sydqiang/X-R1-0.5B

1---2datasets: xiaodongguaAIGC/X-R1-7503library_name: transformers4tags:5- generated_from_trainer6- X-R17licence: license8---9 10# Model Card for None11 12This model is a fine-tuned version of [None](https://huggingface.co/None) on the [xiaodongguaAIGC/X-R1-750](https://huggingface.co/datasets/xiaodongguaAIGC/X-R1-750) dataset.13It has been trained using [TRL](https://github.com/huggingface/trl).14 15## Quick start16 17```python18from transformers import pipeline19 20question = "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?"21generator = pipeline("text-generation", model="None", device="cuda")22output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]23print(output["generated_text"])24```25 26## Training procedure27 28[<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/sydqiang-tsinghua-university/huggingface/runs/3b3peizp) 29 30 31This 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).32 33### Framework versions34 35- TRL: 0.16.0.dev036- Transformers: 4.48.237- Pytorch: 2.5.138- Datasets: 5.0.039- Tokenizers: 0.21.440 41## Citations42 43Cite GRPO as:44 45```bibtex46@article{zhihong2024deepseekmath,47    title        = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},48    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},49    year         = 2024,50    eprint       = {arXiv:2402.03300},51}52 53```54 55Cite TRL as:56    57```bibtex58@misc{vonwerra2022trl,59	title        = {{TRL: Transformer Reinforcement Learning}},60	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 Gallouédec},61	year         = 2020,62	journal      = {GitHub repository},63	publisher    = {GitHub},64	howpublished = {\url{https://github.com/huggingface/trl}}65}66```