Jk, I’m on an instance that disables downvotes. Meaning folks have to write coherent comments for me to be aware they think I’m dumb/wrong/eeeeevil.
I could see tinkering with local models being a very satisfying hobby. I’ve always been curious about how accurate they can get on sentiment. E.g., a southern US woman saying “bless your heart” can have implied condescension while other regions it may be more of a straightforward good-faith comment; how and how well does a model get trained on that sort of thing.
Exactly… the difference in energy use is like driving a Ranger vs. an F-650. Sure, the big chungus models are a lot smarter, but it’s diminishing returns on energy use to capability. Cutting out the fluff and using a much, much smaller and lighter model that is tuned for the task you are using it for is a lot more efficient, and people with macbooks or gaming computers can run a model on their computer instead of subbing to a cloud service.
I’ll take my downvotes now 😞
Jk, I’m on an instance that disables downvotes. Meaning folks have to write coherent comments for me to be aware they think I’m dumb/wrong/eeeeevil.
I could see tinkering with local models being a very satisfying hobby. I’ve always been curious about how accurate they can get on sentiment. E.g., a southern US woman saying “bless your heart” can have implied condescension while other regions it may be more of a straightforward good-faith comment; how and how well does a model get trained on that sort of thing.
Exactly… the difference in energy use is like driving a Ranger vs. an F-650. Sure, the big chungus models are a lot smarter, but it’s diminishing returns on energy use to capability. Cutting out the fluff and using a much, much smaller and lighter model that is tuned for the task you are using it for is a lot more efficient, and people with macbooks or gaming computers can run a model on their computer instead of subbing to a cloud service.