Lex Fridman Podcast
#452 – Dario Amodei: Anthropic CEO on Claude, AGI & the Future of AI & Humanity
Um, and so, um, it turns out this works really well and you can use it to sort of project, um, you know, if you train a sparse autoencoder at a given size, you know, how many tokens should you train on and so on.
0
💬
0
Comments
Log in to comment.
There are no comments yet.