Hacker News Reader

My guess: There is a ton of data online of just plain weights laying around. No human would ever actually read it. LLM's read it all during training.

View on HN

But that would cover finite set of cases. I believe this result is more general.

I mean it would make sense if the training using that data resulted in a general set of equations/patterns, it's still really surprising though...

Reply on news.ycombinator.com