Reinforcement Learning from Human Feedback, Explained Simply | Towards Data Science

The one technique that made ChatGPT so smart