TRL

Supervised Fine-tuning Trainer (SFT) 入門Supervised Fine-tuning (SFT) は、Reinforcement Learning from Human Feedback (RLHF) ...