logoalt Hacker News

p-e-w • today at 1:26 AM • 1 reply • view on HN

Because there is four billion years of evolutionary pre-training.


Replies

bavell • today at 1:43 AM

Instead of pre-training, I'd say it's more like a better architecture, training regime, higher fidelity data, and perhaps better initial "weights".

➕ show 1 reply