Skip to main content
Artificial Intelligence

Finding GPT-4’s mistakes with GPT-4

OpenAI Blog · · 1 views

CriticGPT, a model based on GPT-4, writes critiques of ChatGPT responses to help human trainers spot mistakes during RLHF

CriticGPT, a model based on GPT-4, writes critiques of ChatGPT responses to help human trainers spot mistakes during RLHF