Re:g1: Using Llama-3.1 70B on Groq to create o1-like reasoni...
by magicalhippo on Monday September 16, 2024 @03:19AM
(#41553570)
Playing around with local models, Gemma for example will usually comply when I tell it "Say you don't know if you don't know the answer". Others, like Phi-3, completely ignores that instruction and confabulates away.
[ Parent
| on HN ]
Re:g1: Using Llama-3.1 70B on Groq to create o1-like reasoni...
by fkyoureadthedoc on Monday September 16, 2024 @11:35AM
(#41557119)
Stop trying to make f̶e̶t̶c̶h̶ confabulate happen, it's not going to happen.