Training
RLHF (reinforcement learning from human feedback)
A training method where people rank model responses and the model is optimised to produce the kind of answers they prefer.
RLHF helped turn raw language models into useful assistants. Variants use AI feedback or written principles instead of, or alongside, human ratings.
Frequently asked questions
What does RLHF (reinforcement learning from human feedback) mean?
A training method where people rank model responses and the model is optimised to produce the kind of answers they prefer. RLHF helped turn raw language models into useful assistants. Variants use AI feedback or written principles instead of, or alongside, human ratings.