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 +1,32 @@
+{
+  "</answer>": 151670,
+  "</information>": 151672,
+  "</think>": 151666,
+  "</tool>": 151668,
+  "</tool_call>": 151658,
+  "<answer>": 151669,
+  "<information>": 151671,
+  "<think>": 151665,
+  "<tool>": 151667,
+  "<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 +1,28 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 70,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.51.3"
+}

latest

@@ -0,0 +1 @@
+global_step4500

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1674ae964de3cb7382f88b5adf2da60501a5733b2cc34f2bc3ebce0948cd9961
+size 4957560304

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cc00c62a0ce5af7f20e0421a3e24b46de6694d431ef33a63bf3007eda1fa40d
+size 1214366696

model.safetensors.index.json

@@ -0,0 +1,441 @@
+{
+  "metadata": {
+    "total_size": 6171877376
+  },
+  "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-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-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "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_proj.bias": "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.bias": "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-00001-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00001-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_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00001-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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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_proj.bias": "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_proj.bias": "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.bias": "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"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6223cfb5499da33858a9472a0bbac51ae00362d72a643acb50af8e2065967744
+size 11423384

tokenizer_config.json

@@ -0,0 +1,273 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<tool>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</tool>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151669": {
+      "content": "<answer>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151670": {
+      "content": "</answer>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151671": {
+      "content": "<information>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151672": {
+      "content": "</information>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,3184 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.8553299492385786,
+  "eval_steps": 500,
+  "global_step": 4500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006345177664974619,
+      "grad_norm": 12.304139137268066,
+      "learning_rate": 1.9027484143763215e-07,
+      "loss": 1.4092,
+      "step": 10
+    },
+    {
+      "epoch": 0.012690355329949238,
+      "grad_norm": 10.735240936279297,
+      "learning_rate": 4.0169133192389007e-07,
+      "loss": 1.3444,
+      "step": 20
+    },
+    {
+      "epoch": 0.01903553299492386,
+      "grad_norm": 4.4380784034729,
+      "learning_rate": 6.131078224101481e-07,
+      "loss": 1.2567,
+      "step": 30
+    },
+    {
+      "epoch": 0.025380710659898477,
+      "grad_norm": 3.0971062183380127,
+      "learning_rate": 8.245243128964061e-07,
+      "loss": 1.2201,
+      "step": 40
+    },
+    {
+      "epoch": 0.031725888324873094,
+      "grad_norm": 2.3528785705566406,
+      "learning_rate": 1.0359408033826639e-06,
+      "loss": 1.1005,
+      "step": 50
+    },
+    {
+      "epoch": 0.03807106598984772,
+      "grad_norm": 1.9325449466705322,
+      "learning_rate": 1.2473572938689219e-06,
+      "loss": 1.0258,
+      "step": 60
+    },
+    {
+      "epoch": 0.044416243654822336,
+      "grad_norm": 1.9481005668640137,
+      "learning_rate": 1.4587737843551796e-06,
+      "loss": 0.9549,
+      "step": 70
+    },
+    {
+      "epoch": 0.050761421319796954,
+      "grad_norm": 1.3744746446609497,
+      "learning_rate": 1.6701902748414379e-06,
+      "loss": 0.9397,
+      "step": 80
+    },
+    {
+      "epoch": 0.05710659898477157,
+      "grad_norm": 1.3208822011947632,
+      "learning_rate": 1.8816067653276956e-06,
+      "loss": 0.9581,
+      "step": 90
+    },
+    {
+      "epoch": 0.06345177664974619,
+      "grad_norm": 1.578454613685608,
+      "learning_rate": 2.0930232558139536e-06,
+      "loss": 0.8835,
+      "step": 100
+    },
+    {
+      "epoch": 0.06979695431472081,
+      "grad_norm": 1.7314599752426147,
+      "learning_rate": 2.3044397463002116e-06,
+      "loss": 0.877,
+      "step": 110
+    },
+    {
+      "epoch": 0.07614213197969544,
+      "grad_norm": 1.690652847290039,
+      "learning_rate": 2.5158562367864696e-06,
+      "loss": 0.8674,
+      "step": 120
+    },
+    {
+      "epoch": 0.08248730964467005,
+      "grad_norm": 1.4886319637298584,
+      "learning_rate": 2.7272727272727272e-06,
+      "loss": 0.8124,
+      "step": 130
+    },
+    {
+      "epoch": 0.08883248730964467,
+      "grad_norm": 1.5932313203811646,
+      "learning_rate": 2.9386892177589852e-06,
+      "loss": 0.8825,
+      "step": 140
+    },
+    {
+      "epoch": 0.09517766497461928,
+      "grad_norm": 1.7353770732879639,
+      "learning_rate": 3.1501057082452436e-06,
+      "loss": 0.8381,
+      "step": 150
+    },
+    {
+      "epoch": 0.10152284263959391,
+      "grad_norm": 1.5052095651626587,
+      "learning_rate": 3.3615221987315012e-06,
+      "loss": 0.8094,
+      "step": 160
+    },
+    {
+      "epoch": 0.10786802030456853,
+      "grad_norm": 1.5068026781082153,
+      "learning_rate": 3.5729386892177592e-06,
+      "loss": 0.8088,
+      "step": 170
+    },
+    {
+      "epoch": 0.11421319796954314,
+      "grad_norm": 1.3972314596176147,
+      "learning_rate": 3.7843551797040172e-06,
+      "loss": 0.7807,
+      "step": 180
+    },
+    {
+      "epoch": 0.12055837563451777,
+      "grad_norm": 1.4561253786087036,
+      "learning_rate": 3.995771670190275e-06,
+      "loss": 0.751,
+      "step": 190
+    },
+    {
+      "epoch": 0.12690355329949238,
+      "grad_norm": 1.1900990009307861,
+      "learning_rate": 4.207188160676533e-06,
+      "loss": 0.7526,
+      "step": 200
+    },
+    {
+      "epoch": 0.13324873096446702,
+      "grad_norm": 1.2069578170776367,
+      "learning_rate": 4.418604651162791e-06,
+      "loss": 0.737,
+      "step": 210
+    },
+    {
+      "epoch": 0.13959390862944163,
+      "grad_norm": 1.3006811141967773,
+      "learning_rate": 4.630021141649049e-06,
+      "loss": 0.757,
+      "step": 220
+    },
+    {
+      "epoch": 0.14593908629441624,
+      "grad_norm": 1.1366584300994873,
+      "learning_rate": 4.841437632135307e-06,
+      "loss": 0.7355,
+      "step": 230
+    },
+    {
+      "epoch": 0.15228426395939088,
+      "grad_norm": 1.0923043489456177,
+      "learning_rate": 5.052854122621564e-06,
+      "loss": 0.7273,
+      "step": 240
+    },
+    {
+      "epoch": 0.15862944162436549,
+      "grad_norm": 1.1340067386627197,
+      "learning_rate": 5.264270613107823e-06,
+      "loss": 0.7093,
+      "step": 250
+    },
+    {
+      "epoch": 0.1649746192893401,
+      "grad_norm": 1.0045281648635864,
+      "learning_rate": 5.47568710359408e-06,
+      "loss": 0.709,
+      "step": 260
+    },
+    {
+      "epoch": 0.1713197969543147,
+      "grad_norm": 1.3080400228500366,
+      "learning_rate": 5.687103594080339e-06,
+      "loss": 0.7142,
+      "step": 270
+    },
+    {
+      "epoch": 0.17766497461928935,
+      "grad_norm": 1.4830659627914429,
+      "learning_rate": 5.898520084566597e-06,
+      "loss": 0.7233,
+      "step": 280
+    },
+    {
+      "epoch": 0.18401015228426396,
+      "grad_norm": 1.295798897743225,
+      "learning_rate": 6.109936575052855e-06,
+      "loss": 0.7254,
+      "step": 290
+    },
+    {
+      "epoch": 0.19035532994923857,
+      "grad_norm": 1.1951725482940674,
+      "learning_rate": 6.321353065539113e-06,
+      "loss": 0.7008,
+      "step": 300
+    },
+    {
+      "epoch": 0.1967005076142132,
+      "grad_norm": 1.1962999105453491,
+      "learning_rate": 6.53276955602537e-06,
+      "loss": 0.6697,
+      "step": 310
+    },
+    {
+      "epoch": 0.20304568527918782,
+      "grad_norm": 1.0768781900405884,
+      "learning_rate": 6.744186046511628e-06,
+      "loss": 0.6688,
+      "step": 320
+    },
+    {
+      "epoch": 0.20939086294416243,
+      "grad_norm": 1.2655526399612427,
+      "learning_rate": 6.955602536997886e-06,
+      "loss": 0.7098,
+      "step": 330
+    },
+    {
+      "epoch": 0.21573604060913706,
+      "grad_norm": 1.1732734441757202,
+      "learning_rate": 7.167019027484144e-06,
+      "loss": 0.6961,
+      "step": 340
+    },
+    {
+      "epoch": 0.22208121827411167,
+      "grad_norm": 1.4146960973739624,
+      "learning_rate": 7.378435517970403e-06,
+      "loss": 0.6581,
+      "step": 350
+    },
+    {
+      "epoch": 0.22842639593908629,
+      "grad_norm": 1.0180368423461914,
+      "learning_rate": 7.58985200845666e-06,
+      "loss": 0.636,
+      "step": 360
+    },
+    {
+      "epoch": 0.23477157360406092,
+      "grad_norm": 1.1763561964035034,
+      "learning_rate": 7.801268498942918e-06,
+      "loss": 0.6695,
+      "step": 370
+    },
+    {
+      "epoch": 0.24111675126903553,
+      "grad_norm": 1.120521068572998,
+      "learning_rate": 8.012684989429176e-06,
+      "loss": 0.6658,
+      "step": 380
+    },
+    {
+      "epoch": 0.24746192893401014,
+      "grad_norm": 1.070609450340271,
+      "learning_rate": 8.224101479915433e-06,
+      "loss": 0.6528,
+      "step": 390
+    },
+    {
+      "epoch": 0.25380710659898476,
+      "grad_norm": 1.404994249343872,
+      "learning_rate": 8.435517970401692e-06,
+      "loss": 0.6525,
+      "step": 400
+    },
+    {
+      "epoch": 0.26015228426395937,
+      "grad_norm": 1.3568419218063354,
+      "learning_rate": 8.64693446088795e-06,
+      "loss": 0.6525,
+      "step": 410
+    },
+    {
+      "epoch": 0.26649746192893403,
+      "grad_norm": 1.3468185663223267,
+      "learning_rate": 8.858350951374208e-06,
+      "loss": 0.641,
+      "step": 420
+    },
+    {
+      "epoch": 0.27284263959390864,
+      "grad_norm": 1.0951420068740845,
+      "learning_rate": 9.069767441860465e-06,
+      "loss": 0.6453,
+      "step": 430
+    },
+    {
+      "epoch": 0.27918781725888325,
+      "grad_norm": 1.030259370803833,
+      "learning_rate": 9.281183932346723e-06,
+      "loss": 0.6138,
+      "step": 440
+    },
+    {
+      "epoch": 0.28553299492385786,
+      "grad_norm": 1.1757938861846924,
+      "learning_rate": 9.492600422832982e-06,
+      "loss": 0.6787,
+      "step": 450
+    },
+    {
+      "epoch": 0.2918781725888325,
+      "grad_norm": 1.3138433694839478,
+      "learning_rate": 9.70401691331924e-06,
+      "loss": 0.6633,
+      "step": 460
+    },
+    {
+      "epoch": 0.2982233502538071,
+      "grad_norm": 1.3092707395553589,
+      "learning_rate": 9.915433403805497e-06,
+      "loss": 0.6432,
+      "step": 470
+    },
+    {
+      "epoch": 0.30456852791878175,
+      "grad_norm": 1.2927078008651733,
+      "learning_rate": 9.999950938319974e-06,
+      "loss": 0.6266,
+      "step": 480
+    },
+    {
+      "epoch": 0.31091370558375636,
+      "grad_norm": 1.33150053024292,
+      "learning_rate": 9.999651120428776e-06,
+      "loss": 0.6427,
+      "step": 490
+    },
+    {
+      "epoch": 0.31725888324873097,
+      "grad_norm": 1.2657496929168701,
+      "learning_rate": 9.999078757459388e-06,
+      "loss": 0.6457,
+      "step": 500
+    },
+    {
+      "epoch": 0.3236040609137056,
+      "grad_norm": 1.6883960962295532,
+      "learning_rate": 9.998233880612932e-06,
+      "loss": 0.6137,
+      "step": 510
+    },
+    {
+      "epoch": 0.3299492385786802,
+      "grad_norm": 0.9815077781677246,
+      "learning_rate": 9.997116535946028e-06,
+      "loss": 0.6069,
+      "step": 520
+    },
+    {
+      "epoch": 0.3362944162436548,
+      "grad_norm": 1.3186026811599731,
+      "learning_rate": 9.99572678436828e-06,
+      "loss": 0.6024,
+      "step": 530
+    },
+    {
+      "epoch": 0.3426395939086294,
+      "grad_norm": 1.6290111541748047,
+      "learning_rate": 9.994064701638969e-06,
+      "loss": 0.6273,
+      "step": 540
+    },
+    {
+      "epoch": 0.3489847715736041,
+      "grad_norm": 1.3211804628372192,
+      "learning_rate": 9.992130378362908e-06,
+      "loss": 0.6068,
+      "step": 550
+    },
+    {
+      "epoch": 0.3553299492385787,
+      "grad_norm": 1.619232177734375,
+      "learning_rate": 9.989923919985512e-06,
+      "loss": 0.612,
+      "step": 560
+    },
+    {
+      "epoch": 0.3616751269035533,
+      "grad_norm": 1.0001276731491089,
+      "learning_rate": 9.987445446787049e-06,
+      "loss": 0.5687,
+      "step": 570
+    },
+    {
+      "epoch": 0.3680203045685279,
+      "grad_norm": 1.2668827772140503,
+      "learning_rate": 9.984695093876081e-06,
+      "loss": 0.5723,
+      "step": 580
+    },
+    {
+      "epoch": 0.3743654822335025,
+      "grad_norm": 1.1758859157562256,
+      "learning_rate": 9.981673011182098e-06,
+      "loss": 0.5963,
+      "step": 590
+    },
+    {
+      "epoch": 0.38071065989847713,
+      "grad_norm": 1.4700498580932617,
+      "learning_rate": 9.978379363447348e-06,
+      "loss": 0.5682,
+      "step": 600
+    },
+    {
+      "epoch": 0.3870558375634518,
+      "grad_norm": 1.7378568649291992,
+      "learning_rate": 9.974814330217858e-06,
+      "loss": 0.6286,
+      "step": 610
+    },
+    {
+      "epoch": 0.3934010152284264,
+      "grad_norm": 1.5732265710830688,
+      "learning_rate": 9.970978105833632e-06,
+      "loss": 0.5464,
+      "step": 620
+    },
+    {
+      "epoch": 0.399746192893401,
+      "grad_norm": 1.4477766752243042,
+      "learning_rate": 9.966870899418087e-06,
+      "loss": 0.5806,
+      "step": 630
+    },
+    {
+      "epoch": 0.40609137055837563,
+      "grad_norm": 1.5664384365081787,
+      "learning_rate": 9.96249293486662e-06,
+      "loss": 0.5868,
+      "step": 640
+    },
+    {
+      "epoch": 0.41243654822335024,
+      "grad_norm": 1.242577075958252,
+      "learning_rate": 9.957844450834418e-06,
+      "loss": 0.5943,
+      "step": 650
+    },
+    {
+      "epoch": 0.41878172588832485,
+      "grad_norm": 1.3932079076766968,
+      "learning_rate": 9.952925700723455e-06,
+      "loss": 0.5582,
+      "step": 660
+    },
+    {
+      "epoch": 0.4251269035532995,
+      "grad_norm": 1.4832308292388916,
+      "learning_rate": 9.947736952668667e-06,
+      "loss": 0.561,
+      "step": 670
+    },
+    {
+      "epoch": 0.43147208121827413,
+      "grad_norm": 1.8345366716384888,
+      "learning_rate": 9.942278489523338e-06,
+      "loss": 0.5459,
+      "step": 680
+    },
+    {
+      "epoch": 0.43781725888324874,
+      "grad_norm": 1.1875063180923462,
+      "learning_rate": 9.936550608843685e-06,
+      "loss": 0.5267,
+      "step": 690
+    },
+    {
+      "epoch": 0.44416243654822335,
+      "grad_norm": 1.4732545614242554,
+      "learning_rate": 9.930553622872631e-06,
+      "loss": 0.5814,
+      "step": 700
+    },
+    {
+      "epoch": 0.45050761421319796,
+      "grad_norm": 1.7493573427200317,
+      "learning_rate": 9.924287858522789e-06,
+      "loss": 0.5633,
+      "step": 710
+    },
+    {
+      "epoch": 0.45685279187817257,
+      "grad_norm": 1.4842727184295654,
+      "learning_rate": 9.917753657358638e-06,
+      "loss": 0.53,
+      "step": 720
+    },
+    {
+      "epoch": 0.4631979695431472,
+      "grad_norm": 1.6605039834976196,
+      "learning_rate": 9.910951375577907e-06,
+      "loss": 0.5231,
+      "step": 730
+    },
+    {
+      "epoch": 0.46954314720812185,
+      "grad_norm": 1.6541188955307007,
+      "learning_rate": 9.903881383992153e-06,
+      "loss": 0.5268,
+      "step": 740
+    },
+    {
+      "epoch": 0.47588832487309646,
+      "grad_norm": 1.8268778324127197,
+      "learning_rate": 9.89654406800655e-06,
+      "loss": 0.49,
+      "step": 750
+    },
+    {
+      "epoch": 0.48223350253807107,
+      "grad_norm": 1.4834731817245483,
+      "learning_rate": 9.88893982759888e-06,
+      "loss": 0.5045,
+      "step": 760
+    },
+    {
+      "epoch": 0.4885786802030457,
+      "grad_norm": 1.717140555381775,
+      "learning_rate": 9.881069077297724e-06,
+      "loss": 0.496,
+      "step": 770
+    },
+    {
+      "epoch": 0.4949238578680203,
+      "grad_norm": 1.0741287469863892,
+      "learning_rate": 9.872932246159873e-06,
+      "loss": 0.4679,
+      "step": 780
+    },
+    {
+      "epoch": 0.501269035532995,
+      "grad_norm": 1.2269752025604248,
+      "learning_rate": 9.864529777746929e-06,
+      "loss": 0.4772,
+      "step": 790
+    },
+    {
+      "epoch": 0.5076142131979695,
+      "grad_norm": 1.6613504886627197,
+      "learning_rate": 9.85586213010114e-06,
+      "loss": 0.5008,
+      "step": 800
+    },
+    {
+      "epoch": 0.5139593908629442,
+      "grad_norm": 1.2009035348892212,
+      "learning_rate": 9.846929775720411e-06,
+      "loss": 0.5038,
+      "step": 810
+    },
+    {
+      "epoch": 0.5203045685279187,
+      "grad_norm": 1.5814530849456787,
+      "learning_rate": 9.837733201532565e-06,
+      "loss": 0.5021,
+      "step": 820
+    },
+    {
+      "epoch": 0.5266497461928934,
+      "grad_norm": 1.6952024698257446,
+      "learning_rate": 9.82827290886879e-06,
+      "loss": 0.4845,
+      "step": 830
+    },
+    {
+      "epoch": 0.5329949238578681,
+      "grad_norm": 1.3526102304458618,
+      "learning_rate": 9.818549413436309e-06,
+      "loss": 0.4952,
+      "step": 840
+    },
+    {
+      "epoch": 0.5393401015228426,
+      "grad_norm": 1.7655881643295288,
+      "learning_rate": 9.80856324529027e-06,
+      "loss": 0.4678,
+      "step": 850
+    },
+    {
+      "epoch": 0.5456852791878173,
+      "grad_norm": 1.391158103942871,
+      "learning_rate": 9.79831494880486e-06,
+      "loss": 0.4702,
+      "step": 860
+    },
+    {
+      "epoch": 0.5520304568527918,
+      "grad_norm": 1.3191405534744263,
+      "learning_rate": 9.787805082643604e-06,
+      "loss": 0.4394,
+      "step": 870
+    },
+    {
+      "epoch": 0.5583756345177665,
+      "grad_norm": 1.537750005722046,
+      "learning_rate": 9.777034219728943e-06,
+      "loss": 0.4172,
+      "step": 880
+    },
+    {
+      "epoch": 0.5647208121827412,
+      "grad_norm": 1.953177809715271,
+      "learning_rate": 9.76600294721098e-06,
+      "loss": 0.4846,
+      "step": 890
+    },
+    {
+      "epoch": 0.5710659898477157,
+      "grad_norm": 1.3089863061904907,
+      "learning_rate": 9.754711866435477e-06,
+      "loss": 0.414,
+      "step": 900
+    },
+    {
+      "epoch": 0.5774111675126904,
+      "grad_norm": 1.6026610136032104,
+      "learning_rate": 9.743161592911088e-06,
+      "loss": 0.5243,
+      "step": 910
+    },
+    {
+      "epoch": 0.583756345177665,
+      "grad_norm": 1.7620460987091064,
+      "learning_rate": 9.731352756275781e-06,
+      "loss": 0.4181,
+      "step": 920
+    },
+    {
+      "epoch": 0.5901015228426396,
+      "grad_norm": 1.6068378686904907,
+      "learning_rate": 9.719286000262533e-06,
+      "loss": 0.3713,
+      "step": 930
+    },
+    {
+      "epoch": 0.5964467005076142,
+      "grad_norm": 2.3091704845428467,
+      "learning_rate": 9.706961982664239e-06,
+      "loss": 0.4562,
+      "step": 940
+    },
+    {
+      "epoch": 0.6027918781725888,
+      "grad_norm": 2.353106737136841,
+      "learning_rate": 9.69438137529784e-06,
+      "loss": 0.4361,
+      "step": 950
+    },
+    {
+      "epoch": 0.6091370558375635,
+      "grad_norm": 1.599411129951477,
+      "learning_rate": 9.681544863967713e-06,
+      "loss": 0.4496,
+      "step": 960
+    },
+    {
+      "epoch": 0.6154822335025381,
+      "grad_norm": 1.5869901180267334,
+      "learning_rate": 9.668453148428282e-06,
+      "loss": 0.4046,
+      "step": 970
+    },
+    {
+      "epoch": 0.6218274111675127,
+      "grad_norm": 1.7548712491989136,
+      "learning_rate": 9.65510694234587e-06,
+      "loss": 0.3627,
+      "step": 980
+    },
+    {
+      "epoch": 0.6281725888324873,
+      "grad_norm": 1.3313032388687134,
+      "learning_rate": 9.641506973259798e-06,
+      "loss": 0.4176,
+      "step": 990
+    },
+    {
+      "epoch": 0.6345177664974619,
+      "grad_norm": 3.056716203689575,
+      "learning_rate": 9.627653982542722e-06,
+      "loss": 0.4283,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6408629441624365,
+      "grad_norm": 1.8358234167099,
+      "learning_rate": 9.613548725360224e-06,
+      "loss": 0.4217,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6472081218274112,
+      "grad_norm": 1.823522686958313,
+      "learning_rate": 9.599191970629638e-06,
+      "loss": 0.437,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6535532994923858,
+      "grad_norm": 1.779383897781372,
+      "learning_rate": 9.584584500978144e-06,
+      "loss": 0.3995,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6598984771573604,
+      "grad_norm": 1.7531787157058716,
+      "learning_rate": 9.569727112700093e-06,
+      "loss": 0.4449,
+      "step": 1040
+    },
+    {
+      "epoch": 0.666243654822335,
+      "grad_norm": 2.1453044414520264,
+      "learning_rate": 9.55462061571361e-06,
+      "loss": 0.3754,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6725888324873096,
+      "grad_norm": 1.6521024703979492,
+      "learning_rate": 9.539265833516434e-06,
+      "loss": 0.419,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6789340101522843,
+      "grad_norm": 1.616896152496338,
+      "learning_rate": 9.523663603141032e-06,
+      "loss": 0.4076,
+      "step": 1070
+    },
+    {
+      "epoch": 0.6852791878172588,
+      "grad_norm": 1.219354510307312,
+      "learning_rate": 9.507814775108971e-06,
+      "loss": 0.4092,
+      "step": 1080
+    },
+    {
+      "epoch": 0.6916243654822335,
+      "grad_norm": 22.454200744628906,
+      "learning_rate": 9.49172021338455e-06,
+      "loss": 0.4034,
+      "step": 1090
+    },
+    {
+      "epoch": 0.6979695431472082,
+      "grad_norm": 1.8505566120147705,
+      "learning_rate": 9.475380795327702e-06,
+      "loss": 0.3824,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7043147208121827,
+      "grad_norm": 1.492254376411438,
+      "learning_rate": 9.458797411646176e-06,
+      "loss": 0.3405,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7106598984771574,
+      "grad_norm": 1.774132251739502,
+      "learning_rate": 9.441970966346965e-06,
+      "loss": 0.3425,
+      "step": 1120
+    },
+    {
+      "epoch": 0.7170050761421319,
+      "grad_norm": 1.2463436126708984,
+      "learning_rate": 9.424902376687045e-06,
+      "loss": 0.3594,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7233502538071066,
+      "grad_norm": 1.515215277671814,
+      "learning_rate": 9.407592573123359e-06,
+      "loss": 0.359,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7296954314720813,
+      "grad_norm": 3.103351593017578,
+      "learning_rate": 9.390042499262102e-06,
+      "loss": 0.3554,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7360406091370558,
+      "grad_norm": 1.8471239805221558,
+      "learning_rate": 9.372253111807276e-06,
+      "loss": 0.3251,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7423857868020305,
+      "grad_norm": 1.8411760330200195,
+      "learning_rate": 9.354225380508548e-06,
+      "loss": 0.3233,
+      "step": 1170
+    },
+    {
+      "epoch": 0.748730964467005,
+      "grad_norm": 1.499944806098938,
+      "learning_rate": 9.33596028810838e-06,
+      "loss": 0.3718,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7550761421319797,
+      "grad_norm": 2.158557653427124,
+      "learning_rate": 9.317458830288446e-06,
+      "loss": 0.3463,
+      "step": 1190
+    },
+    {
+      "epoch": 0.7614213197969543,
+      "grad_norm": 1.5045950412750244,
+      "learning_rate": 9.29872201561538e-06,
+      "loss": 0.3682,
+      "step": 1200
+    },
+    {
+      "epoch": 0.7677664974619289,
+      "grad_norm": 1.9903945922851562,
+      "learning_rate": 9.279750865485772e-06,
+      "loss": 0.3149,
+      "step": 1210
+    },
+    {
+      "epoch": 0.7741116751269036,
+      "grad_norm": 1.7139513492584229,
+      "learning_rate": 9.260546414070504e-06,
+      "loss": 0.2947,
+      "step": 1220
+    },
+    {
+      "epoch": 0.7804568527918782,
+      "grad_norm": 2.4074273109436035,
+      "learning_rate": 9.241109708258362e-06,
+      "loss": 0.3451,
+      "step": 1230
+    },
+    {
+      "epoch": 0.7868020304568528,
+      "grad_norm": 1.736325740814209,
+      "learning_rate": 9.221441807598981e-06,
+      "loss": 0.3156,
+      "step": 1240
+    },
+    {
+      "epoch": 0.7931472081218274,
+      "grad_norm": 1.722331166267395,
+      "learning_rate": 9.201543784245076e-06,
+      "loss": 0.2895,
+      "step": 1250
+    },
+    {
+      "epoch": 0.799492385786802,
+      "grad_norm": 1.800851583480835,
+      "learning_rate": 9.181416722893998e-06,
+      "loss": 0.2907,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8058375634517766,
+      "grad_norm": 2.2214279174804688,
+      "learning_rate": 9.161061720728606e-06,
+      "loss": 0.3074,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8121827411167513,
+      "grad_norm": 1.5840632915496826,
+      "learning_rate": 9.140479887357454e-06,
+      "loss": 0.2684,
+      "step": 1280
+    },
+    {
+      "epoch": 0.8185279187817259,
+      "grad_norm": 2.0567562580108643,
+      "learning_rate": 9.119672344754307e-06,
+      "loss": 0.2777,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8248730964467005,
+      "grad_norm": 2.080697774887085,
+      "learning_rate": 9.098640227196978e-06,
+      "loss": 0.294,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8312182741116751,
+      "grad_norm": 2.2059218883514404,
+      "learning_rate": 9.077384681205487e-06,
+      "loss": 0.3483,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8375634517766497,
+      "grad_norm": 1.5565263032913208,
+      "learning_rate": 9.055906865479574e-06,
+      "loss": 0.2744,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8439086294416244,
+      "grad_norm": 1.5794973373413086,
+      "learning_rate": 9.034207950835527e-06,
+      "loss": 0.2803,
+      "step": 1330
+    },
+    {
+      "epoch": 0.850253807106599,
+      "grad_norm": 1.8375296592712402,
+      "learning_rate": 9.01228912014236e-06,
+      "loss": 0.2805,
+      "step": 1340
+    },
+    {
+      "epoch": 0.8565989847715736,
+      "grad_norm": 1.5420727729797363,
+      "learning_rate": 8.99015156825733e-06,
+      "loss": 0.2774,
+      "step": 1350
+    },
+    {
+      "epoch": 0.8629441624365483,
+      "grad_norm": 1.6844383478164673,
+      "learning_rate": 8.967796501960805e-06,
+      "loss": 0.2724,
+      "step": 1360
+    },
+    {
+      "epoch": 0.8692893401015228,
+      "grad_norm": 2.27237606048584,
+      "learning_rate": 8.945225139890468e-06,
+      "loss": 0.2514,
+      "step": 1370
+    },
+    {
+      "epoch": 0.8756345177664975,
+      "grad_norm": 1.6022717952728271,
+      "learning_rate": 8.92243871247491e-06,
+      "loss": 0.2675,
+      "step": 1380
+    },
+    {
+      "epoch": 0.881979695431472,
+      "grad_norm": 1.3979642391204834,
+      "learning_rate": 8.899438461866526e-06,
+      "loss": 0.2404,
+      "step": 1390
+    },
+    {
+      "epoch": 0.8883248730964467,
+      "grad_norm": 1.8629894256591797,
+      "learning_rate": 8.876225641873822e-06,
+      "loss": 0.2744,
+      "step": 1400
+    },
+    {
+      "epoch": 0.8946700507614214,
+      "grad_norm": 1.6122556924819946,
+      "learning_rate": 8.852801517893063e-06,
+      "loss": 0.2814,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9010152284263959,
+      "grad_norm": 2.0331978797912598,
+      "learning_rate": 8.829167366839287e-06,
+      "loss": 0.2728,
+      "step": 1420
+    },
+    {
+      "epoch": 0.9073604060913706,
+      "grad_norm": 1.5905483961105347,
+      "learning_rate": 8.805324477076697e-06,
+      "loss": 0.2503,
+      "step": 1430
+    },
+    {
+      "epoch": 0.9137055837563451,
+      "grad_norm": 1.9675116539001465,
+      "learning_rate": 8.781274148348438e-06,
+      "loss": 0.2241,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9200507614213198,
+      "grad_norm": 1.981604814529419,
+      "learning_rate": 8.757017691705732e-06,
+      "loss": 0.2789,
+      "step": 1450
+    },
+    {
+      "epoch": 0.9263959390862944,
+      "grad_norm": 1.6477928161621094,
+      "learning_rate": 8.732556429436419e-06,
+      "loss": 0.2442,
+      "step": 1460
+    },
+    {
+      "epoch": 0.932741116751269,
+      "grad_norm": 1.875747799873352,
+      "learning_rate": 8.70789169499287e-06,
+      "loss": 0.2372,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9390862944162437,
+      "grad_norm": 1.9763504266738892,
+      "learning_rate": 8.683024832919295e-06,
+      "loss": 0.2493,
+      "step": 1480
+    },
+    {
+      "epoch": 0.9454314720812182,
+      "grad_norm": 2.166445016860962,
+      "learning_rate": 8.657957198778455e-06,
+      "loss": 0.2491,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9517766497461929,
+      "grad_norm": 2.062021493911743,
+      "learning_rate": 8.632690159077758e-06,
+      "loss": 0.2611,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9581218274111675,
+      "grad_norm": 1.5676127672195435,
+      "learning_rate": 8.60722509119478e-06,
+      "loss": 0.2475,
+      "step": 1510
+    },
+    {
+      "epoch": 0.9644670050761421,
+      "grad_norm": 1.734596610069275,
+      "learning_rate": 8.581563383302158e-06,
+      "loss": 0.2499,
+      "step": 1520
+    },
+    {
+      "epoch": 0.9708121827411168,
+      "grad_norm": 2.276888132095337,
+      "learning_rate": 8.555706434291944e-06,
+      "loss": 0.2052,
+      "step": 1530
+    },
+    {
+      "epoch": 0.9771573604060914,
+      "grad_norm": 1.5414533615112305,
+      "learning_rate": 8.529655653699323e-06,
+      "loss": 0.2008,
+      "step": 1540
+    },
+    {
+      "epoch": 0.983502538071066,
+      "grad_norm": 2.0116498470306396,
+      "learning_rate": 8.503412461625792e-06,
+      "loss": 0.2088,
+      "step": 1550
+    },
+    {
+      "epoch": 0.9898477157360406,
+      "grad_norm": 2.507782220840454,
+      "learning_rate": 8.47697828866174e-06,
+      "loss": 0.2212,
+      "step": 1560
+    },
+    {
+      "epoch": 0.9961928934010152,
+      "grad_norm": 1.5416207313537598,
+      "learning_rate": 8.450354575808463e-06,
+      "loss": 0.227,
+      "step": 1570
+    },
+    {
+      "epoch": 1.00253807106599,
+      "grad_norm": 1.7348345518112183,
+      "learning_rate": 8.423542774399606e-06,
+      "loss": 0.2192,
+      "step": 1580
+    },
+    {
+      "epoch": 1.0088832487309645,
+      "grad_norm": 1.8863823413848877,
+      "learning_rate": 8.396544346022055e-06,
+      "loss": 0.159,
+      "step": 1590
+    },
+    {
+      "epoch": 1.015228426395939,
+      "grad_norm": 1.3554282188415527,
+      "learning_rate": 8.36936076243626e-06,
+      "loss": 0.1519,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0215736040609138,
+      "grad_norm": 1.915385127067566,
+      "learning_rate": 8.341993505496e-06,
+      "loss": 0.1667,
+      "step": 1610
+    },
+    {
+      "epoch": 1.0279187817258884,
+      "grad_norm": 2.683910369873047,
+      "learning_rate": 8.314444067067611e-06,
+      "loss": 0.1672,
+      "step": 1620
+    },
+    {
+      "epoch": 1.034263959390863,
+      "grad_norm": 3.2767446041107178,
+      "learning_rate": 8.286713948948646e-06,
+      "loss": 0.151,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0406091370558375,
+      "grad_norm": 1.7172635793685913,
+      "learning_rate": 8.258804662786031e-06,
+      "loss": 0.1365,
+      "step": 1640
+    },
+    {
+      "epoch": 1.0469543147208122,
+      "grad_norm": 1.9492729902267456,
+      "learning_rate": 8.230717729993637e-06,
+      "loss": 0.1521,
+      "step": 1650
+    },
+    {
+      "epoch": 1.0532994923857868,
+      "grad_norm": 1.3974714279174805,
+      "learning_rate": 8.202454681669352e-06,
+      "loss": 0.1784,
+      "step": 1660
+    },
+    {
+      "epoch": 1.0596446700507614,
+      "grad_norm": 1.5528488159179688,
+      "learning_rate": 8.17401705851163e-06,
+      "loss": 0.145,
+      "step": 1670
+    },
+    {
+      "epoch": 1.0659898477157361,
+      "grad_norm": 4.622862815856934,
+      "learning_rate": 8.14540641073548e-06,
+      "loss": 0.149,
+      "step": 1680
+    },
+    {
+      "epoch": 1.0723350253807107,
+      "grad_norm": 1.4450290203094482,
+      "learning_rate": 8.116624297987973e-06,
+      "loss": 0.1354,
+      "step": 1690
+    },
+    {
+      "epoch": 1.0786802030456852,
+      "grad_norm": 1.5473392009735107,
+      "learning_rate": 8.087672289263228e-06,
+      "loss": 0.1355,
+      "step": 1700
+    },
+    {
+      "epoch": 1.0850253807106598,
+      "grad_norm": 1.55717134475708,
+      "learning_rate": 8.058551962816858e-06,
+      "loss": 0.1533,
+      "step": 1710
+    },
+    {
+      "epoch": 1.0913705583756346,
+      "grad_norm": 2.583096742630005,
+      "learning_rate": 8.029264906079962e-06,
+      "loss": 0.1498,
+      "step": 1720
+    },
+    {
+      "epoch": 1.0977157360406091,
+      "grad_norm": 3.534912109375,
+      "learning_rate": 7.99981271557257e-06,
+      "loss": 0.1653,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1040609137055837,
+      "grad_norm": 1.350325345993042,
+      "learning_rate": 7.970196996816622e-06,
+      "loss": 0.1253,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1104060913705585,
+      "grad_norm": 1.4373643398284912,
+      "learning_rate": 7.940419364248445e-06,
+      "loss": 0.1681,
+      "step": 1750
+    },
+    {
+      "epoch": 1.116751269035533,
+      "grad_norm": 2.416491985321045,
+      "learning_rate": 7.910481441130739e-06,
+      "loss": 0.1382,
+      "step": 1760
+    },
+    {
+      "epoch": 1.1230964467005076,
+      "grad_norm": 1.4168888330459595,
+      "learning_rate": 7.880384859464102e-06,
+      "loss": 0.1286,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1294416243654823,
+      "grad_norm": 1.4525187015533447,
+      "learning_rate": 7.850131259898051e-06,
+      "loss": 0.1454,
+      "step": 1780
+    },
+    {
+      "epoch": 1.135786802030457,
+      "grad_norm": 2.431896448135376,
+      "learning_rate": 7.819722291641591e-06,
+      "loss": 0.159,
+      "step": 1790
+    },
+    {
+      "epoch": 1.1421319796954315,
+      "grad_norm": 1.982692837715149,
+      "learning_rate": 7.789159612373317e-06,
+      "loss": 0.1201,
+      "step": 1800
+    },
+    {
+      "epoch": 1.148477157360406,
+      "grad_norm": 1.786580204963684,
+      "learning_rate": 7.758444888151042e-06,
+      "loss": 0.1274,
+      "step": 1810
+    },
+    {
+      "epoch": 1.1548223350253808,
+      "grad_norm": 1.0583122968673706,
+      "learning_rate": 7.727579793320977e-06,
+      "loss": 0.1246,
+      "step": 1820
+    },
+    {
+      "epoch": 1.1611675126903553,
+      "grad_norm": 1.2649511098861694,
+      "learning_rate": 7.69656601042646e-06,
+      "loss": 0.1296,
+      "step": 1830
+    },
+    {
+      "epoch": 1.16751269035533,
+      "grad_norm": 1.5088468790054321,
+      "learning_rate": 7.665405230116232e-06,
+      "loss": 0.1549,
+      "step": 1840
+    },
+    {
+      "epoch": 1.1738578680203045,
+      "grad_norm": 1.6474385261535645,
+      "learning_rate": 7.634099151052283e-06,
+      "loss": 0.1114,
+      "step": 1850
+    },
+    {
+      "epoch": 1.1802030456852792,
+      "grad_norm": 1.665197730064392,
+      "learning_rate": 7.602649479817242e-06,
+      "loss": 0.119,
+      "step": 1860
+    },
+    {
+      "epoch": 1.1865482233502538,
+      "grad_norm": 1.6402256488800049,
+      "learning_rate": 7.5710579308213576e-06,
+      "loss": 0.105,
+      "step": 1870
+    },
+    {
+      "epoch": 1.1928934010152283,
+      "grad_norm": 1.4458770751953125,
+      "learning_rate": 7.539326226209032e-06,
+      "loss": 0.1574,
+      "step": 1880
+    },
+    {
+      "epoch": 1.1992385786802031,
+      "grad_norm": 1.4857584238052368,
+      "learning_rate": 7.507456095764942e-06,
+      "loss": 0.1265,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2055837563451777,
+      "grad_norm": 1.7672957181930542,
+      "learning_rate": 7.475449276819753e-06,
+      "loss": 0.1152,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2119289340101522,
+      "grad_norm": 1.756518006324768,
+      "learning_rate": 7.443307514155402e-06,
+      "loss": 0.1051,
+      "step": 1910
+    },
+    {
+      "epoch": 1.218274111675127,
+      "grad_norm": 2.3999290466308594,
+      "learning_rate": 7.411032559909991e-06,
+      "loss": 0.1249,
+      "step": 1920
+    },
+    {
+      "epoch": 1.2246192893401016,
+      "grad_norm": 2.726649522781372,
+      "learning_rate": 7.378626173482268e-06,
+      "loss": 0.1065,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2309644670050761,
+      "grad_norm": 1.4104615449905396,
+      "learning_rate": 7.346090121435724e-06,
+      "loss": 0.0982,
+      "step": 1940
+    },
+    {
+      "epoch": 1.2373096446700507,
+      "grad_norm": 1.8831905126571655,
+      "learning_rate": 7.313426177402281e-06,
+      "loss": 0.1091,
+      "step": 1950
+    },
+    {
+      "epoch": 1.2436548223350254,
+      "grad_norm": 2.125528573989868,
+      "learning_rate": 7.2806361219856205e-06,
+      "loss": 0.1197,
+      "step": 1960
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 1.8320462703704834,
+      "learning_rate": 7.24772174266411e-06,
+      "loss": 0.0979,
+      "step": 1970
+    },
+    {
+      "epoch": 1.2563451776649746,
+      "grad_norm": 1.6644319295883179,
+      "learning_rate": 7.214684833693362e-06,
+      "loss": 0.1451,
+      "step": 1980
+    },
+    {
+      "epoch": 1.262690355329949,
+      "grad_norm": 1.816611886024475,
+      "learning_rate": 7.181527196008424e-06,
+      "loss": 0.1111,
+      "step": 1990
+    },
+    {
+      "epoch": 1.2690355329949239,
+      "grad_norm": 2.8035154342651367,
+      "learning_rate": 7.148250637125611e-06,
+      "loss": 0.0894,
+      "step": 2000
+    },
+    {
+      "epoch": 1.2753807106598984,
+      "grad_norm": 1.8045902252197266,
+      "learning_rate": 7.114856971043963e-06,
+      "loss": 0.0931,
+      "step": 2010
+    },
+    {
+      "epoch": 1.281725888324873,
+      "grad_norm": 1.637097716331482,
+      "learning_rate": 7.081348018146367e-06,
+      "loss": 0.1572,
+      "step": 2020
+    },
+    {
+      "epoch": 1.2880710659898478,
+      "grad_norm": 1.4267776012420654,
+      "learning_rate": 7.047725605100317e-06,
+      "loss": 0.1071,
+      "step": 2030
+    },
+    {
+      "epoch": 1.2944162436548223,
+      "grad_norm": 2.571660280227661,
+      "learning_rate": 7.01399156475834e-06,
+      "loss": 0.1158,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3007614213197969,
+      "grad_norm": 2.324598789215088,
+      "learning_rate": 6.980147736058083e-06,
+      "loss": 0.0959,
+      "step": 2050
+    },
+    {
+      "epoch": 1.3071065989847717,
+      "grad_norm": 1.4909052848815918,
+      "learning_rate": 6.946195963922064e-06,
+      "loss": 0.1202,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3134517766497462,
+      "grad_norm": 1.6092907190322876,
+      "learning_rate": 6.9121380991571065e-06,
+      "loss": 0.0805,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3197969543147208,
+      "grad_norm": 1.2184277772903442,
+      "learning_rate": 6.877975998353433e-06,
+      "loss": 0.1132,
+      "step": 2080
+    },
+    {
+      "epoch": 1.3261421319796955,
+      "grad_norm": 1.2614070177078247,
+      "learning_rate": 6.8437115237834765e-06,
+      "loss": 0.089,
+      "step": 2090
+    },
+    {
+      "epoch": 1.33248730964467,
+      "grad_norm": 1.7008192539215088,
+      "learning_rate": 6.809346543300346e-06,
+      "loss": 0.0787,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3388324873096447,
+      "grad_norm": 1.3894529342651367,
+      "learning_rate": 6.774882930236015e-06,
+      "loss": 0.0962,
+      "step": 2110
+    },
+    {
+      "epoch": 1.3451776649746192,
+      "grad_norm": 1.7126891613006592,
+      "learning_rate": 6.740322563299195e-06,
+      "loss": 0.0952,
+      "step": 2120
+    },
+    {
+      "epoch": 1.351522842639594,
+      "grad_norm": 1.7561262845993042,
+      "learning_rate": 6.705667326472926e-06,
+      "loss": 0.0989,
+      "step": 2130
+    },
+    {
+      "epoch": 1.3578680203045685,
+      "grad_norm": 1.4162139892578125,
+      "learning_rate": 6.6709191089118685e-06,
+      "loss": 0.1046,
+      "step": 2140
+    },
+    {
+      "epoch": 1.364213197969543,
+      "grad_norm": 1.8884022235870361,
+      "learning_rate": 6.636079804839329e-06,
+      "loss": 0.0847,
+      "step": 2150
+    },
+    {
+      "epoch": 1.3705583756345177,
+      "grad_norm": 1.4617987871170044,
+      "learning_rate": 6.601151313443997e-06,
+      "loss": 0.0858,
+      "step": 2160
+    },
+    {
+      "epoch": 1.3769035532994924,
+      "grad_norm": 1.5476235151290894,
+      "learning_rate": 6.566135538776413e-06,
+      "loss": 0.0907,
+      "step": 2170
+    },
+    {
+      "epoch": 1.383248730964467,
+      "grad_norm": 1.8879975080490112,
+      "learning_rate": 6.531034389645175e-06,
+      "loss": 0.1255,
+      "step": 2180
+    },
+    {
+      "epoch": 1.3895939086294415,
+      "grad_norm": 1.563038945198059,
+      "learning_rate": 6.495849779512879e-06,
+      "loss": 0.084,
+      "step": 2190
+    },
+    {
+      "epoch": 1.3959390862944163,
+      "grad_norm": 2.6775851249694824,
+      "learning_rate": 6.460583626391827e-06,
+      "loss": 0.0957,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4022842639593909,
+      "grad_norm": 5.497508525848389,
+      "learning_rate": 6.4252378527394475e-06,
+      "loss": 0.0882,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4086294416243654,
+      "grad_norm": 2.2709615230560303,
+      "learning_rate": 6.3898143853535145e-06,
+      "loss": 0.1038,
+      "step": 2220
+    },
+    {
+      "epoch": 1.4149746192893402,
+      "grad_norm": 2.0166831016540527,
+      "learning_rate": 6.354315155267105e-06,
+      "loss": 0.0778,
+      "step": 2230
+    },
+    {
+      "epoch": 1.4213197969543148,
+      "grad_norm": 1.4909207820892334,
+      "learning_rate": 6.318742097643336e-06,
+      "loss": 0.1091,
+      "step": 2240
+    },
+    {
+      "epoch": 1.4276649746192893,
+      "grad_norm": 2.3677256107330322,
+      "learning_rate": 6.283097151669869e-06,
+      "loss": 0.1019,
+      "step": 2250
+    },
+    {
+      "epoch": 1.434010152284264,
+      "grad_norm": 3.072751045227051,
+      "learning_rate": 6.247382260453203e-06,
+      "loss": 0.1004,
+      "step": 2260
+    },
+    {
+      "epoch": 1.4403553299492386,
+      "grad_norm": 2.3845341205596924,
+      "learning_rate": 6.211599370912752e-06,
+      "loss": 0.0886,
+      "step": 2270
+    },
+    {
+      "epoch": 1.4467005076142132,
+      "grad_norm": 4.395678997039795,
+      "learning_rate": 6.175750433674708e-06,
+      "loss": 0.1095,
+      "step": 2280
+    },
+    {
+      "epoch": 1.4530456852791878,
+      "grad_norm": 1.326743721961975,
+      "learning_rate": 6.139837402965705e-06,
+      "loss": 0.1021,
+      "step": 2290
+    },
+    {
+      "epoch": 1.4593908629441623,
+      "grad_norm": 1.4270453453063965,
+      "learning_rate": 6.103862236506303e-06,
+      "loss": 0.0744,
+      "step": 2300
+    },
+    {
+      "epoch": 1.465736040609137,
+      "grad_norm": 1.5374149084091187,
+      "learning_rate": 6.067826895404249e-06,
+      "loss": 0.0757,
+      "step": 2310
+    },
+    {
+      "epoch": 1.4720812182741116,
+      "grad_norm": 1.5649033784866333,
+      "learning_rate": 6.031733344047581e-06,
+      "loss": 0.1023,
+      "step": 2320
+    },
+    {
+      "epoch": 1.4784263959390862,
+      "grad_norm": 1.169797420501709,
+      "learning_rate": 5.995583549997542e-06,
+      "loss": 0.0654,
+      "step": 2330
+    },
+    {
+      "epoch": 1.484771573604061,
+      "grad_norm": 1.8578475713729858,
+      "learning_rate": 5.959379483881327e-06,
+      "loss": 0.0819,
+      "step": 2340
+    },
+    {
+      "epoch": 1.4911167512690355,
+      "grad_norm": 1.6423859596252441,
+      "learning_rate": 5.923123119284646e-06,
+      "loss": 0.0663,
+      "step": 2350
+    },
+    {
+      "epoch": 1.49746192893401,
+      "grad_norm": 1.1731383800506592,
+      "learning_rate": 5.886816432644155e-06,
+      "loss": 0.0932,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5038071065989849,
+      "grad_norm": 1.0412118434906006,
+      "learning_rate": 5.850461403139702e-06,
+      "loss": 0.0807,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5101522842639594,
+      "grad_norm": 1.5270987749099731,
+      "learning_rate": 5.814060012586443e-06,
+      "loss": 0.0747,
+      "step": 2380
+    },
+    {
+      "epoch": 1.516497461928934,
+      "grad_norm": 1.9564098119735718,
+      "learning_rate": 5.777614245326802e-06,
+      "loss": 0.0715,
+      "step": 2390
+    },
+    {
+      "epoch": 1.5228426395939088,
+      "grad_norm": 1.6264362335205078,
+      "learning_rate": 5.7411260881223045e-06,
+      "loss": 0.0947,
+      "step": 2400
+    },
+    {
+      "epoch": 1.529187817258883,
+      "grad_norm": 1.0679928064346313,
+      "learning_rate": 5.704597530045272e-06,
+      "loss": 0.0669,
+      "step": 2410
+    },
+    {
+      "epoch": 1.5355329949238579,
+      "grad_norm": 1.393947720527649,
+      "learning_rate": 5.6680305623703926e-06,
+      "loss": 0.089,
+      "step": 2420
+    },
+    {
+      "epoch": 1.5418781725888326,
+      "grad_norm": 1.8824158906936646,
+      "learning_rate": 5.631427178466166e-06,
+      "loss": 0.071,
+      "step": 2430
+    },
+    {
+      "epoch": 1.548223350253807,
+      "grad_norm": 1.060774326324463,
+      "learning_rate": 5.594789373686247e-06,
+      "loss": 0.0747,
+      "step": 2440
+    },
+    {
+      "epoch": 1.5545685279187818,
+      "grad_norm": 1.935646891593933,
+      "learning_rate": 5.5581191452606664e-06,
+      "loss": 0.0671,
+      "step": 2450
+    },
+    {
+      "epoch": 1.5609137055837563,
+      "grad_norm": 1.2591124773025513,
+      "learning_rate": 5.521418492186962e-06,
+      "loss": 0.0796,
+      "step": 2460
+    },
+    {
+      "epoch": 1.5672588832487309,
+      "grad_norm": 2.050698757171631,
+      "learning_rate": 5.484689415121204e-06,
+      "loss": 0.0724,
+      "step": 2470
+    },
+    {
+      "epoch": 1.5736040609137056,
+      "grad_norm": 1.2225536108016968,
+      "learning_rate": 5.447933916268933e-06,
+      "loss": 0.0591,
+      "step": 2480
+    },
+    {
+      "epoch": 1.5799492385786802,
+      "grad_norm": 4.785628318786621,
+      "learning_rate": 5.411153999276016e-06,
+      "loss": 0.0873,
+      "step": 2490
+    },
+    {
+      "epoch": 1.5862944162436547,
+      "grad_norm": 2.2066152095794678,
+      "learning_rate": 5.374351669119425e-06,
+      "loss": 0.057,
+      "step": 2500
+    },
+    {
+      "epoch": 1.5926395939086295,
+      "grad_norm": 1.9447569847106934,
+      "learning_rate": 5.337528931997934e-06,
+      "loss": 0.0548,
+      "step": 2510
+    },
+    {
+      "epoch": 1.598984771573604,
+      "grad_norm": 2.1758713722229004,
+      "learning_rate": 5.3006877952227585e-06,
+      "loss": 0.0674,
+      "step": 2520
+    },
+    {
+      "epoch": 1.6053299492385786,
+      "grad_norm": 1.5067161321640015,
+      "learning_rate": 5.263830267108129e-06,
+      "loss": 0.0583,
+      "step": 2530
+    },
+    {
+      "epoch": 1.6116751269035534,
+      "grad_norm": 1.6991007328033447,
+      "learning_rate": 5.226958356861819e-06,
+      "loss": 0.0521,
+      "step": 2540
+    },
+    {
+      "epoch": 1.618020304568528,
+      "grad_norm": 1.2602826356887817,
+      "learning_rate": 5.190074074475606e-06,
+      "loss": 0.0674,
+      "step": 2550
+    },
+    {
+      "epoch": 1.6243654822335025,
+      "grad_norm": 2.1869382858276367,
+      "learning_rate": 5.153179430615716e-06,
+      "loss": 0.062,
+      "step": 2560
+    },
+    {
+      "epoch": 1.6307106598984773,
+      "grad_norm": 1.6224417686462402,
+      "learning_rate": 5.116276436513201e-06,
+      "loss": 0.0718,
+      "step": 2570
+    },
+    {
+      "epoch": 1.6370558375634516,
+      "grad_norm": 2.291430711746216,
+      "learning_rate": 5.079367103854311e-06,
+      "loss": 0.0722,
+      "step": 2580
+    },
+    {
+      "epoch": 1.6434010152284264,
+      "grad_norm": 1.0190826654434204,
+      "learning_rate": 5.042453444670829e-06,
+      "loss": 0.0612,
+      "step": 2590
+    },
+    {
+      "epoch": 1.649746192893401,
+      "grad_norm": 1.6983177661895752,
+      "learning_rate": 5.005537471230387e-06,
+      "loss": 0.06,
+      "step": 2600
+    },
+    {
+      "epoch": 1.6560913705583755,
+      "grad_norm": 1.5693427324295044,
+      "learning_rate": 4.968621195926779e-06,
+      "loss": 0.0674,
+      "step": 2610
+    },
+    {
+      "epoch": 1.6624365482233503,
+      "grad_norm": 1.4258981943130493,
+      "learning_rate": 4.931706631170246e-06,
+      "loss": 0.0602,
+      "step": 2620
+    },
+    {
+      "epoch": 1.6687817258883249,
+      "grad_norm": 1.9744484424591064,
+      "learning_rate": 4.894795789277789e-06,
+      "loss": 0.0657,
+      "step": 2630
+    },
+    {
+      "epoch": 1.6751269035532994,
+      "grad_norm": 1.0477792024612427,
+      "learning_rate": 4.857890682363461e-06,
+      "loss": 0.0643,
+      "step": 2640
+    },
+    {
+      "epoch": 1.6814720812182742,
+      "grad_norm": 1.2517801523208618,
+      "learning_rate": 4.820993322228691e-06,
+      "loss": 0.0574,
+      "step": 2650
+    },
+    {
+      "epoch": 1.6878172588832487,
+      "grad_norm": 1.339064359664917,
+      "learning_rate": 4.784105720252602e-06,
+      "loss": 0.0639,
+      "step": 2660
+    },
+    {
+      "epoch": 1.6941624365482233,
+      "grad_norm": 1.0788367986679077,
+      "learning_rate": 4.747229887282379e-06,
+      "loss": 0.044,
+      "step": 2670
+    },
+    {
+      "epoch": 1.700507614213198,
+      "grad_norm": 0.8012908697128296,
+      "learning_rate": 4.7103678335236395e-06,
+      "loss": 0.0642,
+      "step": 2680
+    },
+    {
+      "epoch": 1.7068527918781726,
+      "grad_norm": 1.975696086883545,
+      "learning_rate": 4.673521568430859e-06,
+      "loss": 0.0655,
+      "step": 2690
+    },
+    {
+      "epoch": 1.7131979695431472,
+      "grad_norm": 1.7474173307418823,
+      "learning_rate": 4.63669310059783e-06,
+      "loss": 0.0447,
+      "step": 2700
+    },
+    {
+      "epoch": 1.719543147208122,
+      "grad_norm": 0.9429912567138672,
+      "learning_rate": 4.5998844376481665e-06,
+      "loss": 0.0588,
+      "step": 2710
+    },
+    {
+      "epoch": 1.7258883248730963,
+      "grad_norm": 2.345489025115967,
+      "learning_rate": 4.5630975861258605e-06,
+      "loss": 0.0637,
+      "step": 2720
+    },
+    {
+      "epoch": 1.732233502538071,
+      "grad_norm": 0.8988242149353027,
+      "learning_rate": 4.526334551385902e-06,
+      "loss": 0.0613,
+      "step": 2730
+    },
+    {
+      "epoch": 1.7385786802030458,
+      "grad_norm": 2.0134191513061523,
+      "learning_rate": 4.489597337484961e-06,
+      "loss": 0.0533,
+      "step": 2740
+    },
+    {
+      "epoch": 1.7449238578680202,
+      "grad_norm": 1.8432866334915161,
+      "learning_rate": 4.452887947072142e-06,
+      "loss": 0.0684,
+      "step": 2750
+    },
+    {
+      "epoch": 1.751269035532995,
+      "grad_norm": 3.151284694671631,
+      "learning_rate": 4.416208381279812e-06,
+      "loss": 0.0556,
+      "step": 2760
+    },
+    {
+      "epoch": 1.7576142131979695,
+      "grad_norm": 1.051060676574707,
+      "learning_rate": 4.379560639614513e-06,
+      "loss": 0.0498,
+      "step": 2770
+    },
+    {
+      "epoch": 1.763959390862944,
+      "grad_norm": 1.5683525800704956,
+      "learning_rate": 4.3429467198479665e-06,
+      "loss": 0.0524,
+      "step": 2780
+    },
+    {
+      "epoch": 1.7703045685279188,
+      "grad_norm": 1.0461344718933105,
+      "learning_rate": 4.306368617908163e-06,
+      "loss": 0.0445,
+      "step": 2790
+    },
+    {
+      "epoch": 1.7766497461928934,
+      "grad_norm": 1.2296735048294067,
+      "learning_rate": 4.2698283277705655e-06,
+      "loss": 0.0464,
+      "step": 2800
+    },
+    {
+      "epoch": 1.782994923857868,
+      "grad_norm": 0.9869544506072998,
+      "learning_rate": 4.23332784134941e-06,
+      "loss": 0.0506,
+      "step": 2810
+    },
+    {
+      "epoch": 1.7893401015228427,
+      "grad_norm": 2.624345541000366,
+      "learning_rate": 4.196869148389114e-06,
+      "loss": 0.0455,
+      "step": 2820
+    },
+    {
+      "epoch": 1.7956852791878173,
+      "grad_norm": 2.0790648460388184,
+      "learning_rate": 4.160454236355822e-06,
+      "loss": 0.0465,
+      "step": 2830
+    },
+    {
+      "epoch": 1.8020304568527918,
+      "grad_norm": 1.0878472328186035,
+      "learning_rate": 4.124085090329056e-06,
+      "loss": 0.0354,
+      "step": 2840
+    },
+    {
+      "epoch": 1.8083756345177666,
+      "grad_norm": 1.4148125648498535,
+      "learning_rate": 4.087763692893498e-06,
+      "loss": 0.0378,
+      "step": 2850
+    },
+    {
+      "epoch": 1.8147208121827412,
+      "grad_norm": 0.8988755941390991,
+      "learning_rate": 4.051492024030925e-06,
+      "loss": 0.0421,
+      "step": 2860
+    },
+    {
+      "epoch": 1.8210659898477157,
+      "grad_norm": 2.1405270099639893,
+      "learning_rate": 4.015272061012271e-06,
+      "loss": 0.0647,
+      "step": 2870
+    },
+    {
+      "epoch": 1.8274111675126905,
+      "grad_norm": 0.8886227607727051,
+      "learning_rate": 3.979105778289832e-06,
+      "loss": 0.0547,
+      "step": 2880
+    },
+    {
+      "epoch": 1.8337563451776648,
+      "grad_norm": 1.402446985244751,
+      "learning_rate": 3.942995147389648e-06,
+      "loss": 0.0378,
+      "step": 2890
+    },
+    {
+      "epoch": 1.8401015228426396,
+      "grad_norm": 1.283605933189392,
+      "learning_rate": 3.9069421368040115e-06,
+      "loss": 0.0488,
+      "step": 2900
+    },
+    {
+      "epoch": 1.8464467005076142,
+      "grad_norm": 1.226680874824524,
+      "learning_rate": 3.870948711884178e-06,
+      "loss": 0.0382,
+      "step": 2910
+    },
+    {
+      "epoch": 1.8527918781725887,
+      "grad_norm": 1.871385097503662,
+      "learning_rate": 3.835016834733216e-06,
+      "loss": 0.0441,
+      "step": 2920
+    },
+    {
+      "epoch": 1.8591370558375635,
+      "grad_norm": 1.125570297241211,
+      "learning_rate": 3.7991484640990506e-06,
+      "loss": 0.0429,
+      "step": 2930
+    },
+    {
+      "epoch": 1.865482233502538,
+      "grad_norm": 1.131261944770813,
+      "learning_rate": 3.763345555267692e-06,
+      "loss": 0.0404,
+      "step": 2940
+    },
+    {
+      "epoch": 1.8718274111675126,
+      "grad_norm": 1.5131438970565796,
+      "learning_rate": 3.727610059956641e-06,
+      "loss": 0.0359,
+      "step": 2950
+    },
+    {
+      "epoch": 1.8781725888324874,
+      "grad_norm": 0.8379979133605957,
+      "learning_rate": 3.691943926208494e-06,
+      "loss": 0.0508,
+      "step": 2960
+    },
+    {
+      "epoch": 1.884517766497462,
+      "grad_norm": 1.1895625591278076,
+      "learning_rate": 3.6563490982847577e-06,
+      "loss": 0.034,
+      "step": 2970
+    },
+    {
+      "epoch": 1.8908629441624365,
+      "grad_norm": 0.7952091097831726,
+      "learning_rate": 3.620827516559854e-06,
+      "loss": 0.0494,
+      "step": 2980
+    },
+    {
+      "epoch": 1.8972081218274113,
+      "grad_norm": 1.2926766872406006,
+      "learning_rate": 3.58538111741535e-06,
+      "loss": 0.0483,
+      "step": 2990
+    },
+    {
+      "epoch": 1.9035532994923858,
+      "grad_norm": 1.165218472480774,
+      "learning_rate": 3.550011833134399e-06,
+      "loss": 0.0446,
+      "step": 3000
+    },
+    {
+      "epoch": 1.9098984771573604,
+      "grad_norm": 1.2693628072738647,
+      "learning_rate": 3.5147215917964037e-06,
+      "loss": 0.0296,
+      "step": 3010
+    },
+    {
+      "epoch": 1.9162436548223352,
+      "grad_norm": 0.7264485955238342,
+      "learning_rate": 3.4795123171719142e-06,
+      "loss": 0.0488,
+      "step": 3020
+    },
+    {
+      "epoch": 1.9225888324873095,
+      "grad_norm": 0.9121705889701843,
+      "learning_rate": 3.4443859286177545e-06,
+      "loss": 0.0299,
+      "step": 3030
+    },
+    {
+      "epoch": 1.9289340101522843,
+      "grad_norm": 1.2310829162597656,
+      "learning_rate": 3.4093443409723985e-06,
+      "loss": 0.0389,
+      "step": 3040
+    },
+    {
+      "epoch": 1.9352791878172588,
+      "grad_norm": 1.087215542793274,
+      "learning_rate": 3.374389464451583e-06,
+      "loss": 0.0367,
+      "step": 3050
+    },
+    {
+      "epoch": 1.9416243654822334,
+      "grad_norm": 1.1739871501922607,
+      "learning_rate": 3.339523204544176e-06,
+      "loss": 0.0407,
+      "step": 3060
+    },
+    {
+      "epoch": 1.9479695431472082,
+      "grad_norm": 0.9143801927566528,
+      "learning_rate": 3.3047474619083043e-06,
+      "loss": 0.0361,
+      "step": 3070
+    },
+    {
+      "epoch": 1.9543147208121827,
+      "grad_norm": 0.9468094706535339,
+      "learning_rate": 3.2700641322677405e-06,
+      "loss": 0.0309,
+      "step": 3080
+    },
+    {
+      "epoch": 1.9606598984771573,
+      "grad_norm": 1.2729860544204712,
+      "learning_rate": 3.235475106308569e-06,
+      "loss": 0.0194,
+      "step": 3090
+    },
+    {
+      "epoch": 1.967005076142132,
+      "grad_norm": 1.381415843963623,
+      "learning_rate": 3.200982269576111e-06,
+      "loss": 0.0495,
+      "step": 3100
+    },
+    {
+      "epoch": 1.9733502538071066,
+      "grad_norm": 1.4151417016983032,
+      "learning_rate": 3.1665875023721453e-06,
+      "loss": 0.0344,
+      "step": 3110
+    },
+    {
+      "epoch": 1.9796954314720812,
+      "grad_norm": 0.9717885851860046,
+      "learning_rate": 3.1322926796524016e-06,
+      "loss": 0.0376,
+      "step": 3120
+    },
+    {
+      "epoch": 1.986040609137056,
+      "grad_norm": 0.9146430492401123,
+      "learning_rate": 3.0980996709243517e-06,
+      "loss": 0.028,
+      "step": 3130
+    },
+    {
+      "epoch": 1.9923857868020305,
+      "grad_norm": 1.5948601961135864,
+      "learning_rate": 3.0640103401453035e-06,
+      "loss": 0.0511,
+      "step": 3140
+    },
+    {
+      "epoch": 1.998730964467005,
+      "grad_norm": 1.120682716369629,
+      "learning_rate": 3.030026545620787e-06,
+      "loss": 0.0411,
+      "step": 3150
+    },
+    {
+      "epoch": 2.00507614213198,
+      "grad_norm": 0.8402583003044128,
+      "learning_rate": 2.9961501399032546e-06,
+      "loss": 0.0272,
+      "step": 3160
+    },
+    {
+      "epoch": 2.011421319796954,
+      "grad_norm": 1.102598786354065,
+      "learning_rate": 2.9623829696910867e-06,
+      "loss": 0.0207,
+      "step": 3170
+    },
+    {
+      "epoch": 2.017766497461929,
+      "grad_norm": 0.9598972201347351,
+      "learning_rate": 2.928726875727937e-06,
+      "loss": 0.0197,
+      "step": 3180
+    },
+    {
+      "epoch": 2.0241116751269037,
+      "grad_norm": 0.8507049679756165,
+      "learning_rate": 2.8951836927023703e-06,
+      "loss": 0.0161,
+      "step": 3190
+    },
+    {
+      "epoch": 2.030456852791878,
+      "grad_norm": 0.9228895902633667,
+      "learning_rate": 2.861755249147862e-06,
+      "loss": 0.023,
+      "step": 3200
+    },
+    {
+      "epoch": 2.036802030456853,
+      "grad_norm": 0.8271005749702454,
+      "learning_rate": 2.828443367343119e-06,
+      "loss": 0.0148,
+      "step": 3210
+    },
+    {
+      "epoch": 2.0431472081218276,
+      "grad_norm": 1.2311136722564697,
+      "learning_rate": 2.7952498632127324e-06,
+      "loss": 0.0202,
+      "step": 3220
+    },
+    {
+      "epoch": 2.049492385786802,
+      "grad_norm": 1.3220641613006592,
+      "learning_rate": 2.762176546228198e-06,
+      "loss": 0.0235,
+      "step": 3230
+    },
+    {
+      "epoch": 2.0558375634517767,
+      "grad_norm": 1.2385421991348267,
+      "learning_rate": 2.7292252193092693e-06,
+      "loss": 0.0205,
+      "step": 3240
+    },
+    {
+      "epoch": 2.0621827411167515,
+      "grad_norm": 1.238295316696167,
+      "learning_rate": 2.6963976787256726e-06,
+      "loss": 0.0157,
+      "step": 3250
+    },
+    {
+      "epoch": 2.068527918781726,
+      "grad_norm": 0.7305588126182556,
+      "learning_rate": 2.6636957139992003e-06,
+      "loss": 0.0183,
+      "step": 3260
+    },
+    {
+      "epoch": 2.0748730964467006,
+      "grad_norm": 0.8512719869613647,
+      "learning_rate": 2.631121107806144e-06,
+      "loss": 0.0204,
+      "step": 3270
+    },
+    {
+      "epoch": 2.081218274111675,
+      "grad_norm": 0.8006191849708557,
+      "learning_rate": 2.598675635880129e-06,
+      "loss": 0.0223,
+      "step": 3280
+    },
+    {
+      "epoch": 2.0875634517766497,
+      "grad_norm": 1.4886091947555542,
+      "learning_rate": 2.5663610669153043e-06,
+      "loss": 0.0197,
+      "step": 3290
+    },
+    {
+      "epoch": 2.0939086294416245,
+      "grad_norm": 0.7531688213348389,
+      "learning_rate": 2.534179162469924e-06,
+      "loss": 0.0222,
+      "step": 3300
+    },
+    {
+      "epoch": 2.100253807106599,
+      "grad_norm": 0.6706914305686951,
+      "learning_rate": 2.502131676870335e-06,
+      "loss": 0.019,
+      "step": 3310
+    },
+    {
+      "epoch": 2.1065989847715736,
+      "grad_norm": 0.8195891380310059,
+      "learning_rate": 2.470220357115327e-06,
+      "loss": 0.0099,
+      "step": 3320
+    },
+    {
+      "epoch": 2.1129441624365484,
+      "grad_norm": 0.8743392825126648,
+      "learning_rate": 2.438446942780911e-06,
+      "loss": 0.0145,
+      "step": 3330
+    },
+    {
+      "epoch": 2.1192893401015227,
+      "grad_norm": 0.5079776048660278,
+      "learning_rate": 2.4068131659254803e-06,
+      "loss": 0.0164,
+      "step": 3340
+    },
+    {
+      "epoch": 2.1256345177664975,
+      "grad_norm": 0.512514054775238,
+      "learning_rate": 2.3753207509953963e-06,
+      "loss": 0.0287,
+      "step": 3350
+    },
+    {
+      "epoch": 2.1319796954314723,
+      "grad_norm": 0.7019079923629761,
+      "learning_rate": 2.3439714147309845e-06,
+      "loss": 0.0189,
+      "step": 3360
+    },
+    {
+      "epoch": 2.1383248730964466,
+      "grad_norm": 0.8089588284492493,
+      "learning_rate": 2.312766866072947e-06,
+      "loss": 0.0255,
+      "step": 3370
+    },
+    {
+      "epoch": 2.1446700507614214,
+      "grad_norm": 0.9173935651779175,
+      "learning_rate": 2.2817088060692094e-06,
+      "loss": 0.0149,
+      "step": 3380
+    },
+    {
+      "epoch": 2.151015228426396,
+      "grad_norm": 1.1662015914916992,
+      "learning_rate": 2.2507989277821847e-06,
+      "loss": 0.0201,
+      "step": 3390
+    },
+    {
+      "epoch": 2.1573604060913705,
+      "grad_norm": 0.5388917922973633,
+      "learning_rate": 2.2200389161964795e-06,
+      "loss": 0.0198,
+      "step": 3400
+    },
+    {
+      "epoch": 2.1637055837563453,
+      "grad_norm": 1.1195067167282104,
+      "learning_rate": 2.189430448127055e-06,
+      "loss": 0.0196,
+      "step": 3410
+    },
+    {
+      "epoch": 2.1700507614213196,
+      "grad_norm": 0.7136582732200623,
+      "learning_rate": 2.1589751921277925e-06,
+      "loss": 0.0188,
+      "step": 3420
+    },
+    {
+      "epoch": 2.1763959390862944,
+      "grad_norm": 0.773573100566864,
+      "learning_rate": 2.128674808400565e-06,
+      "loss": 0.0212,
+      "step": 3430
+    },
+    {
+      "epoch": 2.182741116751269,
+      "grad_norm": 0.7614580392837524,
+      "learning_rate": 2.098530948704714e-06,
+      "loss": 0.021,
+      "step": 3440
+    },
+    {
+      "epoch": 2.1890862944162435,
+      "grad_norm": 0.6622429490089417,
+      "learning_rate": 2.068545256267015e-06,
+      "loss": 0.0169,
+      "step": 3450
+    },
+    {
+      "epoch": 2.1954314720812182,
+      "grad_norm": 0.3882254660129547,
+      "learning_rate": 2.0387193656921063e-06,
+      "loss": 0.023,
+      "step": 3460
+    },
+    {
+      "epoch": 2.201776649746193,
+      "grad_norm": 1.2883610725402832,
+      "learning_rate": 2.0090549028733685e-06,
+      "loss": 0.0179,
+      "step": 3470
+    },
+    {
+      "epoch": 2.2081218274111674,
+      "grad_norm": 1.0185002088546753,
+      "learning_rate": 1.9795534849043054e-06,
+      "loss": 0.0206,
+      "step": 3480
+    },
+    {
+      "epoch": 2.214467005076142,
+      "grad_norm": 0.7340651154518127,
+      "learning_rate": 1.950216719990383e-06,
+      "loss": 0.0159,
+      "step": 3490
+    },
+    {
+      "epoch": 2.220812182741117,
+      "grad_norm": 0.8917669057846069,
+      "learning_rate": 1.921046207361365e-06,
+      "loss": 0.014,
+      "step": 3500
+    },
+    {
+      "epoch": 2.2271573604060912,
+      "grad_norm": 0.8342999815940857,
+      "learning_rate": 1.8920435371841394e-06,
+      "loss": 0.0168,
+      "step": 3510
+    },
+    {
+      "epoch": 2.233502538071066,
+      "grad_norm": 0.49451372027397156,
+      "learning_rate": 1.8632102904760241e-06,
+      "loss": 0.0202,
+      "step": 3520
+    },
+    {
+      "epoch": 2.239847715736041,
+      "grad_norm": 0.8475871086120605,
+      "learning_rate": 1.8345480390185865e-06,
+      "loss": 0.0228,
+      "step": 3530
+    },
+    {
+      "epoch": 2.246192893401015,
+      "grad_norm": 0.6851008534431458,
+      "learning_rate": 1.806058345271962e-06,
+      "loss": 0.016,
+      "step": 3540
+    },
+    {
+      "epoch": 2.25253807106599,
+      "grad_norm": 1.2128303050994873,
+      "learning_rate": 1.7777427622896764e-06,
+      "loss": 0.0183,
+      "step": 3550
+    },
+    {
+      "epoch": 2.2588832487309647,
+      "grad_norm": 0.3974970877170563,
+      "learning_rate": 1.749602833633992e-06,
+      "loss": 0.0221,
+      "step": 3560
+    },
+    {
+      "epoch": 2.265228426395939,
+      "grad_norm": 0.6373499631881714,
+      "learning_rate": 1.7216400932917544e-06,
+      "loss": 0.0184,
+      "step": 3570
+    },
+    {
+      "epoch": 2.271573604060914,
+      "grad_norm": 0.6473302245140076,
+      "learning_rate": 1.6938560655907743e-06,
+      "loss": 0.0156,
+      "step": 3580
+    },
+    {
+      "epoch": 2.277918781725888,
+      "grad_norm": 0.5753197073936462,
+      "learning_rate": 1.6662522651167345e-06,
+      "loss": 0.0137,
+      "step": 3590
+    },
+    {
+      "epoch": 2.284263959390863,
+      "grad_norm": 0.9094467759132385,
+      "learning_rate": 1.6388301966306215e-06,
+      "loss": 0.0147,
+      "step": 3600
+    },
+    {
+      "epoch": 2.2906091370558377,
+      "grad_norm": 0.5902413725852966,
+      "learning_rate": 1.6115913549867025e-06,
+      "loss": 0.0224,
+      "step": 3610
+    },
+    {
+      "epoch": 2.296954314720812,
+      "grad_norm": 0.875133752822876,
+      "learning_rate": 1.5845372250510287e-06,
+      "loss": 0.0232,
+      "step": 3620
+    },
+    {
+      "epoch": 2.303299492385787,
+      "grad_norm": 1.241910696029663,
+      "learning_rate": 1.557669281620497e-06,
+      "loss": 0.0099,
+      "step": 3630
+    },
+    {
+      "epoch": 2.3096446700507616,
+      "grad_norm": 0.6328564882278442,
+      "learning_rate": 1.5309889893424563e-06,
+      "loss": 0.0132,
+      "step": 3640
+    },
+    {
+      "epoch": 2.315989847715736,
+      "grad_norm": 0.5470057725906372,
+      "learning_rate": 1.5044978026348527e-06,
+      "loss": 0.0164,
+      "step": 3650
+    },
+    {
+      "epoch": 2.3223350253807107,
+      "grad_norm": 1.0264612436294556,
+      "learning_rate": 1.4781971656069665e-06,
+      "loss": 0.0203,
+      "step": 3660
+    },
+    {
+      "epoch": 2.3286802030456855,
+      "grad_norm": 0.6052107810974121,
+      "learning_rate": 1.4520885119806704e-06,
+      "loss": 0.026,
+      "step": 3670
+    },
+    {
+      "epoch": 2.33502538071066,
+      "grad_norm": 0.4180527329444885,
+      "learning_rate": 1.4261732650122795e-06,
+      "loss": 0.0204,
+      "step": 3680
+    },
+    {
+      "epoch": 2.3413705583756346,
+      "grad_norm": 0.6096001267433167,
+      "learning_rate": 1.4004528374149745e-06,
+      "loss": 0.0095,
+      "step": 3690
+    },
+    {
+      "epoch": 2.347715736040609,
+      "grad_norm": 0.5584781765937805,
+      "learning_rate": 1.3749286312817722e-06,
+      "loss": 0.0126,
+      "step": 3700
+    },
+    {
+      "epoch": 2.3540609137055837,
+      "grad_norm": 0.3657080829143524,
+      "learning_rate": 1.349602038009114e-06,
+      "loss": 0.0108,
+      "step": 3710
+    },
+    {
+      "epoch": 2.3604060913705585,
+      "grad_norm": 0.9728971719741821,
+      "learning_rate": 1.3244744382210017e-06,
+      "loss": 0.0104,
+      "step": 3720
+    },
+    {
+      "epoch": 2.3667512690355332,
+      "grad_norm": 0.8524286150932312,
+      "learning_rate": 1.2995472016937405e-06,
+      "loss": 0.0167,
+      "step": 3730
+    },
+    {
+      "epoch": 2.3730964467005076,
+      "grad_norm": 0.6725841164588928,
+      "learning_rate": 1.2748216872812747e-06,
+      "loss": 0.0131,
+      "step": 3740
+    },
+    {
+      "epoch": 2.3794416243654823,
+      "grad_norm": 0.8610649704933167,
+      "learning_rate": 1.2502992428411022e-06,
+      "loss": 0.018,
+      "step": 3750
+    },
+    {
+      "epoch": 2.3857868020304567,
+      "grad_norm": 0.4205199182033539,
+      "learning_rate": 1.2259812051608066e-06,
+      "loss": 0.0158,
+      "step": 3760
+    },
+    {
+      "epoch": 2.3921319796954315,
+      "grad_norm": 0.7805858850479126,
+      "learning_rate": 1.2018688998851802e-06,
+      "loss": 0.0203,
+      "step": 3770
+    },
+    {
+      "epoch": 2.3984771573604062,
+      "grad_norm": 0.2444067746400833,
+      "learning_rate": 1.1779636414439672e-06,
+      "loss": 0.0147,
+      "step": 3780
+    },
+    {
+      "epoch": 2.4048223350253806,
+      "grad_norm": 0.40047794580459595,
+      "learning_rate": 1.1542667329801998e-06,
+      "loss": 0.011,
+      "step": 3790
+    },
+    {
+      "epoch": 2.4111675126903553,
+      "grad_norm": 0.7459643483161926,
+      "learning_rate": 1.130779466279166e-06,
+      "loss": 0.0126,
+      "step": 3800
+    },
+    {
+      "epoch": 2.41751269035533,
+      "grad_norm": 0.6922224760055542,
+      "learning_rate": 1.107503121697997e-06,
+      "loss": 0.0163,
+      "step": 3810
+    },
+    {
+      "epoch": 2.4238578680203045,
+      "grad_norm": 1.863350749015808,
+      "learning_rate": 1.0844389680958533e-06,
+      "loss": 0.0194,
+      "step": 3820
+    },
+    {
+      "epoch": 2.4302030456852792,
+      "grad_norm": 0.29856589436531067,
+      "learning_rate": 1.0615882627647766e-06,
+      "loss": 0.0155,
+      "step": 3830
+    },
+    {
+      "epoch": 2.436548223350254,
+      "grad_norm": 0.377093642950058,
+      "learning_rate": 1.0389522513611372e-06,
+      "loss": 0.015,
+      "step": 3840
+    },
+    {
+      "epoch": 2.4428934010152283,
+      "grad_norm": 0.5333195924758911,
+      "learning_rate": 1.0165321678377332e-06,
+      "loss": 0.0137,
+      "step": 3850
+    },
+    {
+      "epoch": 2.449238578680203,
+      "grad_norm": 0.32329970598220825,
+      "learning_rate": 9.943292343765293e-07,
+      "loss": 0.0084,
+      "step": 3860
+    },
+    {
+      "epoch": 2.4555837563451774,
+      "grad_norm": 0.3231019377708435,
+      "learning_rate": 9.723446613220249e-07,
+      "loss": 0.0126,
+      "step": 3870
+    },
+    {
+      "epoch": 2.4619289340101522,
+      "grad_norm": 0.6870127320289612,
+      "learning_rate": 9.505796471152783e-07,
+      "loss": 0.0137,
+      "step": 3880
+    },
+    {
+      "epoch": 2.468274111675127,
+      "grad_norm": 0.6023297309875488,
+      "learning_rate": 9.290353782285766e-07,
+      "loss": 0.0148,
+      "step": 3890
+    },
+    {
+      "epoch": 2.4746192893401013,
+      "grad_norm": 0.46455860137939453,
+      "learning_rate": 9.077130291007553e-07,
+      "loss": 0.022,
+      "step": 3900
+    },
+    {
+      "epoch": 2.480964467005076,
+      "grad_norm": 0.5320664048194885,
+      "learning_rate": 8.86613762073183e-07,
+      "loss": 0.0096,
+      "step": 3910
+    },
+    {
+      "epoch": 2.487309644670051,
+      "grad_norm": 0.6012682914733887,
+      "learning_rate": 8.657387273263895e-07,
+      "loss": 0.0099,
+      "step": 3920
+    },
+    {
+      "epoch": 2.4936548223350252,
+      "grad_norm": 0.8949501514434814,
+      "learning_rate": 8.450890628173725e-07,
+      "loss": 0.0111,
+      "step": 3930
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.8802683353424072,
+      "learning_rate": 8.246658942175611e-07,
+      "loss": 0.0143,
+      "step": 3940
+    },
+    {
+      "epoch": 2.5063451776649748,
+      "grad_norm": 0.9922573566436768,
+      "learning_rate": 8.04470334851456e-07,
+      "loss": 0.0234,
+      "step": 3950
+    },
+    {
+      "epoch": 2.512690355329949,
+      "grad_norm": 0.23940332233905792,
+      "learning_rate": 7.845034856359368e-07,
+      "loss": 0.011,
+      "step": 3960
+    },
+    {
+      "epoch": 2.519035532994924,
+      "grad_norm": 0.2019755095243454,
+      "learning_rate": 7.647664350202461e-07,
+      "loss": 0.0135,
+      "step": 3970
+    },
+    {
+      "epoch": 2.525380710659898,
+      "grad_norm": 0.17184686660766602,
+      "learning_rate": 7.452602589266583e-07,
+      "loss": 0.0074,
+      "step": 3980
+    },
+    {
+      "epoch": 2.531725888324873,
+      "grad_norm": 0.8647210597991943,
+      "learning_rate": 7.259860206918268e-07,
+      "loss": 0.0101,
+      "step": 3990
+    },
+    {
+      "epoch": 2.5380710659898478,
+      "grad_norm": 0.9781297445297241,
+      "learning_rate": 7.069447710088167e-07,
+      "loss": 0.0147,
+      "step": 4000
+    },
+    {
+      "epoch": 2.5444162436548226,
+      "grad_norm": 0.7230397462844849,
+      "learning_rate": 6.881375478698332e-07,
+      "loss": 0.0159,
+      "step": 4010
+    },
+    {
+      "epoch": 2.550761421319797,
+      "grad_norm": 1.1674317121505737,
+      "learning_rate": 6.695653765096327e-07,
+      "loss": 0.0125,
+      "step": 4020
+    },
+    {
+      "epoch": 2.5571065989847717,
+      "grad_norm": 0.38593119382858276,
+      "learning_rate": 6.512292693496353e-07,
+      "loss": 0.0071,
+      "step": 4030
+    },
+    {
+      "epoch": 2.563451776649746,
+      "grad_norm": 0.3000188171863556,
+      "learning_rate": 6.331302259427418e-07,
+      "loss": 0.0086,
+      "step": 4040
+    },
+    {
+      "epoch": 2.5697969543147208,
+      "grad_norm": 0.6724553108215332,
+      "learning_rate": 6.152692329188297e-07,
+      "loss": 0.0076,
+      "step": 4050
+    },
+    {
+      "epoch": 2.5761421319796955,
+      "grad_norm": 1.0246587991714478,
+      "learning_rate": 5.976472639309888e-07,
+      "loss": 0.02,
+      "step": 4060
+    },
+    {
+      "epoch": 2.5824873096446703,
+      "grad_norm": 0.5962472558021545,
+      "learning_rate": 5.802652796024294e-07,
+      "loss": 0.0208,
+      "step": 4070
+    },
+    {
+      "epoch": 2.5888324873096447,
+      "grad_norm": 0.44684454798698425,
+      "learning_rate": 5.631242274741211e-07,
+      "loss": 0.0179,
+      "step": 4080
+    },
+    {
+      "epoch": 2.5951776649746194,
+      "grad_norm": 0.446123331785202,
+      "learning_rate": 5.46225041953145e-07,
+      "loss": 0.0065,
+      "step": 4090
+    },
+    {
+      "epoch": 2.6015228426395938,
+      "grad_norm": 0.28516885638237,
+      "learning_rate": 5.295686442617442e-07,
+      "loss": 0.0084,
+      "step": 4100
+    },
+    {
+      "epoch": 2.6078680203045685,
+      "grad_norm": 0.42138996720314026,
+      "learning_rate": 5.131559423871191e-07,
+      "loss": 0.0119,
+      "step": 4110
+    },
+    {
+      "epoch": 2.6142131979695433,
+      "grad_norm": 0.857070803642273,
+      "learning_rate": 4.969878310319204e-07,
+      "loss": 0.0116,
+      "step": 4120
+    },
+    {
+      "epoch": 2.6205583756345177,
+      "grad_norm": 0.4262557327747345,
+      "learning_rate": 4.810651915654807e-07,
+      "loss": 0.013,
+      "step": 4130
+    },
+    {
+      "epoch": 2.6269035532994924,
+      "grad_norm": 0.08034439384937286,
+      "learning_rate": 4.6538889197576985e-07,
+      "loss": 0.0085,
+      "step": 4140
+    },
+    {
+      "epoch": 2.6332487309644668,
+      "grad_norm": 0.4999110698699951,
+      "learning_rate": 4.4995978682207396e-07,
+      "loss": 0.0104,
+      "step": 4150
+    },
+    {
+      "epoch": 2.6395939086294415,
+      "grad_norm": 0.47301802039146423,
+      "learning_rate": 4.347787171884149e-07,
+      "loss": 0.013,
+      "step": 4160
+    },
+    {
+      "epoch": 2.6459390862944163,
+      "grad_norm": 0.2837192416191101,
+      "learning_rate": 4.1984651063769864e-07,
+      "loss": 0.0123,
+      "step": 4170
+    },
+    {
+      "epoch": 2.652284263959391,
+      "grad_norm": 0.4908500611782074,
+      "learning_rate": 4.0516398116660196e-07,
+      "loss": 0.0137,
+      "step": 4180
+    },
+    {
+      "epoch": 2.6586294416243654,
+      "grad_norm": 0.38162919878959656,
+      "learning_rate": 3.907319291612027e-07,
+      "loss": 0.0108,
+      "step": 4190
+    },
+    {
+      "epoch": 2.66497461928934,
+      "grad_norm": 0.9448516368865967,
+      "learning_rate": 3.765511413533429e-07,
+      "loss": 0.0139,
+      "step": 4200
+    },
+    {
+      "epoch": 2.6713197969543145,
+      "grad_norm": 0.4047912359237671,
+      "learning_rate": 3.626223907777482e-07,
+      "loss": 0.0147,
+      "step": 4210
+    },
+    {
+      "epoch": 2.6776649746192893,
+      "grad_norm": 0.1890551596879959,
+      "learning_rate": 3.489464367298795e-07,
+      "loss": 0.0135,
+      "step": 4220
+    },
+    {
+      "epoch": 2.684010152284264,
+      "grad_norm": 0.3367404341697693,
+      "learning_rate": 3.3552402472454893e-07,
+      "loss": 0.017,
+      "step": 4230
+    },
+    {
+      "epoch": 2.6903553299492384,
+      "grad_norm": 0.5344458818435669,
+      "learning_rate": 3.2235588645527893e-07,
+      "loss": 0.0201,
+      "step": 4240
+    },
+    {
+      "epoch": 2.696700507614213,
+      "grad_norm": 0.8313795328140259,
+      "learning_rate": 3.094427397544103e-07,
+      "loss": 0.0162,
+      "step": 4250
+    },
+    {
+      "epoch": 2.703045685279188,
+      "grad_norm": 0.35280096530914307,
+      "learning_rate": 2.967852885539768e-07,
+      "loss": 0.0064,
+      "step": 4260
+    },
+    {
+      "epoch": 2.7093908629441623,
+      "grad_norm": 0.6538042426109314,
+      "learning_rate": 2.843842228473293e-07,
+      "loss": 0.0145,
+      "step": 4270
+    },
+    {
+      "epoch": 2.715736040609137,
+      "grad_norm": 0.6905611753463745,
+      "learning_rate": 2.7224021865151996e-07,
+      "loss": 0.0128,
+      "step": 4280
+    },
+    {
+      "epoch": 2.722081218274112,
+      "grad_norm": 0.5076076984405518,
+      "learning_rate": 2.603539379704567e-07,
+      "loss": 0.0171,
+      "step": 4290
+    },
+    {
+      "epoch": 2.728426395939086,
+      "grad_norm": 0.6590428352355957,
+      "learning_rate": 2.4872602875881004e-07,
+      "loss": 0.0077,
+      "step": 4300
+    },
+    {
+      "epoch": 2.734771573604061,
+      "grad_norm": 0.3470360338687897,
+      "learning_rate": 2.373571248866946e-07,
+      "loss": 0.0115,
+      "step": 4310
+    },
+    {
+      "epoch": 2.7411167512690353,
+      "grad_norm": 0.5780541896820068,
+      "learning_rate": 2.262478461051132e-07,
+      "loss": 0.0191,
+      "step": 4320
+    },
+    {
+      "epoch": 2.74746192893401,
+      "grad_norm": 1.4629708528518677,
+      "learning_rate": 2.153987980121719e-07,
+      "loss": 0.0189,
+      "step": 4330
+    },
+    {
+      "epoch": 2.753807106598985,
+      "grad_norm": 1.3563203811645508,
+      "learning_rate": 2.0481057202006992e-07,
+      "loss": 0.0116,
+      "step": 4340
+    },
+    {
+      "epoch": 2.7601522842639596,
+      "grad_norm": 0.4442911744117737,
+      "learning_rate": 1.9448374532285707e-07,
+      "loss": 0.0153,
+      "step": 4350
+    },
+    {
+      "epoch": 2.766497461928934,
+      "grad_norm": 0.26719120144844055,
+      "learning_rate": 1.8441888086497162e-07,
+      "loss": 0.0156,
+      "step": 4360
+    },
+    {
+      "epoch": 2.7728426395939088,
+      "grad_norm": 0.4203988015651703,
+      "learning_rate": 1.7461652731055157e-07,
+      "loss": 0.0162,
+      "step": 4370
+    },
+    {
+      "epoch": 2.779187817258883,
+      "grad_norm": 1.0901730060577393,
+      "learning_rate": 1.650772190135247e-07,
+      "loss": 0.0131,
+      "step": 4380
+    },
+    {
+      "epoch": 2.785532994923858,
+      "grad_norm": 0.3400239944458008,
+      "learning_rate": 1.5580147598848018e-07,
+      "loss": 0.0141,
+      "step": 4390
+    },
+    {
+      "epoch": 2.7918781725888326,
+      "grad_norm": 0.38450250029563904,
+      "learning_rate": 1.4678980388232233e-07,
+      "loss": 0.0099,
+      "step": 4400
+    },
+    {
+      "epoch": 2.798223350253807,
+      "grad_norm": 0.4401623606681824,
+      "learning_rate": 1.3804269394670388e-07,
+      "loss": 0.0166,
+      "step": 4410
+    },
+    {
+      "epoch": 2.8045685279187818,
+      "grad_norm": 0.765143871307373,
+      "learning_rate": 1.295606230112495e-07,
+      "loss": 0.015,
+      "step": 4420
+    },
+    {
+      "epoch": 2.810913705583756,
+      "grad_norm": 0.47553789615631104,
+      "learning_rate": 1.2134405345755773e-07,
+      "loss": 0.0104,
+      "step": 4430
+    },
+    {
+      "epoch": 2.817258883248731,
+      "grad_norm": 1.053678274154663,
+      "learning_rate": 1.1339343319400175e-07,
+      "loss": 0.0085,
+      "step": 4440
+    },
+    {
+      "epoch": 2.8236040609137056,
+      "grad_norm": 0.5694789290428162,
+      "learning_rate": 1.057091956313061e-07,
+      "loss": 0.0131,
+      "step": 4450
+    },
+    {
+      "epoch": 2.8299492385786804,
+      "grad_norm": 0.41042569279670715,
+      "learning_rate": 9.829175965892557e-08,
+      "loss": 0.0162,
+      "step": 4460
+    },
+    {
+      "epoch": 2.8362944162436547,
+      "grad_norm": 0.30753186345100403,
+      "learning_rate": 9.114152962220734e-08,
+      "loss": 0.0085,
+      "step": 4470
+    },
+    {
+      "epoch": 2.8426395939086295,
+      "grad_norm": 1.1423698663711548,
+      "learning_rate": 8.425889530034815e-08,
+      "loss": 0.0111,
+      "step": 4480
+    },
+    {
+      "epoch": 2.848984771573604,
+      "grad_norm": 0.9772459864616394,
+      "learning_rate": 7.764423188515058e-08,
+      "loss": 0.0137,
+      "step": 4490
+    },
+    {
+      "epoch": 2.8553299492385786,
+      "grad_norm": 0.2530859112739563,
+      "learning_rate": 7.129789996056568e-08,
+      "loss": 0.0148,
+      "step": 4500
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 4728,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 391740982493184.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddfb641a3858b4d87703e8b63d389c469e300e2afe6cce5876835fc3ee89fe0e
+size 8145

zero_to_fp32.py

@@ -0,0 +1,760 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device, weights_only=False)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
+
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)