初始化项目，由ModelHub XC社区提供模型

Model: ScienceOne-AI/S1-DeepResearch-8B-Preview Source: Original Platform
2026-09-07 16:28:13 +08:00
commit b32852000c
17 changed files with 1093 additions and 0 deletions
--- a/.gitattributes
+++ b/.gitattributes
@@ -0,0 +1,55 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bin.* filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+ 
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zstandard filter=lfs diff=lfs merge=lfs -text
+*.tfevents* filter=lfs diff=lfs merge=lfs -text
+*.db* filter=lfs diff=lfs merge=lfs -text
+*.ark* filter=lfs diff=lfs merge=lfs -text
+**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text
+**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text
+**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text
+ 
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.gguf* filter=lfs diff=lfs merge=lfs -text
+*.ggml filter=lfs diff=lfs merge=lfs -text
+*.llamafile* filter=lfs diff=lfs merge=lfs -text
+*.pt2 filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+
+model-0003-others-save_rank0.safetensors filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+model-0004-others-save_rank0.safetensors filter=lfs diff=lfs merge=lfs -text
+vocab.json filter=lfs diff=lfs merge=lfs -text
+model-0002-others-save_rank0.safetensors filter=lfs diff=lfs merge=lfs -text
+merges.txt filter=lfs diff=lfs merge=lfs -text
+model-0001-others-save_rank0.safetensors filter=lfs diff=lfs merge=lfs -text
--- a/README.md
+++ b/README.md
@@ -0,0 +1,147 @@
+---
+language:
+- zh
+- en
+license: apache-2.0
+tags:
+- agent
+- deep-research
+- tool-calling
+- long-context
+- reasoning
+pipeline_tag: text-generation
+---
+
+# S1-DeepResearch-8B-Preview
+
+<div align="center">
+
+[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://www.apache.org/licenses/LICENSE-2.0)
+[![GitHub](https://img.shields.io/badge/GitHub-Repository-black?logo=github)](https://github.com/ScienceOne-AI/S1-DeepResearch)
+
+**面向长程深度研究任务的高效能智能体 | High-Performance Deep Research Agent**
+
+[🤗 Model](https://huggingface.co/ScienceOne-AI/S1-DeepResearch-8B-Preview) | [🤖 ModelScope](https://modelscope.cn/models/ScienceOne-AI/S1-DeepResearch-8B-Preview) | [📖 GitHub](https://github.com/ScienceOne-AI/S1-DeepResearch) | [📄 Technical Report (Coming Soon)]()
+
+</div>
+
+---
+
+## 📝 模型简介 | Model Overview
+
+**S1-DeepResearch-8B-Preview** 是由 ScienceOne 团队开发的专门为 **长程深度研究(Deep Research)** 任务设计的开源智能体模型。它可以处理非常复杂的、需要多步推理的信息搜寻任务，并具备强大的连续多轮工具调用能力,能够像人类研究员一样,在海量信息中抽丝剥茧,进行深度的信息检索、阅读、理解与整合。
+
+**S1-DeepResearch is an open-source agent model specifically developed for long-horizon deep research tasks.** it can handle highly complex, multi-step reasoning information retrieval tasks with powerful continuous multi-turn tool calling capabilities, working like a human researcher to perform deep information retrieval, reading, comprehension, and integration.
+
+---
+
+## ✨ 核心特性 | Key Features
+
+- **⚡ 小而不凡 | Small Yet Powerful**
+  - 仅 **8B** 参数量级,保持高推理速度与低部署成本
+  - 在 Agent 任务中表现出优异的主动规划、信息整合能力
+  - Maintains high inference speed and low deployment cost at 8B parameter scale
+  - Exhibits excellent proactive planning and information integration capabilities
+
+- **🏆 性能卓越 | Outstanding Performance**
+  - 在 GAIA、DeepSearch、Browsecomp 等主流深度研究类基准上表现突出
+  - 达到同尺寸模型的 SOTA 水平,并超越一些更大参数量的模型
+  - Achieves SOTA level among same-sized models on mainstream benchmarks
+  - Surpasses some larger parameter models
+
+- **🔄 长链执行 | Long-Chain Execution**
+  - 支持 **128k** 上下文窗口
+  - 能够稳定执行 **100+ 轮**的连续工具调用
+  - 在长链路深度研究推理中保持高度的推理韧性
+  - 128k context length
+  - Capable of stably executing 100+ rounds of continuous tool calls
+
+- **⚙️ 高质量合成数据 | High-Quality Synthetic Data**
+  - 创新的全自动数据合成流水线
+  - 采用搜索-浏览-模糊的合成流程
+  - 知识图谱随机游走策略,自动生成高复杂度、强搜索依赖的问答对数据
+  - Innovative fully automated data synthesis pipeline
+
+---
+
+## 📊 性能评估 | Evaluation
+
+我们在多个权威智能体能力基准上对 S1-DeepResearch 进行了评估。结果表明，S1-DeepResearch 在同参数规模模型中达到领先水平，并在多项任务上凭借高效的推理与规划机制，超越了部分参数规模更大的开源模型。进一步地，在与闭源顶尖模型及专业 Deep Research 系统的对比中，S1-DeepResearch 依然展现出稳定且具竞争力的性能表现。整体结果表明，在合理的智能体架构与推理策略支持下，轻量化但具备高能力密度的模型同样能够有效应对复杂深度研究任务，为深度研究智能体的实际应用提供了一条可行路径。
+
+We evaluate **S1-DeepResearch** across multiple authoritative agent benchmarks. The results show that S1-DeepResearch achieves state-of-the-art performance among models of comparable size, and in several tasks surpasses larger open-source models due to its efficient reasoning and planning mechanisms. Moreover, when compared with leading closed-source models and specialized Deep Research systems, S1-DeepResearch demonstrates consistently competitive performance. These results indicate that **lightweight models with high capability density**, when equipped with well-designed agent architectures and reasoning strategies, can effectively handle complex deep research tasks, offering a practical and scalable solution for real-world research-oriented agent systems.
+
+<div align="center">
+  <img src="https://github.com/ScienceOne-AI/S1-DeepResearch/raw/main/assets/benchmark.png" width="80%" alt="Benchmark Results">
+</div>
+
+### 推理时扩展 | Test-Time Scaling
+
+我们进一步分析了模型在推理时扩展下的性能表现。评测结果显示，S1-DeepResearch 在 Pass@1 到 Pass@3 设置下取得了稳定且显著的性能提升，表明其推理与规划过程在单次推理条件下尚未饱和。该结果说明，通过推理时扩展，模型能够探索更丰富的推理与规划路径，从而有效提升复杂任务的整体成功率。
+
+We further analyze the model’s test-time scaling behavior. S1-DeepResearch shows consistent and significant performance gains from Pass@1 to Pass@3, indicating that its reasoning and planning processes are not saturated under single-sample test. This suggests that test-time expansion enables the model to explore diverse reasoning and planning trajectories, thereby substantially improving task success rates.
+
+<div align="center">
+  <img src="https://github.com/ScienceOne-AI/S1-DeepResearch/raw/main/assets/pass1to3.png" width="80%" alt="Pass@1 to Pass@3 Results">
+</div>
+
+---
+
+
+
+## 🛠️ 使用方式 | Usage
+
+详细的使用说明和示例代码请参考我们的 GitHub 仓库
+
+**For detailed usage instructions and example code, please refer to our GitHub repository:**
+
+👉 **[https://github.com/ScienceOne-AI/S1-DeepResearch](https://github.com/ScienceOne-AI/S1-DeepResearch)**
+
+
+
+---
+
+
+---
+
+## 📜 协议与引用 | License & Citation
+
+### License
+
+本模型采用 **Apache License 2.0** 开源协议。
+
+This model is licensed under **Apache License 2.0**.
+
+### 📑 Citation
+
+
+```bibtex
+@software{s1agent2025,
+    title={S1-DeepResearch: High-Performance Deep Research Agent},
+    author={ScienceOne Team},
+    year={2025},
+    url={https://github.com/ScienceOne-AI/S1-DeepResearch},
+}
+```
+
+---
+
+## 🤝 联系我们 | Contact
+
+- **GitHub Issues**: [https://github.com/ScienceOne-AI/S1-DeepResearch/issues](https://github.com/ScienceOne-AI/S1-DeepResearch/issues)
+
+---
+
+## ⚠️ 免责声明 | Disclaimer
+
+本模型为研究预览版本,可能存在局限性。在生产环境使用前请充分测试。模型输出结果仅供参考,不构成专业建议。
+
+**This model is a research preview and may have limitations. Please test thoroughly before production use. Model outputs are for reference only and do not constitute professional advice.**
+
+---
+
+<div align="center">
+
+**Made with ❤️ by ScienceOne Team**
+
+</div>
+
--- a/added_tokens.json
+++ b/added_tokens.json
@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}
--- a/chat_template.jinja
+++ b/chat_template.jinja
@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# 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' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first 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' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}
--- a/config.json
+++ b/config.json
@@ -0,0 +1,69 @@
+{
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 12288,
+  "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": 131072,
+  "max_window_layers": 36,
+  "model_type": "qwen3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 8,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}
--- a/configuration.json
+++ b/configuration.json
@@ -0,0 +1 @@
+{"framework":"Pytorch","task":"other"}
--- a/generation_config.json
+++ b/generation_config.json
@@ -0,0 +1,13 @@
+{
+    "bos_token_id": 151643,
+    "do_sample": true,
+    "eos_token_id": [
+        151645,
+        151643
+    ],
+    "pad_token_id": 151643,
+    "temperature": 0.6,
+    "top_k": 20,
+    "top_p": 0.95,
+    "transformers_version": "4.51.0"
+}
--- a/merges.txt
+++ b/merges.txt
--- a/model-0001-others-save_rank0.safetensors
+++ b/model-0001-others-save_rank0.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5871eea0a6ae87cc5891393285b17a3fbfac5893b58e11b8122153a1fb78751
+size 4231141088
--- a/model-0002-others-save_rank0.safetensors
+++ b/model-0002-others-save_rank0.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d83edc02ad74befe5274959197d37464133b15838a07a919e444f92e4e7e663
+size 4387458752
--- a/model-0003-others-save_rank0.safetensors
+++ b/model-0003-others-save_rank0.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8566c0e7731ca347228a02df03774f9169eb1c3cb226ba718568fd3c546ec4b
+size 4286779488
--- a/model-0004-others-save_rank0.safetensors
+++ b/model-0004-others-save_rank0.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5450a1b9edd1be2b9278bca8692426f928856d81c6a36764e4d59ff62f6f9ea
+size 3476137368
--- a/model.safetensors.index.json
+++ b/model.safetensors.index.json
@@ -0,0 +1,404 @@
+{
+  "weight_map": {
+    "model.norm.weight": "model-0001-others-save_rank0.safetensors",
+    "lm_head.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.self_attn.q_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.self_attn.k_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.self_attn.q_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.self_attn.k_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.self_attn.q_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.self_attn.k_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.self_attn.q_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.self_attn.k_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.self_attn.q_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.self_attn.k_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.self_attn.q_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.self_attn.k_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.self_attn.q_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.self_attn.k_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.7.self_attn.q_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.7.self_attn.k_norm.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-0001-others-save_rank0.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.self_attn.q_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.self_attn.k_norm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-0002-others-save_rank0.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.19.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.19.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.30.self_attn.q_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.30.self_attn.k_norm.weight": "model-0003-others-save_rank0.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.self_attn.q_norm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.self_attn.k_norm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.self_attn.q_norm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.self_attn.k_norm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.self_attn.q_norm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.self_attn.k_norm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.self_attn.q_norm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.self_attn.k_norm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.self_attn.q_norm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.self_attn.k_norm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-0004-others-save_rank0.safetensors",
+    "model.embed_tokens.weight": "model-0004-others-save_rank0.safetensors"
+  },
+  "metadata": {}
+}
--- a/special_tokens_map.json
+++ b/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
+  }
+}
--- a/tokenizer.json
+++ b/tokenizer.json
--- a/tokenizer_config.json
+++ b/tokenizer_config.json
@@ -0,0 +1,239 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}
--- a/vocab.json
+++ b/vocab.json