#367 – Sam Altman: OpenAI CEO on GPT-4, ChatGPT, and the Future of AI
Lex Fridman Podcast
How RLHF makes ChatGPT better
Chat GPT is a term for a model that is trained on text data and can do amazing things, but is not very useful or easy to use./nRLHF is how humans help to make the model more useful and easier to use./nLess data is required for RLHF to work well than for the original model, and the process of incorporating human feedback is very interesting.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
