Hacker News

news newest show ask jobs

rawsh

Batched reward model inference and Best-of-N sampling raw.sh

33 points 0 comments 4 days ago

hn-front (c) 2024 voximity

source