Skip to content

Top New Best Ask Show Jobs

Top New Best Ask Show Jobs

MegaTrain Full Precision Training of 100B+ Parameter LLMs on a Single GPU

github.com/DLYuanGod

1 pointadulau3 months agodiscuss

Comments

No comments yet.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.