LLMs hand people confident-sounding answers in domains they can't judge. The problem isn't the model being wrong, it's that the person can't tell.
Similar setup here with an M3 Max, it's surprisingly capable. Curious what models you're running on that M4 Pro.
LLMs hand people confident-sounding answers in domains they can't judge. The problem isn't the model being wrong, it's that the person can't tell.
Similar setup here with an M3 Max, it's surprisingly capable. Curious what models you're running on that M4 Pro.