Lex Fridman Podcast
#459 – DeepSeek, China, OpenAI, NVIDIA, xAI, TSMC, Stargate, and AI Megaclusters
Dylan Patel
Now it is the case that there are many model options that are way better than humans at writing detailed and eloquent answers for things like model and code. So They talked about this with the Lama 3 release, where they switched to using Lama 3, 4, or 5B to write their answers for math and code.
0
💬
0
Comments
Log in to comment.
There are no comments yet.