logoalt Hacker News

dcrazy • yesterday at 11:30 PM • 1 reply • view on HN

Yeah, but as you might expect they internally represent numbers probabilistically, so there’s always a nonzero possibility of confusing the inputs or outputs of any operation. Kind of like misremembering your multiplication tables.


Replies

einichi • today at 12:20 AM

I don't think anybody is arguing that LLMs do math better than a traditional processor

➕ show 1 reply