3 ms·
The thing that blows me away is that I woke up one day and was confronted with a chat bot that could communicate in near perfect English. I dunno why exactly b
by Waterluvian 1y ago
The thing that blows me away is that I woke up one day and was confronted with a chat bot that could communicate in near perfect English.
I dunno why exactly but that’s what felt the most stunning about this whole era. It can screw up the number of fingers in an image or the details of a recipe or misidentify elements of an image, etc. but I’ve never seen it make a typo or use improper grammar or whatnot.
- sebmellen 1y agoIn a sense, LLMs emergently figured out the deep structure of language before we did, and that’s the most remarkable thing about them.
- hashmush 1y agoI dunno, it seems you have figured it out too, probably before LLMs? I'd say all speakers of all languages have figured it out and your statement is quite confusing, at least to me.
- Waterluvian 1y agoWe all make grammar mistakes but I’ve yet to see the main LLMs make any.
- sebmellen 1y agoYes of course we’ve implicitly learned those rules, but we have not been able to articulate them fully ala Chomsky. Somehow, LLMs have those rules stored within a finite set of weights. https://slator.com/how-large-language-models-prove-chomsky-wrong-with-steven-piantadosi/ https://slator.com/how-large-language-models-prove-chomsky-w...
- nikolayasdf123 1y agoalso, how quickly we moved from "it types nonsense" to "it can solve symbolic math, write code, test code, write programs, use bash, and tools, plan long-horizon actions, execute autonomously, ..."