Jinse Finance reported that OpenAI has trained a model based on GPT-4, called CriticGPT, to capture errors in the output of ChatGPT code. We are starting to integrate models like CriticGPT into our RLHF to provide clear AI assistance to OpenAI trainers. We plan to further expand the application of RLHF on GPT-4 and put it into practice.
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
OpenAI has released CriticGPT, a model for capturing code errors in ChatGPT.
Jinse Finance reported that OpenAI has trained a model based on GPT-4, called CriticGPT, to capture errors in the output of ChatGPT code. We are starting to integrate models like CriticGPT into our RLHF to provide clear AI assistance to OpenAI trainers. We plan to further expand the application of RLHF on GPT-4 and put it into practice.