OpenAI describes InstructGPT: the model is fine-tuned on human ratings (RLHF), and people prefer the answers of its 1.3-billion-parameter version to those of the 175-billion-parameter GPT-3
translated by AI, reviewed by the editors
· Science
- technology
- Learning from human feedback (RLHF)
- source
- arxiv.org ↗
¶ c585bcdf6