minus-squareDankmemexplorer@alien.topBtoLocalLLaMA•Starling-RM-7B-alpha: New RLAIF Finetuned 7b Model beats Openchat 3.5 and comes close to GPT-4linkfedilinkEnglisharrow-up1·3 years agothe model can have a little of the test data as a treat linkfedilink
minus-squareDankmemexplorer@alien.topBtoLocalLLaMA•RWKV v5 7b, Fully Open-Source, 60% trained, approaching Mistral 7b in abilities or surpassing it.linkfedilinkEnglisharrow-up1·3 years agoim very behind, was 14B not chinchilla optimal? linkfedilink
minus-squareDankmemexplorer@alien.topBtoLocalLLaMA•1.3B with 68.29% Humaneval lol, don't behead me. Part of my project PIC (partner-in-crime)linkfedilinkEnglisharrow-up1·3 years ago“the model can have the test data as a treat” linkfedilink
the model can have a little of the test data as a treat