Vibeleaderboard
← All Intel
Intel / article

Training language models to follow instructions with human feedback

Source
arxiv.org
Author
Long Ouyang et al.
Date

Context

OpenAI's InstructGPT research showed how training for preferred behavior could outweigh model size in human evaluations. The researchers first trained GPT-3 on human-written examples of suitable responses. They then collected rankings of model outputs and used them to further the model with , as the 2022 paper describes.

In that paper, evaluators preferred the 1.3-billion-parameter InstructGPT model to the 175-billion-parameter GPT-3 on the study's prompt distribution. That measures which responses people preferred for those requests, rather than establishing superior capability across every task.

Terms in this piece · Glossary
  • fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.
  • RLHF — Reinforcement learning from human feedback — training a model to prefer answers humans rate as better, which turns a raw text predictor into a usable assistant.
  • LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
Recommended reads
Comments

Checking sign-in…

Loading comments…