6 ms·

> The only way to do that is to have some kind of smarts doing very non-trivial computations to arrive at the next-next-next-next...-next word that makes sense

> The only way to do that is to have some kind of smarts doing very non-trivial computations to arrive at the next-next-next-next...-next word that makes sense within the context of previous words and words that haven't been yet generated/sampled/statistically selected. Do you have any evidence for this claim? Like...I would like to believe this. I would like to say that LLMs are genuinely different in some way. But it just seems very possible that if you throw 10-100x the compute at "raw statistical correlation" it generates much better, specific sequences. > Perhaps there's a lot going on behind the scenes to make the trick work? Perhaps indeed!

roughly 3y ago
> But it just seems very possible that if you throw 10-100x the compute at "raw statistical correlation" it generates much better, specific sequences. This isn't just very possible, it's _what's happening_. They're fantastically complicated statistical machines. We know this - that part, at least, we _did_ write. The actual statistical model, we don't fully know, but we know what the machinery is. There's an impression here of understanding sufficient that it warrants both further investigation and a lot of introspection with regards to the nature of intelligence, but these are not systems with any real autonomy or agency, and in particular the _way_ they're currently used and deployed almost precludes talking about things like understanding or knowledge in any real sense - they're run as one-shot, feed tokens in get token out forward-pass algorithms. This is why I'm beating this drum - it's possible the algorithm that's being used underpins intelligence and understanding in a broad sense, it's possible (and some evidence suggests) the resulting models contain structures or patterns that encode information beyond the strict content of the input data, but when you interact with these systems today - like, in ChatGPT, Copilot, or anything else I'm aware of - you're inserting a string of tokens at the beginning of the model, receiving a predicted next token at the end of the model, and then repeating for however many tokens seem appropriate. This model is not talking to you, it does not understand what you asked, there is no growing mutual understanding, and assuming such will lead you to make assumptions about the behavior of the model that are incorrect.

roughly 3y ago

> But it just seems very possible that if you throw 10-100x the compute at "raw statistical correlation" it generates much better, specific sequences. This isn't just very possible, it's _what's happening_. They're fantastically complicated statistical machines. We know this - that part, at least, we _did_ write. The actual statistical model, we don't fully know, but we know what the machinery is. There's an impression here of understanding sufficient that it warrants both further investigation and a lot of introspection with regards to the nature of intelligence, but these are not systems with any real autonomy or agency, and in particular the _way_ they're currently used and deployed almost precludes talking about things like understanding or knowledge in any real sense - they're run as one-shot, feed tokens in get token out forward-pass algorithms. This is why I'm beating this drum - it's possible the algorithm that's being used underpins intelligence and understanding in a broad sense, it's possible (and some evidence suggests) the resulting models contain structures or patterns that encode information beyond the strict content of the input data, but when you interact with these systems today - like, in ChatGPT, Copilot, or anything else I'm aware of - you're inserting a string of tokens at the beginning of the model, receiving a predicted next token at the end of the model, and then repeating for however many tokens seem appropriate. This model is not talking to you, it does not understand what you asked, there is no growing mutual understanding, and assuming such will lead you to make assumptions about the behavior of the model that are incorrect.