CarperAI/trlx

A repo for distributed training of language models with Reinforcement Learning via Human Feedback (RLHF)

拖动卡牌调整角度，点击翻面查看卡背。