Llama3.1-8B-PRM-Mistral-Data

This is a process-supervised reward (PRM) trained on Mistral-generated data from the project RLHFlow/RLHF-Reward-Modeling

The model is trained from meta-llama/Llama-3.1-8B-Instruct on RLHFlow/Mistral-PRM-Data for 1 epochs. We use a global batch size of 32 and a learning rate of 2e-6, where we pack the samples and split them into chunks of 8192 token. See more training details at https://github.com/RLHFlow/Online-RLHF/blob/main/math/llama-3.1-prm.yaml .

BoN evaluation result for Mistral generator:

Scaling the inference sampling to N=1024 for Deepseek generator:

Visualization

Usage

See https://github.com/RLHFlow/RLHF-Reward-Modeling/tree/main/math-rm for detailed examples.

Citation

The automatic annotation was proposed in the Math-shepherd paper:

@inproceedings{wang2024math,
  title={Math-shepherd: Verify and reinforce llms step-by-step without human annotations},
  author={Wang, Peiyi and Li, Lei and Shao, Zhihong and Xu, Runxin and Dai, Damai and Li, Yifei and Chen, Deli and Wu, Yu and Sui, Zhifang},
  booktitle={Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  pages={9426--9439},
  year={2024}
}

If you find the training recipe useful, please consider cite it as follows.

@misc{xiong2024rlhflowmath,
      author={Wei Xiong and Hanning Zhang and Nan Jiang and Tong Zhang},
  title = {An Implementation of Generative PRM},
  year = {2024},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/RLHFlow/RLHF-Reward-Modeling}}
}