[stevenhuang]

> Language models don’t—if you run the same prompt through a LLM several times you’ll get a slightly different reply every time.

You can get deterministic output (on a given machine) by setting temperature=0. The Chatgpt interface doesn't let you do that, but the playground API does.

[xigency]

This is an interesting argument. It's evident that the LLM /can/ give different answers to the same question and that they are stochastic in nature. But being computer programs we can make them deterministic with respect to input as you suggest.

More to the point, I don't think a "calculator for words" should be deterministic. Operating on language is much more subjective than operating on numbers. If anything, this is a human limitation that we expect only one answer to one question. I'm a contrarian to Chomsky's philosophy, as he's always been pessimistic of statistical language processing and often approaches from the more objective-side like grammar and parsing.

I'm waiting for the point where we can tap knowledge from Deep Learning models to build rule-sets that appease the deterministic crowd (and get the insight of what an LLM is really modeling). A breakthrough here could also help with two big problems a) alignment and b) copyright.

[spencerchubb]

Also if you specify the seed you get deterministic output, assuming that the interface allows to specify a seed.