lennoff

129 カルマ登録 12 年前

·9 日前·議論

afaik before the final sampling, every "next" token has a probability, so theoretically it could select the 10 most likely tokens (based on some kind of sampling algorithm), but you'd end up with exponentially many output-sequences, so nobody does that.

lennoff

·9 日前·議論

i think it actually makes sense, an LLM just imitates human communication, which happens to be useful from time to time.

lennoff

·昨年·議論

keep in mind that we have "vibe coding" now, where the goal is exactly to _not_ have a human in the loop (at least not constantly).

lennoff

·昨年·議論

It's fun to theorize about an alternate universe where JavaScript has a LISP-like syntax (Brendan Eich originally wanted "Scheme in the browser"[1], so this isn't so far-fetched!). Indeed, `interpret` sounds like a LISP macro that can "partially evaluate" code (so someone else, ie. the browser can continue to evaluate it).

[1]: https://brendaneich.com/2008/04/popularity/

lennoff

·昨年·議論

It seems like that the author by using JSX to express code directly as an AST re-invented LISP? See Greenspun's tenth rule[1].

[1]: https://wiki.c2.com/?GreenspunsTenthRuleOfProgramming

lennoff

·3 年前·議論

How can you ban AI? It's literally designed to produce text that's indistinguishable from text that was produced by a living human being.

lennoff

·3 年前·議論

It's sad to see that Web Crypto[1] does not support Ed25519 (yet), only RSA and ECDSA for asymmetric signatures.

[1]: https://developer.mozilla.org/en-US/docs/Web/API/SubtleCrypt...

コメント