No, a well-behaving LLM would do exactly what's seen. The most likely next toxen is "left" and it should deterministically output that unless some other layer like a temperature function makes it non-deterministic in its own way (wholly unrelated to the prompt).
The fantastical AGI precursor that people have been coached into seeing is what you're talking about, and that's (of course) not what an LLM actually is.
This is essentially just one of the easier ways you can expose the parlor trick behind that misconception.
No, that's not what I meant. Although it would still make more sense than what the author did. The problem lies in the way you actually determine probabilities. We know that humans are bad random number generators, but they understand the concept enough to come up with random looking stuff if you give them the chance. The LLMs here were not even given a chance. In essence, the author is complaining that the LLMs are not behaving according to frequentist statistics when he evaluates them in a strictly Bayesian setting.
Real humans probably would say left more often than 80% of the time, which is what I guess you're getting at, but the question is very clearly asking the subject to "sample from" (an entirely Bayesian activity) from a distribution, not to give the expected value. GPT4 gives the expected value and this is simply wrong.
Only at T=0. See my edit above how this changes everything.
I think its pretty simple and straightforward: the model simply fails to understand the question and can reasonably be said to not understand probability.
This is a good point. LLMs are bad at this, okay, but humans aren't great at it either.
On the other hand, does it really matter? With a slight tweak to the prompt, ChatGPT generates some serviceable code:
> Run a function to produce a random number between 1 and 10. What is the number?
import random
# Generate a random number between 1 and 10
random_number = random.randint(1, 10)
random_number
The random number generated between 1 and 10 is 9.Okay so are any GPU compilers intentionally introducing data races in programs that previously exhibited no data races?
Here’s a good jumping off point: https://pytorch.org/docs/stable/generated/torch.use_determin...
I think the author's point stands. They aren't asking "what would you expect from a distribution so described?" The answer to that question is 100% of the time "left.". A well behaving LLM responding to the actual question should distribute the logits across "left" and "right" in the way requested by the user and doesn't.
I think if you chose 1000 random people and prompted them with this question you would get a preponderance of "lefts" compared to the prompt, but not 100% left.