The LLMs are just somewhere between an averaging and a lossy compression of everything on GitHub. There’s nothing about the current paradigm of “AI” that is going to somehow do better than just rehashing that training set but with the inclusion of various classes of errors.
I think it’s better to view it as spicy search rather than any form of intelligence.
The LLMs are just somewhere between an averaging and a lossy compression of everything on GitHub. There’s nothing about the current paradigm of “AI” that is going to somehow do better than just rehashing that training set but with the inclusion of various classes of errors.
I think it’s better to view it as spicy search rather than any form of intelligence.
If you treat it as spicy search it works pretty great though.