zerozeroz/CodeLlama-7b-hf_1000rl_cpp_javascript

1---2library_name: transformers3model_name: CodeLlama-7b-hf_1000rl_cpp_javascript4tags:5- generated_from_trainer6- trl7- grpo8licence: license9---10 11# Model Card for CodeLlama-7b-hf_1000rl_cpp_javascript12 13This model is a fine-tuned version of [None](https://huggingface.co/None).14It has been trained using [TRL](https://github.com/huggingface/trl).15 16## Quick start17 18```python19from transformers import pipeline20 21question = "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?"22generator = pipeline("text-generation", model="zerozeroz/CodeLlama-7b-hf_1000rl_cpp_javascript", device="cuda")23output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]24print(output["generated_text"])25```26 27## Training procedure28 29 30 31 32This 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).33 34### Framework versions35 36- TRL: 0.14.037- Transformers: 4.48.138- Pytorch: 2.5.1+cu12139- Datasets: 3.1.040- Tokenizers: 0.21.041 42## Citations43 44Cite GRPO as:45 46```bibtex47@article{zhihong2024deepseekmath,48    title        = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},49    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},50    year         = 2024,51    eprint       = {arXiv:2402.03300},52}53 54```55 56Cite TRL as:57    58```bibtex59@misc{vonwerra2022trl,60	title        = {{TRL: Transformer Reinforcement Learning}},61	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},62	year         = 2020,63	journal      = {GitHub repository},64	publisher    = {GitHub},65	howpublished = {\url{https://github.com/huggingface/trl}}66}67```