News for hackers. Stuff that matters.
Re:Fine-Tuning LLMs to 1.58bit by patleeman on Wednesday September 18, 2024 @01:51PM (#41583188)
That's awesome. The original discussion of bitnet made it seem like you needed to train a model from scratch but its neat they were able to adapt an existing model. This is quite exciting.
Re:Fine-Tuning LLMs to 1.58bit by cpldcpu on Thursday September 19, 2024 @10:20PM (#41598214)
the performance is still a bit degraded though.