News for hackers. Stuff that matters.

A comment on XGen-7B, a new 7B foundational model trained on up to 8K length for 1.5T tokens. Parent comment

Re:XGen-7B, a new 7B foundational model trained on up to 8K ... by seeknotfind on Wednesday June 28, 2023 @09:34PM (#36515229)
Why not ask the AIs themselves? They are pretty good at explaining this type of thing.
[ Parent | on HN ]
Re:XGen-7B, a new 7B foundational model trained on up to 8K ... by dcre on Wednesday June 28, 2023 @09:49PM (#36515338)
Sounds like a joke but it's true.
[ Parent | on HN ]
Re:XGen-7B, a new 7B foundational model trained on up to 8K ... by deedree on Wednesday June 28, 2023 @10:15PM (#36515532)
But if you don't know a certain amount about a subject already you won't know when it's lying to you. That would probably be the case here.
[ Parent | on HN ]
Re:XGen-7B, a new 7B foundational model trained on up to 8K ... by CamperBob2 on Thursday June 29, 2023 @12:26AM (#36516277)
It's basically what an interactive dialogue with Wikipedia would look like, which is still a darned useful thing.
[ Parent | on HN ]
Re:XGen-7B, a new 7B foundational model trained on up to 8K ... by TeMPOraL on Thursday June 29, 2023 @02:59AM (#36517284)
If Wikipedia had the conviction and focus of a 4 year old. But yeah, while tricky, it's still usually net positive.
[ Parent | on HN ]