PURE
Collection
PRM and fine-tuned LLM used in our PURE github repo: https://github.com/CJReinforce/PURE • 5 items • Updated • 2
🚨 This repo does not include the Process Reward Model (PRM). For access to the PRM, please refer to here.
This repository hosts a fine-tuned LLM optimized for better mathematical reasoning capabilities via only process rewards.
- Downloads last month
- 8