machine-learningdata-ai
fine-tuning-with-trl
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.
maintainer
math-inc
更新於 3/19/2026
星標
1165
分支
96
quick start
Installation and usage
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.
安裝
$ install --globalskills.sh
使用
安裝後，您可以通過在終端運行以下命令來使用此技能：
skills use fine-tuning-with-trl