News for hackers. Stuff that matters.
Re:LLMs are not greater than the sum of their parts: Stanfor... by root_axis on Tuesday May 09, 2023 @01:10AM (#35870396)
try wizard-vicuna-13b in ooba. I think you will be pleasantly surprised.
Re:LLMs are not greater than the sum of their parts: Stanfor... by ineedasername on Tuesday May 09, 2023 @11:00AM (#35875351)
Oh yeah, I have to check that out. I did a little with the 7B model (I think the description said Microsoft was involved in its creation? Which is interesting, shows they’re hedging their Bacha little bit away from open AI in towards true open source options as well) anyway the 7B Wizard 7B model was noticeably better that the other I tried, though also noticeably worse than the 13B “snoozy” model.

I’m trying to see if I can justify getting a new workstation-tower PC. My home computer and work laptop are beefy for my usual work, but not enough to

Re:LLMs are not greater than the sum of their parts: Stanfor... by ineedasername on Wednesday May 10, 2023 @04:11PM (#35892809)
hmmm, I'm having trouble finding a ggml model version... my google-fu force is not with me today. Do you know of a source?