Don't Worry About the Vase Podcast cover image

On Dwarkesh Patel's Podcast With Andrej Karpathy

Don't Worry About the Vase Podcast

00:00

How Much Training Data Is Memorized?

Discussion on massive compression from training tokens into parameters and when full context improves precision.

Play episode from 08:29
Transcript

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app