fine-tuning-with-trl

【效率与协作】Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准