logoalt Hacker News

jimmoorestoday at 8:49 PM3 repliesview on HN

These people have zero idea what they're doing. Not a single mention of pipeline parallelism that would actually make the setup useful to run a big model.


Replies

itkovian_today at 9:07 PM

I can’t stand it. Very engineering-y over specified formal language around a complete lack of core understanding. Is damaging other people read this and try to learn things from it.

schaefertoday at 9:01 PM

can you point to a write up that discusses what you're talking about?

because I would read it.

kelmorantoday at 8:54 PM

I feel like you would want to run 8 smaller models separately for quantity of raw output. 1 big model is slow and isnt guaranteed to make no mistakes.

show 2 replies