.893289535563093e-07,
+      "loss": 0.978,
+      "step": 2930
+    },
+    {
+      "epoch": 1.8812799999999998,
+      "grad_norm": 2.234375,
+      "learning_rate": 2.6049907817304155e-07,
+      "loss": 0.9441,
+      "step": 2940
+    },
+    {
+      "epoch": 1.88768,
+      "grad_norm": 3.03125,
+      "learning_rate": 2.331693767541432e-07,
+      "loss": 0.9926,
+      "step": 2950
+    },
+    {
+      "epoch": 1.89408,
+      "grad_norm": 1.6015625,
+      "learning_rate": 2.0734263089356276e-07,
+      "loss": 0.9828,
+      "step": 2960
+    },
+    {
+      "epoch": 1.90048,
+      "grad_norm": 2.703125,
+      "learning_rate": 1.830214692156934e-07,
+      "loss": 1.025,
+      "step": 2970
+    },
+    {
+      "epoch": 1.9068800000000001,
+      "grad_norm": 1.84375,
+      "learning_rate": 1.6020836710784692e-07,
+      "loss": 0.9925,
+      "step": 2980
+    },
+    {
+      "epoch": 1.9132799999999999,
+      "grad_norm": 1.9609375,
+      "learning_rate": 1.3890564646830194e-07,
+      "loss": 0.9416,
+      "step": 2990
+    },
+    {
+      "epoch": 1.91968,
+      "grad_norm": 1.921875,
+      "learning_rate": 1.1911547546998636e-07,
+      "loss": 1.0084,
+      "step": 3000
+    },
+    {
+      "epoch": 1.92608,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.008398683398054e-07,
+      "loss": 0.9642,
+      "step": 3010
+    },
+    {
+      "epoch": 1.93248,
+      "grad_norm": 2.34375,
+      "learning_rate": 8.408068515362411e-08,
+      "loss": 1.0143,
+      "step": 3020
+    },
+    {
+      "epoch": 1.9388800000000002,
+      "grad_norm": 2.125,
+      "learning_rate": 6.883963164696527e-08,
+      "loss": 0.9897,
+      "step": 3030
+    },
+    {
+      "epoch": 1.94528,
+      "grad_norm": 2.109375,
+      "learning_rate": 5.5118259041391784e-08,
+      "loss": 0.9477,
+      "step": 3040
+    },
+    {
+      "epoch": 1.95168,
+      "grad_norm": 1.9296875,
+      "learning_rate": 4.291796388662949e-08,
+      "loss": 1.0092,
+      "step": 3050
+    },
+    {
+      "epoch": 1.95808,
+      "grad_norm": 2.046875,
+      "learning_rate": 3.223998791842098e-08,
+      "loss": 1.0761,
+      "step": 3060
+    },
+    {
+      "epoch": 1.96448,
+      "grad_norm": 2.859375,
+      "learning_rate": 2.3085417932148246e-08,
+      "loss": 1.009,
+      "step": 3070
+    },
+    {
+      "epoch": 1.97088,
+      "grad_norm": 1.734375,
+      "learning_rate": 1.5455185672219595e-08,
+      "loss": 0.9905,
+      "step": 3080
+    },
+    {
+      "epoch": 1.97728,
+      "grad_norm": 2.34375,
+      "learning_rate": 9.350067737233791e-09,
+      "loss": 1.0572,
+      "step": 3090
+    },
+    {
+      "epoch": 1.98368,
+      "grad_norm": 2.609375,
+      "learning_rate": 4.7706855009416186e-09,
+      "loss": 0.9107,
+      "step": 3100
+    },
+    {
+      "epoch": 1.9900799999999998,
+      "grad_norm": 2.21875,
+      "learning_rate": 1.717505049003143e-09,
+      "loss": 1.0182,
+      "step": 3110
+    },
+    {
+      "epoch": 1.99648,
+      "grad_norm": 1.9609375,
+      "learning_rate": 1.9083713154732785e-10,
+      "loss": 0.9777,
+      "step": 3120
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3124,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.9759783028737065e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d6f4e87766ba870a680c3fd431812c531c3a7b25a0830b9595b1602dab95217
+size 5969