News for hackers. Stuff that matters.

A comment on OPT: Open Pre-trained Transformer Language Models. Parent comment

Re:OPT: Open Pre-trained Transformer Language Models by lumost on Tuesday May 03, 2022 @01:13AM (#31244904)
A 175 billion parameter model might be a couple hundred gigs on disk. The file is probably just too big for GitHub/other standard FB services.
[ Parent | on HN ]
Re:OPT: Open Pre-trained Transformer Language Models by cma on Tuesday May 03, 2022 @03:44AM (#31245755)
They could just torrent.
[ Parent | on HN ]
Re:OPT: Open Pre-trained Transformer Language Models by lumost on Wednesday May 04, 2022 @08:54AM (#31259736)
They could, and their might be a torrent in the future - but torrents lose tracking info. I'm sure the researchers want to know who is downloading their models even if they don't care who it is.
[ Parent | on HN ]