Finding GPT-4’s mistakes with GPT-4
CriticGPT, a model based on GPT-4, writes critiques of ChatGPT responses to help human trainers spot mistakes during RLHF
Originally reported by OpenAI Blog
Read original
CriticGPT, a model based on GPT-4, writes critiques of ChatGPT responses to help human trainers spot mistakes during RLHF
