AILLM训练流水线

对齐 (Alignment)

订阅后可继续阅读剩余内容立即订阅

预训练 (Pretraining)

Previous Page

提示工程 (Prompt Engineering)

Next Page

On this page

1. SFT (Supervised Fine-Tuning, 监督微调)2. 偏好对齐 (Preference Alignment)RLHF (Reinforcement Learning from Human Feedback)DPO (Direct Preference Optimization)对齐税 (Alignment Tax)