logoalt Hacker News

conorberginyesterday at 10:49 PM4 repliesview on HN

LLMs are deterministic, the same model under the same conditions will produce the same output, unless some randomness is purposefully injected. Neural networks in general can be thought of as universal function approximators.


Replies

mrobtoday at 12:08 AM

Whenever somebody calls LLMs "non-deterministic", assume they meant "chaotic", in the informal sense of being a system where small changes of input can cause large changes to output, and the only way to find out if it will happen is by running the full calculation.

For many applications, this is equally troublesome as true non-determinism.

show 1 reply
2ndorderthoughtyesterday at 11:46 PM

That's not really true. If you turn a few knobs you can make them deterministic. Namely setting temperature to zero, and turning off all history. But none of the cloud providers do this. Because it's not a product as far as they are concerned. So in practice - not so much.

show 2 replies
alansaberyesterday at 11:45 PM

Yes theres a good thinking machines lab blog about this

0-_-0yesterday at 11:38 PM

You're being downvoted, but you're right. Determinism is a different concept and doesn't characterise LLMs well. You can have deterministic random number generators for example.