logoalt Hacker News

skerittoday at 8:17 PM3 repliesview on HN

So in theory you could feed it incomplete text, and then ask it for the probabilities of what the next character could be?


Replies

mckngbrdtoday at 8:54 PM

I think the joke here is getting missed

show 1 reply
CompleteSkeptictoday at 8:25 PM

you could, but it the model is not optimized for text

this is complex, but generating text is highly complicated and requires mode dropping to make long cohesive text

vatsachaktoday at 8:19 PM

If you provide it an AST of the english language, yes.