6 ms·
Maybe get yourself checked for chatbot psychosis. I am using current models productively, every day, and have 0 (and I mean precisely, literally 0) issue with c
by negura 18d ago
Maybe get yourself checked for chatbot psychosis. I am using current models productively, every day, and have 0 (and I mean precisely, literally 0) issue with calling it a stochastic parrot, one which lacks any kind of mentality whatsoever. There is not a shred of doubt in my mind that this is purely a statistical model, generating sequences of words, that happen to make sense in our actual mentality.
- sneak 16d agoWhat about my proposed psychosis would make the tests go green when it implements something complex in one shot? In any case, thanks for validating my previous assumption that people like this still exist even in the age of Opus/Fable.
- negura 16d ago> What about my proposed psychosis would make the tests go green when it implements something complex in one shot? Nothing. That's why it's a psychosis. It has nothing to do with the reality of how language models generate words without mental reasoning.
- sneak 15d agoThe “something complex” is critical here - I mean specifically an implementation that requires an advanced and reasoned mental model to write. How does it oneshot those and pass comprehensive tests if it isn’t doing advanced reasoning?
- negura 15d agoBecause it's a statistical model of the words used in similar problem domains.