RPT DeepSeek R1 0528 Qwen3 8B

🚀 RPT-DeepSeek-R1-0528-Qwen3-8B模型卡片

本模型是 deepseek-ai/DeepSeek-R1-0528-Qwen3-8B 的微調版本。它使用 TRL 進行訓練。該模型採用GRPO方法進行訓練，此方法在論文 DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models 中被提出。GRPO訓練遵循RPT（強化預訓練論文）中介紹的方法：https://huggingface.co/papers/2506.08007 。

📚 詳細文檔

模型信息

框架版本

• TRL: 0.19.0
• Transformers: 4.52.4
• Pytorch: 2.6.0+cu124
• Datasets: 3.6.0
• Tokenizers: 0.21.2
• vllm: 0.8.5.post1

📄 許可證

本模型使用的許可證為 apache-2.0。

📚 引用信息

引用GRPO

@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},
}

引用TRL

@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}}
}