Add files using upload-large-folder tool

Blood/seed_0/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/README.md

@@ -0,0 +1,67 @@
+---
+library_name: transformers
+model_name: LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized
+tags:
+- generated_from_trainer
+- grpo
+- trl
+licence: license
+---
+
+# Model Card for LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized
+
+This model is a fine-tuned version of [None](https://huggingface.co/None).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/baileyyeah326/MLLM/runs/hkmta5lx) 
+
+
+This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
+
+### Framework versions
+
+- TRL: 0.21.0
+- Transformers: 4.55.2
+- Pytorch: 2.4.0+cu124
+- Datasets: 4.0.0
+- Tokenizers: 0.21.4
+
+## Citations
+
+Cite GRPO as:
+
+```bibtex
+@article{zhihong2024deepseekmath,
+    title        = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
+    author       = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
+    year         = 2024,
+    eprint       = {arXiv:2402.03300},
+}
+
+```
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

Blood/seed_0/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/added_tokens.json

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

Blood/seed_0/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

Blood/seed_0/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/config.json

@@ -0,0 +1,66 @@
+{
+  "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,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.2",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Blood/seed_0/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/model.safetensors.index.json

@@ -0,0 +1,443 @@
+{
+  "metadata": {
+    "total_parameters": 3085938688,
+    "total_size": 6794207232
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "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"
+  }
+}

Blood/seed_0/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/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
+  }
+}

Blood/seed_0/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 2048,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

Blood/seed_0/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/trainer_state.json

@@ -0,0 +1,2331 @@
+{
+  "best_global_step": 400,
+  "best_metric": 0.7866666666666666,
+  "best_model_checkpoint": "/data/chenlang/Med3DVLM/llm/Blood/seeds/new/seed_0/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/checkpoint-400",
+  "epoch": 3.0,
+  "eval_steps": 100,
+  "global_step": 3588,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 490.5,
+      "eval_completions/max_terminated_length": 490.5,
+      "eval_completions/mean_length": 417.03125,
+      "eval_completions/mean_terminated_length": 417.03125,
+      "eval_completions/min_length": 321.0,
+      "eval_completions/min_terminated_length": 321.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 0.0,
+      "eval_reward": 0.7265625,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.7265625,
+      "eval_rewards/accuracy_reward/std": 0.44706645607948303,
+      "eval_runtime": 56.9822,
+      "eval_samples_per_second": 1.316,
+      "eval_steps_per_second": 0.035,
+      "step": 0
+    },
+    {
+      "clip_ratio/high_max": 0.0007908888192105223,
+      "clip_ratio/high_mean": 0.00039544440960526117,
+      "clip_ratio/low_mean": 4.6173167902452406e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0004000617263955064,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 476.9375,
+      "completions/max_terminated_length": 476.9375,
+      "completions/mean_length": 415.46875,
+      "completions/mean_terminated_length": 415.46875,
+      "completions/min_length": 356.75,
+      "completions/min_terminated_length": 356.75,
+      "entropy": 0.09742542379535735,
+      "epoch": 0.05351170568561873,
+      "frac_reward_zero_std": 0.84375,
+      "grad_norm": 0.07067517191171646,
+      "kl": 0.0019223751488652852,
+      "learning_rate": 1.9648829431438127e-06,
+      "loss": 0.0004,
+      "num_tokens": 346848.0,
+      "reward": 0.849609375,
+      "reward_std": 0.052045637741684914,
+      "rewards/accuracy_reward/mean": 0.849609375,
+      "rewards/accuracy_reward/std": 0.18268440011888742,
+      "step": 64
+    },
+    {
+      "epoch": 0.08361204013377926,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 509.0,
+      "eval_completions/max_terminated_length": 509.0,
+      "eval_completions/mean_length": 416.984375,
+      "eval_completions/mean_terminated_length": 416.984375,
+      "eval_completions/min_length": 327.5,
+      "eval_completions/min_terminated_length": 327.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 541184.0,
+      "eval_reward": 0.7578125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.7578125,
+      "eval_rewards/accuracy_reward/std": 0.4309118092060089,
+      "eval_runtime": 58.9151,
+      "eval_samples_per_second": 1.273,
+      "eval_steps_per_second": 0.034,
+      "step": 100
+    },
+    {
+      "clip_ratio/high_max": 0.0003593869956343302,
+      "clip_ratio/high_mean": 0.0003047454187513462,
+      "clip_ratio/low_mean": 0.0001279211137443781,
+      "clip_ratio/low_min": 4.644250397437385e-05,
+      "clip_ratio/region_mean": 0.0004326665324957243,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 470.57142857142856,
+      "completions/max_terminated_length": 470.57142857142856,
+      "completions/mean_length": 413.88392857142856,
+      "completions/mean_terminated_length": 413.88392857142856,
+      "completions/min_length": 325.0,
+      "completions/min_terminated_length": 325.0,
+      "entropy": 0.09871216783566135,
+      "epoch": 0.10702341137123746,
+      "frac_reward_zero_std": 0.8571428571428571,
+      "grad_norm": 0.08249606937170029,
+      "kl": 0.006351065875995638,
+      "learning_rate": 1.9292084726867336e-06,
+      "loss": -0.0037,
+      "num_tokens": 692662.0,
+      "reward": 0.9866071428571429,
+      "reward_std": 0.04225464378084455,
+      "rewards/accuracy_reward/mean": 0.9866071428571429,
+      "rewards/accuracy_reward/std": 0.042306367840085714,
+      "step": 128
+    },
+    {
+      "clip_ratio/high_max": 0.00019411579887673724,
+      "clip_ratio/high_mean": 9.705789943836862e-05,
+      "clip_ratio/low_mean": 0.0002299350276189216,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00032699292478355346,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 477.125,
+      "completions/max_terminated_length": 477.125,
+      "completions/mean_length": 427.84765625,
+      "completions/mean_terminated_length": 427.84765625,
+      "completions/min_length": 375.1875,
+      "completions/min_terminated_length": 375.1875,
+      "entropy": 0.09535896719899029,
+      "epoch": 0.1605351170568562,
+      "frac_reward_zero_std": 0.90625,
+      "grad_norm": 2.7755444049835205,
+      "kl": 0.008481005030262168,
+      "learning_rate": 1.8935340022296544e-06,
+      "loss": 0.0005,
+      "num_tokens": 1045752.0,
+      "reward": 0.845703125,
+      "reward_std": 0.0234375,
+      "rewards/accuracy_reward/mean": 0.845703125,
+      "rewards/accuracy_reward/std": 0.16967455577105284,
+      "step": 192
+    },
+    {
+      "epoch": 0.16722408026755853,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 507.0,
+      "eval_completions/max_terminated_length": 507.0,
+      "eval_completions/mean_length": 411.984375,
+      "eval_completions/mean_terminated_length": 411.984375,
+      "eval_completions/min_length": 329.0,
+      "eval_completions/min_terminated_length": 329.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 1088259.0,
+      "eval_reward": 0.7265625,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.7265625,
+      "eval_rewards/accuracy_reward/std": 0.4484642744064331,
+      "eval_runtime": 59.3904,
+      "eval_samples_per_second": 1.263,
+      "eval_steps_per_second": 0.034,
+      "step": 200
+    },
+    {
+      "clip_ratio/high_max": 0.0006860934559621715,
+      "clip_ratio/high_mean": 0.00034304672798108574,
+      "clip_ratio/low_mean": 0.0001461557621951215,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0004892024849790946,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 461.85714285714283,
+      "completions/max_terminated_length": 461.85714285714283,
+      "completions/mean_length": 405.6763392857143,
+      "completions/mean_terminated_length": 405.6763392857143,
+      "completions/min_length": 341.07142857142856,
+      "completions/min_terminated_length": 341.07142857142856,
+      "entropy": 0.09618141941194024,
+      "epoch": 0.2140468227424749,
+      "frac_reward_zero_std": 0.8928571428571429,
+      "grad_norm": 0.05106227472424507,
+      "kl": 0.015624415898595803,
+      "learning_rate": 1.8578595317725752e-06,
+      "loss": 0.0031,
+      "num_tokens": 1387378.0,
+      "reward": 0.8973214285714286,
+      "reward_std": 0.046613442046301704,
+      "rewards/accuracy_reward/mean": 0.8973214285714286,
+      "rewards/accuracy_reward/std": 0.14556341192552022,
+      "step": 256
+    },
+    {
+      "epoch": 0.2508361204013378,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 497.5,
+      "eval_completions/max_terminated_length": 497.5,
+      "eval_completions/mean_length": 413.3828125,
+      "eval_completions/mean_terminated_length": 413.3828125,
+      "eval_completions/min_length": 317.5,
+      "eval_completions/min_terminated_length": 317.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 1628714.0,
+      "eval_reward": 0.75,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.75,
+      "eval_rewards/accuracy_reward/std": 0.43605661392211914,
+      "eval_runtime": 57.5325,
+      "eval_samples_per_second": 1.304,
+      "eval_steps_per_second": 0.035,
+      "step": 300
+    },
+    {
+      "clip_ratio/high_max": 0.00016558784700464456,
+      "clip_ratio/high_mean": 8.279392350232228e-05,
+      "clip_ratio/low_mean": 0.0005344158911611885,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0006172098161187023,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 447.6,
+      "completions/max_terminated_length": 447.6,
+      "completions/mean_length": 397.49375,
+      "completions/mean_terminated_length": 397.49375,
+      "completions/min_length": 365.2,
+      "completions/min_terminated_length": 365.2,
+      "entropy": 0.08891196176409721,
+      "epoch": 0.26755852842809363,
+      "frac_reward_zero_std": 0.9,
+      "grad_norm": 0.10075017064809799,
+      "kl": 0.019166122027672827,
+      "learning_rate": 1.822185061315496e-06,
+      "loss": 0.0015,
+      "num_tokens": 1734249.0,
+      "reward": 0.425,
+      "reward_std": 0.04472135901451111,
+      "rewards/accuracy_reward/mean": 0.425,
+      "rewards/accuracy_reward/std": 0.1999739706516266,
+      "step": 320
+    },
+    {
+      "clip_ratio/high_max": 0.0003638084035628708,
+      "clip_ratio/high_mean": 0.0001819042017814354,
+      "clip_ratio/low_mean": 0.0002965977841995482,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00047850198188825743,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 468.75,
+      "completions/max_terminated_length": 468.75,
+      "completions/mean_length": 408.599609375,
+      "completions/mean_terminated_length": 408.599609375,
+      "completions/min_length": 358.375,
+      "completions/min_terminated_length": 358.375,
+      "entropy": 0.09487172344233841,
+      "epoch": 0.3210702341137124,
+      "frac_reward_zero_std": 0.875,
+      "grad_norm": 1.0121594667434692,
+      "kl": 0.021883822595555102,
+      "learning_rate": 1.786510590858417e-06,
+      "loss": 0.0023,
+      "num_tokens": 2077548.0,
+      "reward": 0.67578125,
+      "reward_std": 0.044233137741684914,
+      "rewards/accuracy_reward/mean": 0.67578125,
+      "rewards/accuracy_reward/std": 0.3040844462811947,
+      "step": 384
+    },
+    {
+      "epoch": 0.33444816053511706,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 493.0,
+      "eval_completions/max_terminated_length": 493.0,
+      "eval_completions/mean_length": 417.265625,
+      "eval_completions/mean_terminated_length": 417.265625,
+      "eval_completions/min_length": 321.5,
+      "eval_completions/min_terminated_length": 321.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 2168096.0,
+      "eval_reward": 0.75,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.75,
+      "eval_rewards/accuracy_reward/std": 0.4364357888698578,
+      "eval_runtime": 57.8802,
+      "eval_samples_per_second": 1.296,
+      "eval_steps_per_second": 0.035,
+      "step": 400
+    },
+    {
+      "clip_ratio/high_max": 0.0008308086750427416,
+      "clip_ratio/high_mean": 0.000467669527400479,
+      "clip_ratio/low_mean": 0.0007246423750378502,
+      "clip_ratio/low_min": 0.00011243560341730093,
+      "clip_ratio/region_mean": 0.0011923118945560418,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 472.5,
+      "completions/max_terminated_length": 472.5,
+      "completions/mean_length": 418.3307291666667,
+      "completions/mean_terminated_length": 418.3307291666667,
+      "completions/min_length": 360.8333333333333,
+      "completions/min_terminated_length": 360.8333333333333,
+      "entropy": 0.1242803984011213,
+      "epoch": 0.3745819397993311,
+      "frac_reward_zero_std": 0.7083333333333334,
+      "grad_norm": 0.8793007135391235,
+      "kl": 0.07019452709937468,
+      "learning_rate": 1.7508361204013377e-06,
+      "loss": 0.0067,
+      "num_tokens": 2429423.0,
+      "reward": 0.734375,
+      "reward_std": 0.11149696509043376,
+      "rewards/accuracy_reward/mean": 0.734375,
+      "rewards/accuracy_reward/std": 0.25355326011776924,
+      "step": 448
+    },
+    {
+      "epoch": 0.4180602006688963,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 495.5,
+      "eval_completions/max_terminated_length": 495.5,
+      "eval_completions/mean_length": 413.9140625,
+      "eval_completions/mean_terminated_length": 413.9140625,
+      "eval_completions/min_length": 318.5,
+      "eval_completions/min_terminated_length": 318.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 2717672.0,
+      "eval_reward": 0.734375,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.734375,
+      "eval_rewards/accuracy_reward/std": 0.44479964673519135,
+      "eval_runtime": 58.746,
+      "eval_samples_per_second": 1.277,
+      "eval_steps_per_second": 0.034,
+      "step": 500
+    },
+    {
+      "clip_ratio/high_max": 0.000866840787542363,
+      "clip_ratio/high_mean": 0.0004334203937711815,
+      "clip_ratio/low_mean": 0.00020594061546338102,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0006393609849813705,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 479.3333333333333,
+      "completions/max_terminated_length": 479.3333333333333,
+      "completions/mean_length": 420.9791666666667,
+      "completions/mean_terminated_length": 420.9791666666667,
+      "completions/min_length": 365.6666666666667,
+      "completions/min_terminated_length": 365.6666666666667,
+      "entropy": 0.12591193988919258,
+      "epoch": 0.4280936454849498,
+      "frac_reward_zero_std": 0.8333333333333334,
+      "grad_norm": 0.09410286694765091,
+      "kl": 0.05266569306453069,
+      "learning_rate": 1.7151616499442586e-06,
+      "loss": 0.0056,
+      "num_tokens": 2783286.0,
+      "reward": 0.96875,
+      "reward_std": 0.0671854813893636,
+      "rewards/accuracy_reward/mean": 0.96875,
+      "rewards/accuracy_reward/std": 0.09871485829353333,
+      "step": 512
+    },
+    {
+      "clip_ratio/high_max": 7.435129555233289e-05,
+      "clip_ratio/high_mean": 3.717564777616644e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 3.717564777616644e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 475.8125,
+      "completions/max_terminated_length": 475.8125,
+      "completions/mean_length": 421.255859375,
+      "completions/mean_terminated_length": 421.255859375,
+      "completions/min_length": 365.25,
+      "completions/min_terminated_length": 365.25,
+      "entropy": 0.11762141226790845,
+      "epoch": 0.4816053511705686,
+      "frac_reward_zero_std": 0.96875,
+      "grad_norm": 0.08674240112304688,
+      "kl": 0.030779874985455535,
+      "learning_rate": 1.6794871794871794e-06,
+      "loss": 0.003,
+      "num_tokens": 3133097.0,
+      "reward": 0.80859375,
+      "reward_std": 0.010673906654119492,
+      "rewards/accuracy_reward/mean": 0.80859375,
+      "rewards/accuracy_reward/std": 0.06325103715062141,
+      "step": 576
+    },
+    {
+      "epoch": 0.5016722408026756,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 491.5,
+      "eval_completions/max_terminated_length": 491.5,
+      "eval_completions/mean_length": 415.6484375,
+      "eval_completions/mean_terminated_length": 415.6484375,
+      "eval_completions/min_length": 318.0,
+      "eval_completions/min_terminated_length": 318.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 3263273.0,
+      "eval_reward": 0.7421875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.7421875,
+      "eval_rewards/accuracy_reward/std": 0.440796360373497,
+      "eval_runtime": 56.5774,
+      "eval_samples_per_second": 1.326,
+      "eval_steps_per_second": 0.035,
+      "step": 600
+    },
+    {
+      "clip_ratio/high_max": 4.976774798706174e-05,
+      "clip_ratio/high_mean": 2.488387399353087e-05,
+      "clip_ratio/low_mean": 0.00016385155322495848,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00018873543303925543,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 474.7,
+      "completions/max_terminated_length": 474.7,
+      "completions/mean_length": 411.803125,
+      "completions/mean_terminated_length": 411.803125,
+      "completions/min_length": 346.0,
+      "completions/min_terminated_length": 346.0,
+      "entropy": 0.11736641991883516,
+      "epoch": 0.5351170568561873,
+      "frac_reward_zero_std": 0.95,
+      "grad_norm": 1.3029202222824097,
+      "kl": 0.026357294467743486,
+      "learning_rate": 1.6438127090301002e-06,
+      "loss": 0.0028,
+      "num_tokens": 3478874.0,
+      "reward": 0.609375,
+      "reward_std": 0.02015564441680908,
+      "rewards/accuracy_reward/mean": 0.609375,
+      "rewards/accuracy_reward/std": 0.30389933586120604,
+      "step": 640
+    },
+    {
+      "epoch": 0.5852842809364549,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 491.0,
+      "eval_completions/max_terminated_length": 491.0,
+      "eval_completions/mean_length": 408.3203125,
+      "eval_completions/mean_terminated_length": 408.3203125,
+      "eval_completions/min_length": 305.0,
+      "eval_completions/min_terminated_length": 305.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 3800664.0,
+      "eval_reward": 0.703125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.703125,
+      "eval_rewards/accuracy_reward/std": 0.4552833288908005,
+      "eval_runtime": 57.1263,
+      "eval_samples_per_second": 1.313,
+      "eval_steps_per_second": 0.035,
+      "step": 700
+    },
+    {
+      "clip_ratio/high_max": 0.001476035060477443,
+      "clip_ratio/high_mean": 0.0007380175302387215,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0007380175302387215,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 507.0,
+      "completions/max_terminated_length": 507.0,
+      "completions/mean_length": 434.03125,
+      "completions/mean_terminated_length": 434.03125,
+      "completions/min_length": 369.0,
+      "completions/min_terminated_length": 369.0,
+      "entropy": 0.1387515664100647,
+      "epoch": 0.5886287625418061,
+      "frac_reward_zero_std": 0.5,
+      "grad_norm": 0.545124351978302,
+      "kl": 0.09541250206530094,
+      "learning_rate": 1.608138238573021e-06,
+      "loss": 0.0109,
+      "num_tokens": 3822969.0,
+      "reward": 0.96875,
+      "reward_std": 0.125,
+      "rewards/accuracy_reward/mean": 0.96875,
+      "rewards/accuracy_reward/std": 0.1767766922712326,
+      "step": 704
+    },
+    {
+      "clip_ratio/high_max": 0.00048521713688387536,
+      "clip_ratio/high_mean": 0.00024260856844193768,
+      "clip_ratio/low_mean": 7.099746426320053e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00031360603679786436,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 469.3125,
+      "completions/max_terminated_length": 469.3125,
+      "completions/mean_length": 403.490234375,
+      "completions/mean_terminated_length": 403.490234375,
+      "completions/min_length": 345.0625,
+      "completions/min_terminated_length": 345.0625,
+      "entropy": 0.12115569307934493,
+      "epoch": 0.6421404682274248,
+      "frac_reward_zero_std": 0.875,
+      "grad_norm": 0.09112782031297684,
+      "kl": 0.03751233006914845,
+      "learning_rate": 1.5724637681159421e-06,
+      "loss": 0.0046,
+      "num_tokens": 4163780.0,
+      "reward": 0.8359375,
+      "reward_std": 0.05667105130851269,
+      "rewards/accuracy_reward/mean": 0.8359375,
+      "rewards/accuracy_reward/std": 0.21587339136749506,
+      "step": 768
+    },
+    {
+      "epoch": 0.6688963210702341,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 498.0,
+      "eval_completions/max_terminated_length": 498.0,
+      "eval_completions/mean_length": 418.78125,
+      "eval_completions/mean_terminated_length": 418.78125,
+      "eval_completions/min_length": 325.0,
+      "eval_completions/min_terminated_length": 325.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 4336339.0,
+      "eval_reward": 0.7109375,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.7109375,
+      "eval_rewards/accuracy_reward/std": 0.4561667740345001,
+      "eval_runtime": 58.6319,
+      "eval_samples_per_second": 1.279,
+      "eval_steps_per_second": 0.034,
+      "step": 800
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 457.25,
+      "completions/max_terminated_length": 457.25,
+      "completions/mean_length": 402.98046875,
+      "completions/mean_terminated_length": 402.98046875,
+      "completions/min_length": 344.0,
+      "completions/min_terminated_length": 344.0,
+      "entropy": 0.11502773710526526,
+      "epoch": 0.6956521739130435,
+      "frac_reward_zero_std": 1.0,
+      "grad_norm": 0.23805662989616394,
+      "kl": 0.01497033271152759,
+      "learning_rate": 1.5367892976588628e-06,
+      "loss": 0.0029,
+      "num_tokens": 4506622.0,
+      "reward": 0.875,
+      "reward_std": 0.0,
+      "rewards/accuracy_reward/mean": 0.875,
+      "rewards/accuracy_reward/std": 0.1270001232624054,
+      "step": 832
+    },
+    {
+      "clip_ratio/high_max": 0.0009068004947039299,
+      "clip_ratio/high_mean": 0.0004959487878295477,
+      "clip_ratio/low_mean": 5.776368061560788e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.000553712471628387,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 471.5625,
+      "completions/max_terminated_length": 471.5625,
+      "completions/mean_length": 414.232421875,
+      "completions/mean_terminated_length": 414.232421875,
+      "completions/min_length": 365.375,
+      "completions/min_terminated_length": 365.375,
+      "entropy": 0.129343748325482,
+      "epoch": 0.7491638795986622,
+      "frac_reward_zero_std": 0.75,
+      "grad_norm": 0.22267663478851318,
+      "kl": 0.06127101270976709,
+      "learning_rate": 1.5011148272017838e-06,
+      "loss": 0.0099,
+      "num_tokens": 4852805.0,
+      "reward": 0.873046875,
+      "reward_std": 0.08301629684865475,
+      "rewards/accuracy_reward/mean": 0.873046875,
+      "rewards/accuracy_reward/std": 0.19705284852534533,
+      "step": 896
+    },
+    {
+      "epoch": 0.7525083612040134,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 493.0,
+      "eval_completions/max_terminated_length": 493.0,
+      "eval_completions/mean_length": 413.3203125,
+      "eval_completions/mean_terminated_length": 413.3203125,
+      "eval_completions/min_length": 329.5,
+      "eval_completions/min_terminated_length": 329.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 4873883.0,
+      "eval_reward": 0.75,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.75,
+      "eval_rewards/accuracy_reward/std": 0.43491509556770325,
+      "eval_runtime": 56.8767,
+      "eval_samples_per_second": 1.319,
+      "eval_steps_per_second": 0.035,
+      "step": 900
+    },
+    {
+      "clip_ratio/high_max": 0.0004170371171009416,
+      "clip_ratio/high_mean": 0.0002085185585504708,
+      "clip_ratio/low_mean": 0.00011040495689182231,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00031892351592735697,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 466.3333333333333,
+      "completions/max_terminated_length": 466.3333333333333,
+      "completions/mean_length": 403.9375,
+      "completions/mean_terminated_length": 403.9375,
+      "completions/min_length": 349.8,
+      "completions/min_terminated_length": 349.8,
+      "entropy": 0.13929289107521375,
+      "epoch": 0.802675585284281,
+      "frac_reward_zero_std": 0.8333333333333334,
+      "grad_norm": 0.45232445001602173,
+      "kl": 0.04927637780395647,
+      "learning_rate": 1.4654403567447044e-06,
+      "loss": 0.0042,
+      "num_tokens": 5193645.0,
+      "reward": 0.8520833333333333,
+      "reward_std": 0.06744743982950846,
+      "rewards/accuracy_reward/mean": 0.8520833333333333,
+      "rewards/accuracy_reward/std": 0.20930681228637696,
+      "step": 960
+    },
+    {
+      "epoch": 0.8361204013377926,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 498.0,
+      "eval_completions/max_terminated_length": 498.0,
+      "eval_completions/mean_length": 414.2265625,
+      "eval_completions/mean_terminated_length": 414.2265625,
+      "eval_completions/min_length": 327.0,
+      "eval_completions/min_terminated_length": 327.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 5410791.0,
+      "eval_reward": 0.671875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.671875,
+      "eval_rewards/accuracy_reward/std": 0.47205156087875366,
+      "eval_runtime": 57.8412,
+      "eval_samples_per_second": 1.297,
+      "eval_steps_per_second": 0.035,
+      "step": 1000
+    },
+    {
+      "clip_ratio/high_max": 0.0008678070929211875,
+      "clip_ratio/high_mean": 0.00043390354646059376,
+      "clip_ratio/low_mean": 0.0002630315866554156,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0006969351331160093,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 474.8333333333333,
+      "completions/max_terminated_length": 474.8333333333333,
+      "completions/mean_length": 412.7291666666667,
+      "completions/mean_terminated_length": 412.7291666666667,
+      "completions/min_length": 354.8333333333333,
+      "completions/min_terminated_length": 354.8333333333333,
+      "entropy": 0.14330726861953735,
+      "epoch": 0.8561872909698997,
+      "frac_reward_zero_std": 0.75,
+      "grad_norm": 1.0059927701950073,
+      "kl": 0.05156718186723689,
+      "learning_rate": 1.4297658862876255e-06,
+      "loss": 0.0074,
+      "num_tokens": 5540323.0,
+      "reward": 0.7135416666666666,
+      "reward_std": 0.10508963962395985,
+      "rewards/accuracy_reward/mean": 0.7135416666666666,
+      "rewards/accuracy_reward/std": 0.1636701002717018,
+      "step": 1024
+    },
+    {
+      "clip_ratio/high_max": 0.0006484732912213076,
+      "clip_ratio/high_mean": 0.0003492480500426609,
+      "clip_ratio/low_mean": 0.00020737830254802248,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0005566263575929042,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 478.6875,
+      "completions/max_terminated_length": 478.6875,
+      "completions/mean_length": 417.869140625,
+      "completions/mean_terminated_length": 417.869140625,
+      "completions/min_length": 353.6875,
+      "completions/min_terminated_length": 353.6875,
+      "entropy": 0.1362351783318445,
+      "epoch": 0.9096989966555183,
+      "frac_reward_zero_std": 0.6875,
+      "grad_norm": 1.5935475826263428,
+      "kl": 0.049660728640446905,
+      "learning_rate": 1.394091415830546e-06,
+      "loss": 0.0083,
+      "num_tokens": 5888384.0,
+      "reward": 0.80859375,
+      "reward_std": 0.10727976821362972,
+      "rewards/accuracy_reward/mean": 0.80859375,
+      "rewards/accuracy_reward/std": 0.24826626759022474,
+      "step": 1088
+    },
+    {
+      "epoch": 0.919732441471572,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 481.5,
+      "eval_completions/max_terminated_length": 481.5,
+      "eval_completions/mean_length": 414.1484375,
+      "eval_completions/mean_terminated_length": 414.1484375,
+      "eval_completions/min_length": 323.0,
+      "eval_completions/min_terminated_length": 323.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 5954092.0,
+      "eval_reward": 0.6796875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6796875,
+      "eval_rewards/accuracy_reward/std": 0.46960316598415375,
+      "eval_runtime": 55.3948,
+      "eval_samples_per_second": 1.354,
+      "eval_steps_per_second": 0.036,
+      "step": 1100
+    },
+    {
+      "clip_ratio/high_max": 0.00026838650228455663,
+      "clip_ratio/high_mean": 0.00013941049781771234,
+      "clip_ratio/low_mean": 0.0001278050331166014,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00026721552645680134,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 487.38461538461536,
+      "completions/max_terminated_length": 487.38461538461536,
+      "completions/mean_length": 433.93028846153845,
+      "completions/mean_terminated_length": 433.93028846153845,
+      "completions/min_length": 371.3076923076923,
+      "completions/min_terminated_length": 371.3076923076923,
+      "entropy": 0.2032681694970681,
+      "epoch": 0.9632107023411371,
+      "frac_reward_zero_std": 0.7307692307692307,
+      "grad_norm": 1.172003149986267,
+      "kl": 0.582390272954049,
+      "learning_rate": 1.3584169453734671e-06,
+      "loss": 0.0459,
+      "num_tokens": 6243567.0,
+      "reward": 0.8245192307692307,
+      "reward_std": 0.10265854688791129,
+      "rewards/accuracy_reward/mean": 0.8245192307692307,
+      "rewards/accuracy_reward/std": 0.1861229149194864,
+      "step": 1152
+    },
+    {
+      "epoch": 1.0033444816053512,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 484.0,
+      "eval_completions/max_terminated_length": 484.0,
+      "eval_completions/mean_length": 409.8359375,
+      "eval_completions/mean_terminated_length": 409.8359375,
+      "eval_completions/min_length": 290.5,
+      "eval_completions/min_terminated_length": 290.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 6498032.0,
+      "eval_reward": 0.7265625,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.7265625,
+      "eval_rewards/accuracy_reward/std": 0.4484642744064331,
+      "eval_runtime": 56.3146,
+      "eval_samples_per_second": 1.332,
+      "eval_steps_per_second": 0.036,
+      "step": 1200
+    },
+    {
+      "clip_ratio/high_max": 4.045307287015021e-05,
+      "clip_ratio/high_mean": 2.0226536435075104e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 2.0226536435075104e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 436.0,
+      "completions/max_terminated_length": 436.0,
+      "completions/mean_length": 365.421875,
+      "completions/mean_terminated_length": 365.421875,
+      "completions/min_length": 313.5,
+      "completions/min_terminated_length": 313.5,
+      "entropy": 0.2158376593142748,
+      "epoch": 1.0167224080267558,
+      "frac_reward_zero_std": 0.75,
+      "grad_norm": 0.1277727633714676,
+      "kl": 0.15024185786023736,
+      "learning_rate": 1.3227424749163878e-06,
+      "loss": 0.0152,
+      "num_tokens": 6578326.0,
+      "reward": 0.8828125,
+      "reward_std": 0.09308473765850067,
+      "rewards/accuracy_reward/mean": 0.8828125,
+      "rewards/accuracy_reward/std": 0.18623139336705208,
+      "step": 1216
+    },
+    {
+      "clip_ratio/high_max": 9.294958454120206e-05,
+      "clip_ratio/high_mean": 4.647479227060103e-05,
+      "clip_ratio/low_mean": 9.177984338748502e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 5.565277660934953e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 478.5,
+      "completions/max_terminated_length": 478.5,
+      "completions/mean_length": 400.9921875,
+      "completions/mean_terminated_length": 400.9921875,
+      "completions/min_length": 326.5,
+      "completions/min_terminated_length": 326.5,
+      "entropy": 0.24563083425164223,
+      "epoch": 1.0702341137123745,
+      "frac_reward_zero_std": 0.625,
+      "grad_norm": 2.6599202156066895,
+      "kl": 0.1877364907413721,
+      "learning_rate": 1.2870680044593088e-06,
+      "loss": 0.0201,
+      "num_tokens": 6917794.0,
+      "reward": 0.849609375,
+      "reward_std": 0.12969218753278255,
+      "rewards/accuracy_reward/mean": 0.849609375,
+      "rewards/accuracy_reward/std": 0.2124346774071455,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0869565217391304,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 497.0,
+      "eval_completions/max_terminated_length": 497.0,
+      "eval_completions/mean_length": 404.3125,
+      "eval_completions/mean_terminated_length": 404.3125,
+      "eval_completions/min_length": 283.5,
+      "eval_completions/min_terminated_length": 283.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 7025182.0,
+      "eval_reward": 0.6875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6875,
+      "eval_rewards/accuracy_reward/std": 0.467176616191864,
+      "eval_runtime": 58.4461,
+      "eval_samples_per_second": 1.283,
+      "eval_steps_per_second": 0.034,
+      "step": 1300
+    },
+    {
+      "clip_ratio/high_max": 8.207485503093763e-05,
+      "clip_ratio/high_mean": 4.1037427515468814e-05,
+      "clip_ratio/low_mean": 6.5723751586946575e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 4.760980267416347e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 474.09090909090907,
+      "completions/max_terminated_length": 474.09090909090907,
+      "completions/mean_length": 394.9261363636364,
+      "completions/mean_terminated_length": 394.9261363636364,
+      "completions/min_length": 320.09090909090907,
+      "completions/min_terminated_length": 320.09090909090907,
+      "entropy": 0.2421540587463162,
+      "epoch": 1.1237458193979932,
+      "frac_reward_zero_std": 0.6363636363636364,
+      "grad_norm": 0.14775635302066803,
+      "kl": 0.1778960299085487,
+      "learning_rate": 1.2513935340022296e-06,
+      "loss": 0.0183,
+      "num_tokens": 7256404.0,
+      "reward": 0.71875,
+      "reward_std": 0.15349717302755875,
+      "rewards/accuracy_reward/mean": 0.71875,
+      "rewards/accuracy_reward/std": 0.3336510712450201,
+      "step": 1344
+    },
+    {
+      "epoch": 1.1705685618729098,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 512.0,
+      "eval_completions/max_terminated_length": 512.0,
+      "eval_completions/mean_length": 407.3515625,
+      "eval_completions/mean_terminated_length": 407.3515625,
+      "eval_completions/min_length": 305.5,
+      "eval_completions/min_terminated_length": 305.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 7560502.0,
+      "eval_reward": 0.71875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.71875,
+      "eval_rewards/accuracy_reward/std": 0.4518062025308609,
+      "eval_runtime": 58.6946,
+      "eval_samples_per_second": 1.278,
+      "eval_steps_per_second": 0.034,
+      "step": 1400
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 469.5,
+      "completions/max_terminated_length": 469.5,
+      "completions/mean_length": 400.265625,
+      "completions/mean_terminated_length": 400.265625,
+      "completions/min_length": 287.0,
+      "completions/min_terminated_length": 287.0,
+      "entropy": 0.23906511440873146,
+      "epoch": 1.1772575250836121,
+      "frac_reward_zero_std": 0.75,
+      "grad_norm": 0.17765256762504578,
+      "kl": 0.17118846625089645,
+      "learning_rate": 1.2157190635451505e-06,
+      "loss": 0.0136,
+      "num_tokens": 7602903.0,
+      "reward": 0.875,
+      "reward_std": 0.12909944355487823,
+      "rewards/accuracy_reward/mean": 0.875,
+      "rewards/accuracy_reward/std": 0.2199706733226776,
+      "step": 1408
+    },
+    {
+      "clip_ratio/high_max": 8.924065696191974e-05,
+      "clip_ratio/high_mean": 4.462032848095987e-05,
+      "clip_ratio/low_mean": 4.155584974796511e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 4.877591345575638e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 488.0625,
+      "completions/max_terminated_length": 488.0625,
+      "completions/mean_length": 419.373046875,
+      "completions/mean_terminated_length": 419.373046875,
+      "completions/min_length": 349.3125,
+      "completions/min_terminated_length": 349.3125,
+      "entropy": 0.2390368201304227,
+      "epoch": 1.2307692307692308,
+      "frac_reward_zero_std": 0.6875,
+      "grad_norm": 0.1258598417043686,
+      "kl": 0.15962096699513495,
+      "learning_rate": 1.1800445930880713e-06,
+      "loss": 0.0152,
+      "num_tokens": 7951686.0,
+      "reward": 0.751953125,
+      "reward_std": 0.11525858379900455,
+      "rewards/accuracy_reward/mean": 0.751953125,
+      "rewards/accuracy_reward/std": 0.30761036090552807,
+      "step": 1472
+    },
+    {
+      "epoch": 1.254180602006689,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 489.0,
+      "eval_completions/max_terminated_length": 489.0,
+      "eval_completions/mean_length": 402.109375,
+      "eval_completions/mean_terminated_length": 402.109375,
+      "eval_completions/min_length": 298.5,
+      "eval_completions/min_terminated_length": 298.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 8103527.0,
+      "eval_reward": 0.6875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6875,
+      "eval_rewards/accuracy_reward/std": 0.465938538312912,
+      "eval_runtime": 57.2645,
+      "eval_samples_per_second": 1.31,
+      "eval_steps_per_second": 0.035,
+      "step": 1500
+    },
+    {
+      "clip_ratio/high_max": 3.321839095507231e-05,
+      "clip_ratio/high_mean": 1.6609195477536155e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 1.6609195477536155e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 470.22222222222223,
+      "completions/max_terminated_length": 470.22222222222223,
+      "completions/mean_length": 394.00347222222223,
+      "completions/mean_terminated_length": 394.00347222222223,
+      "completions/min_length": 324.22222222222223,
+      "completions/min_terminated_length": 324.22222222222223,
+      "entropy": 0.23824424006872708,
+      "epoch": 1.2842809364548495,
+      "frac_reward_zero_std": 0.6111111111111112,
+      "grad_norm": 0.1226065531373024,
+      "kl": 0.16477220774524742,
+      "learning_rate": 1.1443701226309922e-06,
+      "loss": 0.0164,
+      "num_tokens": 8292504.0,
+      "reward": 0.6944444444444444,
+      "reward_std": 0.17361230651537576,
+      "rewards/accuracy_reward/mean": 0.6944444444444444,
+      "rewards/accuracy_reward/std": 0.3317638602521684,
+      "step": 1536
+    },
+    {
+      "clip_ratio/high_max": 2.729532752709929e-05,
+      "clip_ratio/high_mean": 1.3647663763549645e-05,
+      "clip_ratio/low_mean": 9.310579571319977e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 2.295824333486962e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 463.8125,
+      "completions/max_terminated_length": 463.8125,
+      "completions/mean_length": 403.76171875,
+      "completions/mean_terminated_length": 403.76171875,
+      "completions/min_length": 339.0625,
+      "completions/min_terminated_length": 339.0625,
+      "entropy": 0.2335241783875972,
+      "epoch": 1.3377926421404682,
+      "frac_reward_zero_std": 0.6875,
+      "grad_norm": 0.6611215472221375,
+      "kl": 0.14638517287676223,
+      "learning_rate": 1.108695652173913e-06,
+      "loss": 0.0139,
+      "num_tokens": 8633278.0,
+      "reward": 0.80859375,
+      "reward_std": 0.11795367486774921,
+      "rewards/accuracy_reward/mean": 0.80859375,
+      "rewards/accuracy_reward/std": 0.2854210138320923,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3377926421404682,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 489.5,
+      "eval_completions/max_terminated_length": 489.5,
+      "eval_completions/mean_length": 411.0390625,
+      "eval_completions/mean_terminated_length": 411.0390625,
+      "eval_completions/min_length": 306.0,
+      "eval_completions/min_terminated_length": 306.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 8633278.0,
+      "eval_reward": 0.7265625,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.7265625,
+      "eval_rewards/accuracy_reward/std": 0.4484642744064331,
+      "eval_runtime": 55.6551,
+      "eval_samples_per_second": 1.348,
+      "eval_steps_per_second": 0.036,
+      "step": 1600
+    },
+    {
+      "clip_ratio/high_max": 7.189887583081145e-05,
+      "clip_ratio/high_mean": 4.0447135688737035e-05,
+      "clip_ratio/low_mean": 4.950887159793638e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 4.539802284853067e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 483.125,
+      "completions/max_terminated_length": 483.125,
+      "completions/mean_length": 416.267578125,
+      "completions/mean_terminated_length": 416.267578125,
+      "completions/min_length": 343.0625,
+      "completions/min_terminated_length": 343.0625,
+      "entropy": 0.2306650117971003,
+      "epoch": 1.391304347826087,
+      "frac_reward_zero_std": 0.78125,
+      "grad_norm": 0.12204845994710922,
+      "kl": 0.1382605164544657,
+      "learning_rate": 1.0730211817168338e-06,
+      "loss": 0.0131,
+      "num_tokens": 8980599.0,
+      "reward": 0.798828125,
+      "reward_std": 0.08617058955132961,
+      "rewards/accuracy_reward/mean": 0.798828125,
+      "rewards/accuracy_reward/std": 0.25321623776108027,
+      "step": 1664
+    },
+    {
+      "epoch": 1.4214046822742474,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 483.0,
+      "eval_completions/max_terminated_length": 483.0,
+      "eval_completions/mean_length": 409.21875,
+      "eval_completions/mean_terminated_length": 409.21875,
+      "eval_completions/min_length": 306.5,
+      "eval_completions/min_terminated_length": 306.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 9169717.0,
+      "eval_reward": 0.71875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.71875,
+      "eval_rewards/accuracy_reward/std": 0.4500993341207504,
+      "eval_runtime": 54.3262,
+      "eval_samples_per_second": 1.381,
+      "eval_steps_per_second": 0.037,
+      "step": 1700
+    },
+    {
+      "clip_ratio/high_max": 2.1605737856589258e-05,
+      "clip_ratio/high_mean": 1.0802868928294629e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 1.0802868928294629e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 450.0,
+      "completions/max_terminated_length": 450.0,
+      "completions/mean_length": 392.3125,
+      "completions/mean_terminated_length": 392.3125,
+      "completions/min_length": 338.7142857142857,
+      "completions/min_terminated_length": 338.7142857142857,
+      "entropy": 0.21413783409765788,
+      "epoch": 1.4448160535117056,
+      "frac_reward_zero_std": 0.9285714285714286,
+      "grad_norm": 0.15873290598392487,
+      "kl": 0.11360293932791267,
+      "learning_rate": 1.0373467112597547e-06,
+      "loss": 0.0126,
+      "num_tokens": 9316395.0,
+      "reward": 0.9196428571428571,
+      "reward_std": 0.024397500923701694,
+      "rewards/accuracy_reward/mean": 0.9196428571428571,
+      "rewards/accuracy_reward/std": 0.10770502473626818,
+      "step": 1728
+    },
+    {
+      "clip_ratio/high_max": 6.776595000701491e-05,
+      "clip_ratio/high_mean": 3.3882975003507454e-05,
+      "clip_ratio/low_mean": 4.966624146618415e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 3.884959915012587e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 476.875,
+      "completions/max_terminated_length": 476.875,
+      "completions/mean_length": 413.81640625,
+      "completions/mean_terminated_length": 413.81640625,
+      "completions/min_length": 353.8125,
+      "completions/min_terminated_length": 353.8125,
+      "entropy": 0.2277033186983317,
+      "epoch": 1.4983277591973243,
+      "frac_reward_zero_std": 0.71875,
+      "grad_norm": 0.18755804002285004,
+      "kl": 0.15047808474628255,
+      "learning_rate": 1.0016722408026757e-06,
+      "loss": 0.0188,
+      "num_tokens": 9662429.0,
+      "reward": 0.88671875,
+      "reward_std": 0.1241629384458065,
+      "rewards/accuracy_reward/mean": 0.88671875,
+      "rewards/accuracy_reward/std": 0.18500914610922337,
+      "step": 1792
+    },
+    {
+      "epoch": 1.5050167224080266,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 480.0,
+      "eval_completions/max_terminated_length": 480.0,
+      "eval_completions/mean_length": 403.1328125,
+      "eval_completions/mean_terminated_length": 403.1328125,
+      "eval_completions/min_length": 277.0,
+      "eval_completions/min_terminated_length": 277.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 9705865.0,
+      "eval_reward": 0.7109375,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.7109375,
+      "eval_rewards/accuracy_reward/std": 0.4548390805721283,
+      "eval_runtime": 53.6282,
+      "eval_samples_per_second": 1.399,
+      "eval_steps_per_second": 0.037,
+      "step": 1800
+    },
+    {
+      "clip_ratio/high_max": 4.0922807654299375e-05,
+      "clip_ratio/high_mean": 2.5359139856716085e-05,
+      "clip_ratio/low_mean": 4.974134623937841e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 3.033327344123141e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 471.92857142857144,
+      "completions/max_terminated_length": 471.92857142857144,
+      "completions/mean_length": 415.390625,
+      "completions/mean_terminated_length": 415.390625,
+      "completions/min_length": 353.7142857142857,
+      "completions/min_terminated_length": 353.7142857142857,
+      "entropy": 0.21820709614881448,
+      "epoch": 1.551839464882943,
+      "frac_reward_zero_std": 0.8214285714285714,
+      "grad_norm": 0.9635913372039795,
+      "kl": 0.10167988536081143,
+      "learning_rate": 9.659977703455963e-07,
+      "loss": 0.0108,
+      "num_tokens": 10009128.0,
+      "reward": 0.7924107142857143,
+      "reward_std": 0.06061477107661111,
+      "rewards/accuracy_reward/mean": 0.7924107142857143,
+      "rewards/accuracy_reward/std": 0.263948740703719,
+      "step": 1856
+    },
+    {
+      "epoch": 1.588628762541806,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 489.0,
+      "eval_completions/max_terminated_length": 489.0,
+      "eval_completions/mean_length": 404.2890625,
+      "eval_completions/mean_terminated_length": 404.2890625,
+      "eval_completions/min_length": 275.5,
+      "eval_completions/min_terminated_length": 275.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 10243905.0,
+      "eval_reward": 0.6953125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6953125,
+      "eval_rewards/accuracy_reward/std": 0.46320293843746185,
+      "eval_runtime": 55.0353,
+      "eval_samples_per_second": 1.363,
+      "eval_steps_per_second": 0.036,
+      "step": 1900
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 469.4,
+      "completions/max_terminated_length": 469.4,
+      "completions/mean_length": 421.41875,
+      "completions/mean_terminated_length": 421.41875,
+      "completions/min_length": 351.0,
+      "completions/min_terminated_length": 351.0,
+      "entropy": 0.21203163117170334,
+      "epoch": 1.605351170568562,
+      "frac_reward_zero_std": 1.0,
+      "grad_norm": 0.15916664898395538,
+      "kl": 0.1034720666706562,
+      "learning_rate": 9.303232998885172e-07,
+      "loss": 0.0135,
+      "num_tokens": 10353220.0,
+      "reward": 0.8,
+      "reward_std": 0.0,
+      "rewards/accuracy_reward/mean": 0.8,
+      "rewards/accuracy_reward/std": 0.20320019721984864,
+      "step": 1920
+    },
+    {
+      "clip_ratio/high_max": 3.993847894889768e-05,
+      "clip_ratio/high_mean": 1.996923947444884e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 1.996923947444884e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 485.25,
+      "completions/max_terminated_length": 485.25,
+      "completions/mean_length": 421.94140625,
+      "completions/mean_terminated_length": 421.94140625,
+      "completions/min_length": 333.25,
+      "completions/min_terminated_length": 333.25,
+      "entropy": 0.22447309363633394,
+      "epoch": 1.6588628762541806,
+      "frac_reward_zero_std": 0.78125,
+      "grad_norm": 0.1262553632259369,
+      "kl": 0.1332392282783985,
+      "learning_rate": 8.94648829431438e-07,
+      "loss": 0.0129,
+      "num_tokens": 10703478.0,
+      "reward": 0.712890625,
+      "reward_std": 0.09271937422454357,
+      "rewards/accuracy_reward/mean": 0.712890625,
+      "rewards/accuracy_reward/std": 0.28171301539987326,
+      "step": 1984
+    },
+    {
+      "epoch": 1.6722408026755853,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 488.0,
+      "eval_completions/max_terminated_length": 488.0,
+      "eval_completions/mean_length": 409.9375,
+      "eval_completions/mean_terminated_length": 409.9375,
+      "eval_completions/min_length": 301.5,
+      "eval_completions/min_terminated_length": 301.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 10788515.0,
+      "eval_reward": 0.6953125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6953125,
+      "eval_rewards/accuracy_reward/std": 0.46193525195121765,
+      "eval_runtime": 55.3758,
+      "eval_samples_per_second": 1.354,
+      "eval_steps_per_second": 0.036,
+      "step": 2000
+    },
+    {
+      "clip_ratio/high_max": 7.684052737507348e-05,
+      "clip_ratio/high_mean": 3.842026368753674e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 3.842026368753674e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 466.75,
+      "completions/max_terminated_length": 466.75,
+      "completions/mean_length": 408.203125,
+      "completions/mean_terminated_length": 408.203125,
+      "completions/min_length": 340.0833333333333,
+      "completions/min_terminated_length": 340.0833333333333,
+      "entropy": 0.22362264028439918,
+      "epoch": 1.7123745819397993,
+      "frac_reward_zero_std": 0.6666666666666666,
+      "grad_norm": 0.12759657204151154,
+      "kl": 0.12007095486236115,
+      "learning_rate": 8.589743589743588e-07,
+      "loss": 0.0105,
+      "num_tokens": 11045953.0,
+      "reward": 0.8671875,
+      "reward_std": 0.11398912469546,
+      "rewards/accuracy_reward/mean": 0.8671875,
+      "rewards/accuracy_reward/std": 0.22504575674732527,
+      "step": 2048
+    },
+    {
+      "epoch": 1.7558528428093645,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 501.5,
+      "eval_completions/max_terminated_length": 501.5,
+      "eval_completions/mean_length": 408.03125,
+      "eval_completions/mean_terminated_length": 408.03125,
+      "eval_completions/min_length": 316.0,
+      "eval_completions/min_terminated_length": 316.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 11325630.0,
+      "eval_reward": 0.6796875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6796875,
+      "eval_rewards/accuracy_reward/std": 0.46960316598415375,
+      "eval_runtime": 56.2491,
+      "eval_samples_per_second": 1.333,
+      "eval_steps_per_second": 0.036,
+      "step": 2100
+    },
+    {
+      "clip_ratio/high_max": 5.1824211065346994e-05,
+      "clip_ratio/high_mean": 2.5912105532673497e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 2.5912105532673497e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 480.3333333333333,
+      "completions/max_terminated_length": 480.3333333333333,
+      "completions/mean_length": 419.3854166666667,
+      "completions/mean_terminated_length": 419.3854166666667,
+      "completions/min_length": 327.3333333333333,
+      "completions/min_terminated_length": 327.3333333333333,
+      "entropy": 0.21756159514188766,
+      "epoch": 1.7658862876254182,
+      "frac_reward_zero_std": 0.6666666666666666,
+      "grad_norm": 0.12993162870407104,
+      "kl": 0.1064981979628404,
+      "learning_rate": 8.232998885172798e-07,
+      "loss": 0.0143,
+      "num_tokens": 11391091.0,
+      "reward": 0.9791666666666666,
+      "reward_std": 0.08333333333333333,
+      "rewards/accuracy_reward/mean": 0.9791666666666666,
+      "rewards/accuracy_reward/std": 0.11785112818082173,
+      "step": 2112
+    },
+    {
+      "clip_ratio/high_max": 9.864267667580862e-06,
+      "clip_ratio/high_mean": 4.932133833790431e-06,
+      "clip_ratio/low_mean": 4.852484380535316e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 9.784618214325747e-06,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 479.6875,
+      "completions/max_terminated_length": 479.6875,
+      "completions/mean_length": 400.6796875,
+      "completions/mean_terminated_length": 400.6796875,
+      "completions/min_length": 332.8125,
+      "completions/min_terminated_length": 332.8125,
+      "entropy": 0.21871562604792416,
+      "epoch": 1.819397993311037,
+      "frac_reward_zero_std": 0.75,
+      "grad_norm": 0.11841385066509247,
+      "kl": 0.12797342258272693,
+      "learning_rate": 7.876254180602006e-07,
+      "loss": 0.0127,
+      "num_tokens": 11730399.0,
+      "reward": 0.82421875,
+      "reward_std": 0.11090045981109142,
+      "rewards/accuracy_reward/mean": 0.82421875,
+      "rewards/accuracy_reward/std": 0.2643054760992527,
+      "step": 2176
+    },
+    {
+      "epoch": 1.839464882943144,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 480.5,
+      "eval_completions/max_terminated_length": 480.5,
+      "eval_completions/mean_length": 409.5546875,
+      "eval_completions/mean_terminated_length": 409.5546875,
+      "eval_completions/min_length": 296.0,
+      "eval_completions/min_terminated_length": 296.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 11860817.0,
+      "eval_reward": 0.75,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.75,
+      "eval_rewards/accuracy_reward/std": 0.43491509556770325,
+      "eval_runtime": 53.894,
+      "eval_samples_per_second": 1.392,
+      "eval_steps_per_second": 0.037,
+      "step": 2200
+    },
+    {
+      "clip_ratio/high_max": 4.147470317548141e-05,
+      "clip_ratio/high_mean": 2.0737351587740704e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 2.0737351587740704e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 486.0,
+      "completions/max_terminated_length": 486.0,
+      "completions/mean_length": 419.878125,
+      "completions/mean_terminated_length": 419.878125,
+      "completions/min_length": 339.9,
+      "completions/min_terminated_length": 339.9,
+      "entropy": 0.2102437574416399,
+      "epoch": 1.8729096989966556,
+      "frac_reward_zero_std": 0.9,
+      "grad_norm": 0.09409788995981216,
+      "kl": 0.09403365906327962,
+      "learning_rate": 7.519509476031214e-07,
+      "loss": 0.009,
+      "num_tokens": 12078906.0,
+      "reward": 0.79375,
+      "reward_std": 0.025,
+      "rewards/accuracy_reward/mean": 0.79375,
+      "rewards/accuracy_reward/std": 0.23855553567409515,
+      "step": 2240
+    },
+    {
+      "epoch": 1.9230769230769231,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 495.0,
+      "eval_completions/max_terminated_length": 495.0,
+      "eval_completions/mean_length": 411.8125,
+      "eval_completions/mean_terminated_length": 411.8125,
+      "eval_completions/min_length": 308.0,
+      "eval_completions/min_terminated_length": 308.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 12400892.0,
+      "eval_reward": 0.6875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6875,
+      "eval_rewards/accuracy_reward/std": 0.46219292283058167,
+      "eval_runtime": 55.4173,
+      "eval_samples_per_second": 1.353,
+      "eval_steps_per_second": 0.036,
+      "step": 2300
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 467.0,
+      "completions/max_terminated_length": 467.0,
+      "completions/mean_length": 437.0625,
+      "completions/mean_terminated_length": 437.0625,
+      "completions/min_length": 359.0,
+      "completions/min_terminated_length": 359.0,
+      "entropy": 0.23047670722007751,
+      "epoch": 1.9264214046822743,
+      "frac_reward_zero_std": 1.0,
+      "grad_norm": 0.20002007484436035,
+      "kl": 0.11390126869082451,
+      "learning_rate": 7.162764771460424e-07,
+      "loss": 0.0122,
+      "num_tokens": 12423262.0,
+      "reward": 1.0,
+      "reward_std": 0.0,
+      "rewards/accuracy_reward/mean": 1.0,
+      "rewards/accuracy_reward/std": 0.0,
+      "step": 2304
+    },
+    {
+      "clip_ratio/high_max": 2.8518005819933023e-05,
+      "clip_ratio/high_mean": 1.4259002909966512e-05,
+      "clip_ratio/low_mean": 4.495109351410065e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 1.8754112261376576e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 473.0625,
+      "completions/max_terminated_length": 473.0625,
+      "completions/mean_length": 413.14453125,
+      "completions/mean_terminated_length": 413.14453125,
+      "completions/min_length": 345.125,
+      "completions/min_terminated_length": 345.125,
+      "entropy": 0.21623935969546437,
+      "epoch": 1.979933110367893,
+      "frac_reward_zero_std": 0.8125,
+      "grad_norm": 0.15597644448280334,
+      "kl": 0.11163879834930412,
+      "learning_rate": 6.806020066889632e-07,
+      "loss": 0.0159,
+      "num_tokens": 12768968.0,
+      "reward": 0.677734375,
+      "reward_std": 0.07654774375259876,
+      "rewards/accuracy_reward/mean": 0.677734375,
+      "rewards/accuracy_reward/std": 0.33232067339122295,
+      "step": 2368
+    },
+    {
+      "epoch": 2.0066889632107023,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 485.0,
+      "eval_completions/max_terminated_length": 485.0,
+      "eval_completions/mean_length": 409.28125,
+      "eval_completions/mean_terminated_length": 409.28125,
+      "eval_completions/min_length": 292.5,
+      "eval_completions/min_terminated_length": 292.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 12941163.0,
+      "eval_reward": 0.6953125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6953125,
+      "eval_rewards/accuracy_reward/std": 0.46320293843746185,
+      "eval_runtime": 54.8202,
+      "eval_samples_per_second": 1.368,
+      "eval_steps_per_second": 0.036,
+      "step": 2400
+    },
+    {
+      "clip_ratio/high_max": 7.531194933108054e-05,
+      "clip_ratio/high_mean": 3.765597466554027e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 3.765597466554027e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 477.125,
+      "completions/max_terminated_length": 477.125,
+      "completions/mean_length": 418.265625,
+      "completions/mean_terminated_length": 418.265625,
+      "completions/min_length": 350.75,
+      "completions/min_terminated_length": 350.75,
+      "entropy": 0.21743145305663347,
+      "epoch": 2.0334448160535117,
+      "frac_reward_zero_std": 0.75,
+      "grad_norm": 0.14243286848068237,
+      "kl": 0.12919681682251394,
+      "learning_rate": 6.449275362318841e-07,
+      "loss": 0.011,
+      "num_tokens": 13115359.0,
+      "reward": 0.8125,
+      "reward_std": 0.08636415377259254,
+      "rewards/accuracy_reward/mean": 0.8125,
+      "rewards/accuracy_reward/std": 0.2475989516824484,
+      "step": 2432
+    },
+    {
+      "clip_ratio/high_max": 4.564970004139468e-05,
+      "clip_ratio/high_mean": 2.282485002069734e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 2.282485002069734e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 474.5,
+      "completions/max_terminated_length": 474.5,
+      "completions/mean_length": 420.361328125,
+      "completions/mean_terminated_length": 420.361328125,
+      "completions/min_length": 346.4375,
+      "completions/min_terminated_length": 346.4375,
+      "entropy": 0.20763462036848068,
+      "epoch": 2.0869565217391304,
+      "frac_reward_zero_std": 0.875,
+      "grad_norm": 0.05451633036136627,
+      "kl": 0.08778653664921876,
+      "learning_rate": 6.092530657748049e-07,
+      "loss": 0.0081,
+      "num_tokens": 13464632.0,
+      "reward": 0.73828125,
+      "reward_std": 0.0457378551363945,
+      "rewards/accuracy_reward/mean": 0.73828125,
+      "rewards/accuracy_reward/std": 0.17558534815907478,
+      "step": 2496
+    },
+    {
+      "epoch": 2.0903010033444818,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 487.0,
+      "eval_completions/max_terminated_length": 487.0,
+      "eval_completions/mean_length": 405.203125,
+      "eval_completions/mean_terminated_length": 405.203125,
+      "eval_completions/min_length": 277.0,
+      "eval_completions/min_terminated_length": 277.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 13486071.0,
+      "eval_reward": 0.71875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.71875,
+      "eval_rewards/accuracy_reward/std": 0.4518062025308609,
+      "eval_runtime": 55.4822,
+      "eval_samples_per_second": 1.352,
+      "eval_steps_per_second": 0.036,
+      "step": 2500
+    },
+    {
+      "clip_ratio/high_max": 7.478842744603753e-05,
+      "clip_ratio/high_mean": 4.234863299643621e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 4.234863299643621e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 472.6,
+      "completions/max_terminated_length": 472.6,
+      "completions/mean_length": 407.50208333333336,
+      "completions/mean_terminated_length": 407.50208333333336,
+      "completions/min_length": 320.0,
+      "completions/min_terminated_length": 320.0,
+      "entropy": 0.21952359055479367,
+      "epoch": 2.140468227424749,
+      "frac_reward_zero_std": 0.7333333333333333,
+      "grad_norm": 0.4182104468345642,
+      "kl": 0.11582428424929579,
+      "learning_rate": 5.735785953177257e-07,
+      "loss": 0.0094,
+      "num_tokens": 13807544.0,
+      "reward": 0.7583333333333333,
+      "reward_std": 0.09178131222724914,
+      "rewards/accuracy_reward/mean": 0.7583333333333333,
+      "rewards/accuracy_reward/std": 0.2630104998747508,
+      "step": 2560
+    },
+    {
+      "epoch": 2.1739130434782608,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 490.0,
+      "eval_completions/max_terminated_length": 490.0,
+      "eval_completions/mean_length": 410.9453125,
+      "eval_completions/mean_terminated_length": 410.9453125,
+      "eval_completions/min_length": 314.5,
+      "eval_completions/min_terminated_length": 314.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 14028729.0,
+      "eval_reward": 0.703125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.703125,
+      "eval_rewards/accuracy_reward/std": 0.45757468044757843,
+      "eval_runtime": 55.028,
+      "eval_samples_per_second": 1.363,
+      "eval_steps_per_second": 0.036,
+      "step": 2600
+    },
+    {
+      "clip_ratio/high_max": 5.208333216918012e-05,
+      "clip_ratio/high_mean": 2.604166608459006e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 2.604166608459006e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 481.1666666666667,
+      "completions/max_terminated_length": 481.1666666666667,
+      "completions/mean_length": 409.8020833333333,
+      "completions/mean_terminated_length": 409.8020833333333,
+      "completions/min_length": 329.8333333333333,
+      "completions/min_terminated_length": 329.8333333333333,
+      "entropy": 0.21271449948350588,
+      "epoch": 2.1939799331103678,
+      "frac_reward_zero_std": 0.75,
+      "grad_norm": 0.1287287324666977,
+      "kl": 0.10981122233594458,
+      "learning_rate": 5.379041248606467e-07,
+      "loss": 0.0116,
+      "num_tokens": 14157731.0,
+      "reward": 0.6614583333333334,
+      "reward_std": 0.10372315843900044,
+      "rewards/accuracy_reward/mean": 0.6614583333333334,
+      "rewards/accuracy_reward/std": 0.2969101319710414,
+      "step": 2624
+    },
+    {
+      "clip_ratio/high_max": 2.8780045795429032e-05,
+      "clip_ratio/high_mean": 1.4390022897714516e-05,
+      "clip_ratio/low_mean": 4.342690317571396e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 1.8732713215285912e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 480.0625,
+      "completions/max_terminated_length": 480.0625,
+      "completions/mean_length": 421.123046875,
+      "completions/mean_terminated_length": 421.123046875,
+      "completions/min_length": 357.0,
+      "completions/min_terminated_length": 357.0,
+      "entropy": 0.21469741920009255,
+      "epoch": 2.2474916387959865,
+      "frac_reward_zero_std": 0.6875,
+      "grad_norm": 0.14622721076011658,
+      "kl": 0.12252736231312156,
+      "learning_rate": 5.022296544035675e-07,
+      "loss": 0.0168,
+      "num_tokens": 14507554.0,
+      "reward": 0.7890625,
+      "reward_std": 0.11960325017571449,
+      "rewards/accuracy_reward/mean": 0.7890625,
+      "rewards/accuracy_reward/std": 0.25777294766157866,
+      "step": 2688
+    },
+    {
+      "epoch": 2.25752508361204,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 482.0,
+      "eval_completions/max_terminated_length": 482.0,
+      "eval_completions/mean_length": 409.46875,
+      "eval_completions/mean_terminated_length": 409.46875,
+      "eval_completions/min_length": 295.0,
+      "eval_completions/min_terminated_length": 295.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 14570788.0,
+      "eval_reward": 0.7109375,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.7109375,
+      "eval_rewards/accuracy_reward/std": 0.4561667740345001,
+      "eval_runtime": 54.3849,
+      "eval_samples_per_second": 1.379,
+      "eval_steps_per_second": 0.037,
+      "step": 2700
+    },
+    {
+      "clip_ratio/high_max": 8.321751710564758e-05,
+      "clip_ratio/high_mean": 4.160875855282379e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 4.160875855282379e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 470.46153846153845,
+      "completions/max_terminated_length": 470.46153846153845,
+      "completions/mean_length": 406.8245192307692,
+      "completions/mean_terminated_length": 406.8245192307692,
+      "completions/min_length": 338.0769230769231,
+      "completions/min_terminated_length": 338.0769230769231,
+      "entropy": 0.20195185393095016,
+      "epoch": 2.3010033444816056,
+      "frac_reward_zero_std": 0.8076923076923077,
+      "grad_norm": 0.10907543450593948,
+      "kl": 0.11125540396628472,
+      "learning_rate": 4.665551839464883e-07,
+      "loss": 0.0098,
+      "num_tokens": 14848971.0,
+      "reward": 0.9206730769230769,
+      "reward_std": 0.0706413984298706,
+      "rewards/accuracy_reward/mean": 0.9206730769230769,
+      "rewards/accuracy_reward/std": 0.14741014861143553,
+      "step": 2752
+    },
+    {
+      "epoch": 2.3411371237458196,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 490.5,
+      "eval_completions/max_terminated_length": 490.5,
+      "eval_completions/mean_length": 411.5078125,
+      "eval_completions/mean_terminated_length": 411.5078125,
+      "eval_completions/min_length": 291.5,
+      "eval_completions/min_terminated_length": 291.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 15102974.0,
+      "eval_reward": 0.703125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.703125,
+      "eval_rewards/accuracy_reward/std": 0.46017006039619446,
+      "eval_runtime": 55.6132,
+      "eval_samples_per_second": 1.349,
+      "eval_steps_per_second": 0.036,
+      "step": 2800
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 479.5,
+      "completions/max_terminated_length": 479.5,
+      "completions/mean_length": 419.90625,
+      "completions/mean_terminated_length": 419.90625,
+      "completions/min_length": 302.75,
+      "completions/min_terminated_length": 302.75,
+      "entropy": 0.20193158090114594,
+      "epoch": 2.3545150501672243,
+      "frac_reward_zero_std": 1.0,
+      "grad_norm": 0.11240160465240479,
+      "kl": 0.08153703552670777,
+      "learning_rate": 4.308807134894091e-07,
+      "loss": 0.0101,
+      "num_tokens": 15190242.0,
+      "reward": 0.625,
+      "reward_std": 0.0,
+      "rewards/accuracy_reward/mean": 0.625,
+      "rewards/accuracy_reward/std": 0.1270001232624054,
+      "step": 2816
+    },
+    {
+      "clip_ratio/high_max": 1.9505164345900994e-05,
+      "clip_ratio/high_mean": 9.752582172950497e-06,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 9.752582172950497e-06,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 506.125,
+      "completions/max_terminated_length": 506.125,
+      "completions/mean_length": 423.416015625,
+      "completions/mean_terminated_length": 423.416015625,
+      "completions/min_length": 333.1875,
+      "completions/min_terminated_length": 333.1875,
+      "entropy": 0.21712527703493834,
+      "epoch": 2.408026755852843,
+      "frac_reward_zero_std": 0.875,
+      "grad_norm": 0.1890118569135666,
+      "kl": 0.10883855499560013,
+      "learning_rate": 3.9520624303232996e-07,
+      "loss": 0.0118,
+      "num_tokens": 15541351.0,
+      "reward": 0.849609375,
+      "reward_std": 0.043847277760505676,
+      "rewards/accuracy_reward/mean": 0.849609375,
+      "rewards/accuracy_reward/std": 0.09456285089254379,
+      "step": 2880
+    },
+    {
+      "epoch": 2.4247491638795986,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 497.0,
+      "eval_completions/max_terminated_length": 497.0,
+      "eval_completions/mean_length": 410.9921875,
+      "eval_completions/mean_terminated_length": 410.9921875,
+      "eval_completions/min_length": 258.5,
+      "eval_completions/min_terminated_length": 258.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 15648422.0,
+      "eval_reward": 0.6875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6875,
+      "eval_rewards/accuracy_reward/std": 0.465938538312912,
+      "eval_runtime": 56.9931,
+      "eval_samples_per_second": 1.316,
+      "eval_steps_per_second": 0.035,
+      "step": 2900
+    },
+    {
+      "clip_ratio/high_max": 2.8449102749370716e-05,
+      "clip_ratio/high_mean": 1.4224551374685358e-05,
+      "clip_ratio/low_mean": 6.500935838134452e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 2.072548721281981e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 474.45454545454544,
+      "completions/max_terminated_length": 474.45454545454544,
+      "completions/mean_length": 420.01704545454544,
+      "completions/mean_terminated_length": 420.01704545454544,
+      "completions/min_length": 370.1818181818182,
+      "completions/min_terminated_length": 370.1818181818182,
+      "entropy": 0.20619109814817255,
+      "epoch": 2.4615384615384617,
+      "frac_reward_zero_std": 0.7727272727272727,
+      "grad_norm": 0.18583235144615173,
+      "kl": 0.1027019578455524,
+      "learning_rate": 3.595317725752508e-07,
+      "loss": 0.0091,
+      "num_tokens": 15888348.0,
+      "reward": 0.8892045454545454,
+      "reward_std": 0.08273993567986922,
+      "rewards/accuracy_reward/mean": 0.8892045454545454,
+      "rewards/accuracy_reward/std": 0.15522026880220932,
+      "step": 2944
+    },
+    {
+      "epoch": 2.508361204013378,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 484.5,
+      "eval_completions/max_terminated_length": 484.5,
+      "eval_completions/mean_length": 407.7890625,
+      "eval_completions/mean_terminated_length": 407.7890625,
+      "eval_completions/min_length": 260.0,
+      "eval_completions/min_terminated_length": 260.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 16190533.0,
+      "eval_reward": 0.703125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.703125,
+      "eval_rewards/accuracy_reward/std": 0.4591996520757675,
+      "eval_runtime": 55.9308,
+      "eval_samples_per_second": 1.341,
+      "eval_steps_per_second": 0.036,
+      "step": 3000
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 485.5,
+      "completions/max_terminated_length": 485.5,
+      "completions/mean_length": 397.703125,
+      "completions/mean_terminated_length": 397.703125,
+      "completions/min_length": 323.0,
+      "completions/min_terminated_length": 323.0,
+      "entropy": 0.1908447314053774,
+      "epoch": 2.5150501672240804,
+      "frac_reward_zero_std": 0.5,
+      "grad_norm": 0.294365257024765,
+      "kl": 0.06209766957908869,
+      "learning_rate": 3.238573021181717e-07,
+      "loss": 0.0104,
+      "num_tokens": 16232754.0,
+      "reward": 0.96875,
+      "reward_std": 0.125,
+      "rewards/accuracy_reward/mean": 0.96875,
+      "rewards/accuracy_reward/std": 0.1767766922712326,
+      "step": 3008
+    },
+    {
+      "clip_ratio/high_max": 2.0032051907037385e-05,
+      "clip_ratio/high_mean": 1.0016025953518692e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 1.0016025953518692e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 474.75,
+      "completions/max_terminated_length": 474.75,
+      "completions/mean_length": 414.232421875,
+      "completions/mean_terminated_length": 414.232421875,
+      "completions/min_length": 342.9375,
+      "completions/min_terminated_length": 342.9375,
+      "entropy": 0.20259337592869997,
+      "epoch": 2.568561872909699,
+      "frac_reward_zero_std": 0.8125,
+      "grad_norm": 0.11983966082334518,
+      "kl": 0.08080932262237184,
+      "learning_rate": 2.881828316610925e-07,
+      "loss": 0.0084,
+      "num_tokens": 16578905.0,
+      "reward": 0.787109375,
+      "reward_std": 0.05452118441462517,
+      "rewards/accuracy_reward/mean": 0.787109375,
+      "rewards/accuracy_reward/std": 0.1910215886309743,
+      "step": 3072
+    },
+    {
+      "epoch": 2.591973244147157,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 494.0,
+      "eval_completions/max_terminated_length": 494.0,
+      "eval_completions/mean_length": 413.71875,
+      "eval_completions/mean_terminated_length": 413.71875,
+      "eval_completions/min_length": 321.5,
+      "eval_completions/min_terminated_length": 321.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 16730197.0,
+      "eval_reward": 0.703125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.703125,
+      "eval_rewards/accuracy_reward/std": 0.45757468044757843,
+      "eval_runtime": 56.085,
+      "eval_samples_per_second": 1.337,
+      "eval_steps_per_second": 0.036,
+      "step": 3100
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 477.6666666666667,
+      "completions/max_terminated_length": 477.6666666666667,
+      "completions/mean_length": 407.5729166666667,
+      "completions/mean_terminated_length": 407.5729166666667,
+      "completions/min_length": 336.8888888888889,
+      "completions/min_terminated_length": 336.8888888888889,
+      "entropy": 0.2053673474325074,
+      "epoch": 2.6220735785953178,
+      "frac_reward_zero_std": 0.8333333333333334,
+      "grad_norm": 0.09832081943750381,
+      "kl": 0.08688689861446619,
+      "learning_rate": 2.5250836120401335e-07,
+      "loss": 0.0105,
+      "num_tokens": 16923050.0,
+      "reward": 0.7395833333333334,
+      "reward_std": 0.05017293824089898,
+      "rewards/accuracy_reward/mean": 0.7395833333333334,
+      "rewards/accuracy_reward/std": 0.20655767785178292,
+      "step": 3136
+    },
+    {
+      "clip_ratio/high_max": 8.985049134935252e-06,
+      "clip_ratio/high_mean": 4.492524567467626e-06,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 4.492524567467626e-06,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 479.0,
+      "completions/max_terminated_length": 479.0,
+      "completions/mean_length": 424.021484375,
+      "completions/mean_terminated_length": 424.021484375,
+      "completions/min_length": 347.125,
+      "completions/min_terminated_length": 347.125,
+      "entropy": 0.20460248668678105,
+      "epoch": 2.6755852842809364,
+      "frac_reward_zero_std": 0.875,
+      "grad_norm": 0.14690326154232025,
+      "kl": 0.09142500828602351,
+      "learning_rate": 2.1683389074693424e-07,
+      "loss": 0.0088,
+      "num_tokens": 17274293.0,
+      "reward": 0.765625,
+      "reward_std": 0.04243633709847927,
+      "rewards/accuracy_reward/mean": 0.765625,
+      "rewards/accuracy_reward/std": 0.18196979723870754,
+      "step": 3200
+    },
+    {
+      "epoch": 2.6755852842809364,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 475.5,
+      "eval_completions/max_terminated_length": 475.5,
+      "eval_completions/mean_length": 407.9765625,
+      "eval_completions/mean_terminated_length": 407.9765625,
+      "eval_completions/min_length": 306.5,
+      "eval_completions/min_terminated_length": 306.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 17274293.0,
+      "eval_reward": 0.6953125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6953125,
+      "eval_rewards/accuracy_reward/std": 0.4638346880674362,
+      "eval_runtime": 53.8229,
+      "eval_samples_per_second": 1.393,
+      "eval_steps_per_second": 0.037,
+      "step": 3200
+    },
+    {
+      "clip_ratio/high_max": 9.895503353618551e-06,
+      "clip_ratio/high_mean": 4.9477516768092755e-06,
+      "clip_ratio/low_mean": 4.5767428673570976e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 9.524494544166373e-06,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 471.1875,
+      "completions/max_terminated_length": 471.1875,
+      "completions/mean_length": 406.287109375,
+      "completions/mean_terminated_length": 406.287109375,
+      "completions/min_length": 330.75,
+      "completions/min_terminated_length": 330.75,
+      "entropy": 0.193967598490417,
+      "epoch": 2.729096989966555,
+      "frac_reward_zero_std": 0.90625,
+      "grad_norm": 0.15071481466293335,
+      "kl": 0.08262389054289088,
+      "learning_rate": 1.8115942028985507e-07,
+      "loss": 0.007,
+      "num_tokens": 17616488.0,
+      "reward": 0.90625,
+      "reward_std": 0.03996079042553902,
+      "rewards/accuracy_reward/mean": 0.90625,
+      "rewards/accuracy_reward/std": 0.11948453821241856,
+      "step": 3264
+    },
+    {
+      "epoch": 2.759197324414716,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 482.0,
+      "eval_completions/max_terminated_length": 482.0,
+      "eval_completions/mean_length": 409.1171875,
+      "eval_completions/mean_terminated_length": 409.1171875,
+      "eval_completions/min_length": 288.5,
+      "eval_completions/min_terminated_length": 288.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 17808921.0,
+      "eval_reward": 0.7109375,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.7109375,
+      "eval_rewards/accuracy_reward/std": 0.4561667740345001,
+      "eval_runtime": 54.9248,
+      "eval_samples_per_second": 1.366,
+      "eval_steps_per_second": 0.036,
+      "step": 3300
+    },
+    {
+      "clip_ratio/high_max": 4.422420793811658e-05,
+      "clip_ratio/high_mean": 2.211210396905829e-05,
+      "clip_ratio/low_mean": 1.0725010984710284e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 3.2837114953768575e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 471.7142857142857,
+      "completions/max_terminated_length": 471.7142857142857,
+      "completions/mean_length": 405.16964285714283,
+      "completions/mean_terminated_length": 405.16964285714283,
+      "completions/min_length": 344.85714285714283,
+      "completions/min_terminated_length": 344.85714285714283,
+      "entropy": 0.20200447046330997,
+      "epoch": 2.782608695652174,
+      "frac_reward_zero_std": 0.6428571428571429,
+      "grad_norm": 0.13082604110240936,
+      "kl": 0.1005759423465601,
+      "learning_rate": 1.454849498327759e-07,
+      "loss": 0.0146,
+      "num_tokens": 17958399.0,
+      "reward": 0.7142857142857143,
+      "reward_std": 0.1647304126194545,
+      "rewards/accuracy_reward/mean": 0.7142857142857143,
+      "rewards/accuracy_reward/std": 0.260468738419669,
+      "step": 3328
+    },
+    {
+      "clip_ratio/high_max": 1.0028883480117656e-05,
+      "clip_ratio/high_mean": 5.014441740058828e-06,
+      "clip_ratio/low_mean": 4.266793894203147e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 9.281235634261975e-06,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 481.8125,
+      "completions/max_terminated_length": 481.8125,
+      "completions/mean_length": 418.5859375,
+      "completions/mean_terminated_length": 418.5859375,
+      "completions/min_length": 341.9375,
+      "completions/min_terminated_length": 341.9375,
+      "entropy": 0.20107114291749895,
+      "epoch": 2.8361204013377925,
+      "frac_reward_zero_std": 0.90625,
+      "grad_norm": 0.18976449966430664,
+      "kl": 0.08409578536520712,
+      "learning_rate": 1.0981047937569676e-07,
+      "loss": 0.0092,
+      "num_tokens": 18306923.0,
+      "reward": 0.8671875,
+      "reward_std": 0.02960042469203472,
+      "rewards/accuracy_reward/mean": 0.8671875,
+      "rewards/accuracy_reward/std": 0.12586778588593006,
+      "step": 3392
+    },
+    {
+      "epoch": 2.842809364548495,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 480.5,
+      "eval_completions/max_terminated_length": 480.5,
+      "eval_completions/mean_length": 409.984375,
+      "eval_completions/mean_terminated_length": 409.984375,
+      "eval_completions/min_length": 312.0,
+      "eval_completions/min_terminated_length": 312.0,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 18348241.0,
+      "eval_reward": 0.6875,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6875,
+      "eval_rewards/accuracy_reward/std": 0.46219292283058167,
+      "eval_runtime": 54.1878,
+      "eval_samples_per_second": 1.384,
+      "eval_steps_per_second": 0.037,
+      "step": 3400
+    },
+    {
+      "clip_ratio/high_max": 2.266245220588254e-05,
+      "clip_ratio/high_mean": 1.133122610294127e-05,
+      "clip_ratio/low_mean": 5.305152236750084e-06,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 1.6636378339691354e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 482.2142857142857,
+      "completions/max_terminated_length": 482.2142857142857,
+      "completions/mean_length": 414.65401785714283,
+      "completions/mean_terminated_length": 414.65401785714283,
+      "completions/min_length": 347.7142857142857,
+      "completions/min_terminated_length": 347.7142857142857,
+      "entropy": 0.20981859575424874,
+      "epoch": 2.8896321070234112,
+      "frac_reward_zero_std": 0.7857142857142857,
+      "grad_norm": 1.1070315837860107,
+      "kl": 0.10563170131562012,
+      "learning_rate": 7.41360089186176e-08,
+      "loss": 0.0135,
+      "num_tokens": 18651446.0,
+      "reward": 0.7857142857142857,
+      "reward_std": 0.0999031024319785,
+      "rewards/accuracy_reward/mean": 0.7857142857142857,
+      "rewards/accuracy_reward/std": 0.2894010437386377,
+      "step": 3456
+    },
+    {
+      "epoch": 2.9264214046822743,
+      "eval_completions/clipped_ratio": 0.0,
+      "eval_completions/max_length": 479.5,
+      "eval_completions/max_terminated_length": 479.5,
+      "eval_completions/mean_length": 409.65625,
+      "eval_completions/mean_terminated_length": 409.65625,
+      "eval_completions/min_length": 302.5,
+      "eval_completions/min_terminated_length": 302.5,
+      "eval_frac_reward_zero_std": 0.0,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 18887628.0,
+      "eval_reward": 0.6953125,
+      "eval_reward_std": 0.0,
+      "eval_rewards/accuracy_reward/mean": 0.6953125,
+      "eval_rewards/accuracy_reward/std": 0.46193525195121765,
+      "eval_runtime": 53.6776,
+      "eval_samples_per_second": 1.397,
+      "eval_steps_per_second": 0.037,
+      "step": 3500
+    },
+    {
+      "clip_ratio/high_max": 6.487761856988072e-05,
+      "clip_ratio/high_mean": 3.243880928494036e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 3.243880928494036e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 465.0,
+      "completions/max_terminated_length": 465.0,
+      "completions/mean_length": 402.0875,
+      "completions/mean_terminated_length": 402.0875,
+      "completions/min_length": 321.8,
+      "completions/min_terminated_length": 321.8,
+      "entropy": 0.2007051318883896,
+      "epoch": 2.94314381270903,
+      "frac_reward_zero_std": 0.9,
+      "grad_norm": 0.10725326836109161,
+      "kl": 0.09607526948675513,
+      "learning_rate": 3.846153846153846e-08,
+      "loss": 0.0081,
+      "num_tokens": 18993898.0,
+      "reward": 0.76875,
+      "reward_std": 0.047871357202529906,
+      "rewards/accuracy_reward/mean": 0.76875,
+      "rewards/accuracy_reward/std": 0.1981118381023407,
+      "step": 3520
+    },
+    {
+      "clip_ratio/high_max": 2.1599041247100104e-05,
+      "clip_ratio/high_mean": 1.0799520623550052e-05,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 1.0799520623550052e-05,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 480.0,
+      "completions/max_terminated_length": 480.0,
+      "completions/mean_length": 411.005859375,
+      "completions/mean_terminated_length": 411.005859375,
+      "completions/min_length": 343.625,
+      "completions/min_terminated_length": 343.625,
+      "entropy": 0.2045821927022189,
+      "epoch": 2.9966555183946486,
+      "frac_reward_zero_std": 0.84375,
+      "grad_norm": 0.10782357305288315,
+      "kl": 0.08409121469594538,
+      "learning_rate": 2.787068004459309e-09,
+      "loss": 0.0098,
+      "num_tokens": 19338349.0,
+      "reward": 0.880859375,
+      "reward_std": 0.05523413047194481,
+      "rewards/accuracy_reward/mean": 0.880859375,
+      "rewards/accuracy_reward/std": 0.17677514534443617,
+      "step": 3584
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 484.0,
+      "completions/max_terminated_length": 484.0,
+      "completions/mean_length": 422.125,
+      "completions/mean_terminated_length": 422.125,
+      "completions/min_length": 360.0,
+      "completions/min_terminated_length": 360.0,
+      "entropy": 0.20116707310080528,
+      "epoch": 3.0,
+      "frac_reward_zero_std": 1.0,
+      "kl": 0.06294638197869062,
+      "num_tokens": 19360225.0,
+      "reward": 0.5,
+      "reward_std": 0.0,
+      "rewards/accuracy_reward/mean": 0.5,
+      "rewards/accuracy_reward/std": 0.5080004930496216,
+      "step": 3588,
+      "total_flos": 0.0,
+      "train_loss": 0.010405275567721289,
+      "train_runtime": 34829.3301,
+      "train_samples_per_second": 0.052,
+      "train_steps_per_second": 0.103
+    }
+  ],
+  "logging_steps": 64,
+  "max_steps": 3588,
+  "num_input_tokens_seen": 19360225,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

Blood/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/added_tokens.json

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

Blood/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

Blood/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/config.json

@@ -0,0 +1,66 @@
+{
+  "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,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Blood/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/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": "4.55.2"
+}

Blood/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/model.safetensors.index.json

@@ -0,0 +1,443 @@
+{
+  "metadata": {
+    "total_parameters": 3085938688,
+    "total_size": 6794207232
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "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"
+  }
+}

Blood/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/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
+  }
+}

Blood/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 2048,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

Blood/seed_15/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/added_tokens.json

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

Blood/seed_15/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

Blood/seed_15/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/config.json

@@ -0,0 +1,66 @@
+{
+  "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,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Blood/seed_15/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/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": "4.55.2"
+}

Blood/seed_15/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/model.safetensors.index.json

@@ -0,0 +1,443 @@
+{
+  "metadata": {
+    "total_parameters": 3085938688,
+    "total_size": 6794207232
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "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"
+  }
+}

Blood/seed_15/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 2048,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

Blood/seed_2/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/README.md

@@ -0,0 +1,67 @@
+---
+library_name: transformers
+model_name: LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized
+tags:
+- generated_from_trainer
+- grpo
+- trl
+licence: license
+---
+
+# Model Card for LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized
+
+This model is a fine-tuned version of [None](https://huggingface.co/None).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/baileyyeah326/MLLM/runs/4xam9tu2) 
+
+
+This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
+
+### Framework versions
+
+- TRL: 0.21.0
+- Transformers: 4.55.2
+- Pytorch: 2.4.0+cu124
+- Datasets: 4.0.0
+- Tokenizers: 0.21.4
+
+## Citations
+
+Cite GRPO as:
+
+```bibtex
+@article{zhihong2024deepseekmath,
+    title        = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
+    author       = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
+    year         = 2024,
+    eprint       = {arXiv:2402.03300},
+}
+
+```
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

Blood/seed_2/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/added_tokens.json

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

Blood/seed_2/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

Blood/seed_2/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/config.json

@@ -0,0 +1,66 @@
+{
+  "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,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.2",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Blood/seed_2/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/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": "4.55.2"
+}

Blood/seed_2/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/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": "4.55.2"
+}

Blood/seed_2/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/model.safetensors.index.json

@@ -0,0 +1,443 @@
+{
+  "metadata": {
+    "total_parameters": 3085938688,
+    "total_size": 6794207232
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "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"
+  }
+}

Blood/seed_2/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/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
+  }
+}

Blood/seed_2/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 2048,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

Blood/seed_4/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/README.md

@@ -0,0 +1,67 @@
+---
+library_name: transformers
+model_name: LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized
+tags:
+- generated_from_trainer
+- grpo
+- trl
+licence: license
+---
+
+# Model Card for LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized
+
+This model is a fine-tuned version of [None](https://huggingface.co/None).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/baileyyeah326/MLLM/runs/vdh21tyh) 
+
+
+This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
+
+### Framework versions
+
+- TRL: 0.21.0
+- Transformers: 4.55.2
+- Pytorch: 2.4.0+cu124
+- Datasets: 4.0.0
+- Tokenizers: 0.21.4
+
+## Citations
+
+Cite GRPO as:
+
+```bibtex
+@article{zhihong2024deepseekmath,
+    title        = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
+    author       = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
+    year         = 2024,
+    eprint       = {arXiv:2402.03300},
+}
+
+```
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

Blood/seed_4/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/added_tokens.json

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

Blood/seed_4/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

Blood/seed_4/GRPO/LLM-Qwen-2.5-3B-GRPO-decision-tree-Blood-serialized/config.json

@@ -0,0 +1,66 @@
+{
+  "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,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.2",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Blood/seed_4/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/added_tokens.json

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

Blood/seed_4/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

Blood/seed_4/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/config.json

@@ -0,0 +1,66 @@
+{
+  "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,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Blood/seed_4/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/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": "4.55.2"
+}

Blood/seed_50/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/added_tokens.json

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

Blood/seed_50/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

Blood/seed_50/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/config.json

@@ -0,0 +1,66 @@
+{
+  "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,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Blood/seed_50/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Blood-serialized/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": "4.55.2"
+}

Creditg/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Creditg-serialized/added_tokens.json

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

Creditg/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Creditg-serialized/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

Creditg/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Creditg-serialized/config.json

@@ -0,0 +1,66 @@
+{
+  "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,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Creditg/seed_0/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Creditg-serialized/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": "4.55.2"
+}

Creditg/seed_4/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Creditg-serialized/runs/Dec26_23-58-50_csce-yang-s2.engr.tamu.edu/events.out.tfevents.1766815143.csce-yang-s2.engr.tamu.edu.3648358.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63c500daef15e1649376263aaec8c28e14ea7b9b2668053e9a12f3242015fbb8
+size 206473

Creditg/seed_50/SFT/LLM-Qwen-2.5-3B-SFT-decision-tree-Creditg-serialized/runs/Dec27_07-28-09_csce-yang-s2.engr.tamu.edu/events.out.tfevents.1766842103.csce-yang-s2.engr.tamu.edu.3873179.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5514891fbf86d3a176725ac2df323f71ed0600e70d35faee1355408455d0c05
+size 186280