Finding GPT-4’s mistakes with GPT-4

T1Covered by 1 source · first reported by OpenAI News at 27 Jun 2024, 10:00 UTC

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

Sources

• T1Finding GPT-4’s mistakes with GPT-4OpenAI News27 Jun 2024, 10:00