Are the results of the matmuls really that far apart in size that you have to lose significant bits when adding them up at FP32?
Defeating Nondeterminism in LLM Inference
131–137 of 137 posts
Re: Defeating Nondeterminism in LLM Inference
#132Earlier quoted context omitted.
I used exponential golomb coding, so the rank 0 logit is encoded with a single bit, ranks 1 and 2 are encoded with three bits, ranks 3-6 are encoded with 5 bits, etc. In terms of performance, I've not done any serious testing, but e.g. the wikipedia article on volcanos compresses to about 20% using GPT2. I've seen other strings compress even further. The big issue is that while encoding is not unreasonable, decoding…
I don't know about golomb coding, but with Arithmetic coding you can do stream decoding(AC), if I remember correctly. I supervised a student's project whose goal was exactly that : implement compression with LLMs using AC. Since AC is optimal, if your LLM has an average cross entropy x on some dataset, you can expect that the compression will compress data using x nats per token on average!
Arithmetic coding looks like an extremely interesting approach, given that you can use the model at each step to give you the probabilities of each token.
Re: Defeating Nondeterminism in LLM Inference
#133Fixing "theoretical" nondeterminism for a totally closed individual input-output pair doesn't solve the two "practical" nondeterminism problems, where the exact same input gives different results given different preceding context, and where a slightly transformed input doesn't give a correctly transformed result. Until those are addressed, closed-system nondeterminism doesn't really help except in cases where a looku…
There is no such thing as "exactly the same input, but with different preceding context". The preceding context is input! If you were to obtain exactly the same output for a given input prompt, regardless of context, then that would mean that the context is being ignored, which is indistinguishable from the session not maintaining any context such that each prompt is in a brand new empty context. Now what some people…
Sure. But the context always starts with the first input, right?
And how can you guarantee—or why should you guarantee—that the reply to the first input will always be the same?
And if that’s not the case, how can we ensure the preceding context remains consistent?
Re: Defeating Nondeterminism in LLM Inference
#134Earlier quoted context omitted.
Ambiguity : The request/prompt leaves a lot of room for interpretation. Many qualitatively different answers may be correct, relative to the prompt. Different or non-deterministic models will return highly variance results. Determinism : If a model is given the exact same request/prompt twice, its two responses will also be identical. Whether or not the consistent response qualifies as correct. The two concepts are v…
Maybe I got this wrong but I thought ambiguity refered to the input. So in a deterministic system I would assume that a input of "Give an example of a dice roll" Will always output the exact same example (unless the model also gets the context of the message history). Ambiguity is what happens when you change the prompt slightly, e.g. by adding a word: "Give an example of a single dice roll". Now as a human our expec…
> Ambiguity: [...] Different or non-deterministic models will return highly variance results.
Yes, and thanks. That was my intended point - but you point out a better example. Slightly different prompts may also produce highly varied responses.
(My subsequent comments on ambiguous models was in case I was misinterpreting the comment I was replying to. I also generally think of ambiguity as a property of input. Either way, ambiguity is not the same as non-deterministic.)
Re: Defeating Nondeterminism in LLM Inference
#135Deterministic reproducibility is very different from replicability, and imo the latter is more important; even if the details of the reproducibility are interesting I think they're irrelevant.
There's a similar situation in other scientific disciplines. People want source code and data so they can reproduce results - that basically tells you someone didn't cheat and they documented everything. But it does not tell you whether a real phenomenon was observed.
It's much more interesting to know if roughly the same cause and effect relationships exist so we can predict behavior.
Concretely, there are studies that show e.g. randomly capitalizing letters can lead to completely different responses from and LLM. That speaks to a fragility that doesn't have anything to do with deterministic reproduction.
Re: Defeating Nondeterminism in LLM Inference
#136Earlier quoted context omitted.
Then we need something else. This is not how LLMs work. They are simple statistical predictors, now universal anwsering machines.
> They are simple statistical predictors, now universal anwsering machines. I see this a lot. I kinda' doubt the "simple" part, but even beyond that, is there any evidence that statistical predictor can't be a universal answering machine? I think there's plenty of evidence that our thinking is at least partially a statistical predictor (e.g. when you see a black sheep you don't think "at least one side of this sheep…
Does your definition of "universal answering machine" include the answers being correct?
If it does, statistical predictors can't help you because they're not always correct or even meaningful (correlation does not imply causation).
If it doesn't then, by all means, enjoy your infinite monkeys
Re: Defeating Nondeterminism in LLM Inference
#137[dead]