geepytee

joined 1 year ago
sorted by: new top controversial old
Starling-RM-7B-alpha: New RLAIF Finetuned 7b Model beats Openchat 3.5 and comes close to GPT-4 in c/localllama@poweruser.forum
[–] geepytee@alien.top 1 points 11 months ago

It is pretty good, just not as good

Starling-RM-7B-alpha: New RLAIF Finetuned 7b Model beats Openchat 3.5 and comes close to GPT-4 in c/localllama@poweruser.forum
[–] geepytee@alien.top 1 points 11 months ago

They tested to MT Bench with GPT-4 as judge

I built Copilot replacement in c/localllama@poweruser.forum
[–] geepytee@alien.top 1 points 11 months ago

I built a new that has required preprocessing

Nice! How are you preprocessing the code?