News for hackers. Stuff that matters.

A comment on Replit's new Code LLM: Open Source, 77% smaller than Codex, trained in 1 week. Parent comment

Re:Replit's new Code LLM: Open Source, 77% smaller than Code... by sanderjd on Wednesday May 03, 2023 @05:10PM (#35807851)
You seem to know your stuff some, so I'll ask you a question on this: Are there any good books on all the different approaches in this space, or is it all too new and fast moving for such a thing?
[ Parent | on HN ]
Re:Replit's new Code LLM: Open Source, 77% smaller than Code... by nl on Wednesday May 03, 2023 @09:48PM (#35810299)
There are no books on Large LMs but almost any resource about neural networks covers fine tuning. I like the FastAI courses, and these do cover language models.
[ Parent | on HN ]
Re:Replit's new Code LLM: Open Source, 77% smaller than Code... by osanseviero on Thursday May 04, 2023 @01:51AM (#35811671)
You can also check the NLP with transformers book
[ Parent | on HN ]