We’ve trained a model called ChatGPT which interacts in a conversational way. The dialogue format makes it possible for ChatGPT to answer followup questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests.https://docs.google.com/spreadsheets/d/198a_3Ue8oM0f4WVRRpKzc1l7qucYF9PzZqDqz02IdsA/edit?gid=0#gid=0
