News for hackers. Stuff that matters.

A comment on Cargo Cult AI. Parent comment

Re:Cargo Cult AI by hospitalhusband on Thursday May 18, 2023 @04:38PM (#35993722)
> In what sense does an LLM think one word at a time that doesn't also apply to a person typing at a keyboard?

Because you already have the thought formed before you started typing.

> When I read my brain presumably ingests one word at a time (not sure if it's one exactly, but it can't be much more than one)

And these models ingest many words at a time, up to the context length.

Seems to me it's fundamentally inverted from how we operate,both input and output.

[ Parent | on HN ]
Re:Cargo Cult AI by sebzim4500 on Thursday May 18, 2023 @04:57PM (#35993949)
>Because you already have the thought formed before you started typing.

Can you prove that GPT-4 doesn't? Clearly there is a sense in which thinks more than one word ahead, since as I mentioned above it would not otherwise be able to use 'a' vs 'an' correctly.

As far as I am aware, exactly to what extent these models have determined what tokens will be generated before they produce anything is an open question in mechanistic interpratability research. I would be very interested if you knew of some work that answers this question empirically.

[ Parent | on HN ]