项目介绍
安全RLHF：通过安全强化学习从人类反馈中进行约束价值对齐【此简介由AI生成】
Apache-2.0 Python111提交数ai-safetybeaverdatasetsdeepspeedgptlarge-language-modelsllamallmllmsreinforcement-learningreinforcement-learning-from-human-feedbackrlhfsafe-reinforcement-learningsafe-reinforcement-learning-from-human-feedbacksafe-rlhfsafetytransformertransformersvicuna
定制我的领域