I honestly wasn’t trying to attack you. I think we should be careful when we talk about LLMs, because it’s important for people to know that it’s just a bunch of math in a computer program. A lot of people have a tendency to anthropomorphize it.

You said “that’s not how they work”. But that is how they work. Same prompt = same output. Throw some random data in there to jumble things around and you get a little variance. That’s the seed, and we only need to do that because LLMs are inherently deterministic.

Same reason Minecraft has a random seed for world generation, and block cipher algorithms use an initialization vector and/or feedback loop. We don’t want the same thing every time.

I did say that you’re right, because the tooling we use around the LLM itself does exactly what you’re talking about. So, in practice, you’re right.

People confuse LLMs for being non-deterministic all the time. The comment implied that. I want people to understand that LLMs are deterministic, because I don’t want people confusing them with actual intelligence or some sort of “black magic”. But sure man, whatever excuse you need to be an asshole to a stranger.

Please read the last sentence of my comment. I am not saying that the interpretation is wrong, I’m saying the statement that that’s not how LLMs work is wrong. That is how LLMs work. They are deterministic. The only reason they don’t do that in practice is because we purposefully seed them with random data to make them not do that.

Fair enough. Still, why would you misrepresent my view so radically unless you were being uncharitable on purpose?

It’s called pedantry, and I have never failed to own it. At this point, I feel like you’re trying to be overly abrasive.

Was it not your interpretation that the messages seem to not be from LLMs, because they’re identical? Because that’s literally what you said.

How did you possibly get that from what I said? Are you purposefully being as uncharitable as possible?

No, I clearly was not talking about this situation. I was clarifying how your interpretation was correct, but you were factually incorrect.

LLMs won’t spit out the same word-for-word response for the same prompt

You can give an LLM the same seed and it will spit out the same word-for-word response. That’s how they work. It’s just a bunch of math.

You are right, but you are also wrong. If they’re given the same seed, they certainly will. They are 100% deterministic. But in reality, the seed is randomly generated, so yeah, it won’t be exactly the same every time.