The chess engine's training is anchored by the win/lose outcome of the game. LLM training is anchored by what humans like to read and write. This means that a human needs to somehow be in the loop.
itsybitesyspider
joined 1 year ago
The chess engine's training is anchored by the win/lose outcome of the game. LLM training is anchored by what humans like to read and write. This means that a human needs to somehow be in the loop.
I like library providers that can provide mechanical upgrade instructions. For example:
model.adjust(x,1,y)is nowmodel.single(Adjustment.Foo, x).with_attribute(y)Or whatever. Then people can go through your instructions find-and-replacing the changes, or even better, have an automated tool do it.
Also you pay some of the maintenance burden by writing all this documentation, so you have a some stake in keeping the changes minimal.