yo-cuddles
97
26
yo-cuddles has not written any posts yet.
yo-cuddles has not written any posts yet.
There's an improvement in LLM's I've seen that is important but has wildly inflated people's expectations beyond what's reasonable:
LLM's have hit a point in some impressive tests where they don't reliably fail past the threshold of being unrecoverable. They are conservative enough that they can do search on a problem, fail a million times until they mumble into an answer.
I'm going to try writing something of at least not-embarrassing quality about my thoughts on this but I am really confused by people's hype around this sort of thing, this feels like directed randomness
Gotcha, you didn't sound OVER confident so I assumed it was much-less-than-certain, still refreshingly concrete
Ah, okay.
I'll throw in my moderately strong disagreement for future bayes points, respect for the short term, unambiguous prediction!
This is not going to be a high quality answer, sorry in advance.
I noticed this with someone in my office who is learning robotic process automation: people are very bad at measuring their productivity, they are better at seeing certain kinds of gains and certain kinds of losses. I know someone who swears emphatically that they are many times as productive but have become almost totally unreliable. He's in denial over it, and a couple people now have openly told me they try to remove him from workflows for all the problems he causes.
I think the situation is like this:
If you finish a task very quickly using automated methods, that feels viscerally... (read more)
By "solve", what do you mean? Like, provably secure systems, create a AAA game from scratch, etc?
I feel like any system that could do that would implicitly have what the OP says these systems might lack, but you seem to be in half agreeance with them. Am I misunderstanding something?
Definitely! However, there is more money and "hype" in the direction of wanting these to scale into AGI.
Hype and anti-hype don't cancel each other out, if someone invests a billion dollars into LLM's, someone else can't spend negative 1 billion and it cancels out: the billion dollar spender is the one moving markets, and getting a lot of press attention.
We have Yudkowsky going on destiny, I guess?
I think there's some miscommunication here, on top of a fundamental disagreement on whether more compute takes us to AGI.
On miscommunication, we're not talking about the lowering cost per flop, we're talking about a world where openai either does or does not have a price war eating it's margins.
On fundamental disagreement, I assume you don't take very seriously the idea that AI labs are seeing a breakdown of scaling laws? No problem if so, reality should resolve that disagreement relatively soon!
This is actually a good use case, which fits with what gpt does well, where very cheap tokens help!
Pending some time for people to pick at it to test it's limits, this might be really good. My instinct is legal research, case law etc. will be the test of how good it is, if it does well this might be it's foothold into real commercial use that actually generates profit.
My prediction is that we will be glad this exists. It will not be "phd level", a phrase which defaces all who utter it, but it will save some people a lot of time and effort
Where I think we disagree: This will likely not... (read more)
Also, Amodei needs to cool it. There's a reading of the things he's been saying lately that could be taken as sane but a plausible reading that makes him look like a buffoon. Credibility is a scarce resource
No, sorry, that's not a typo that's a linguistic norm that i probably assumed was more common than it actually is
Me and the people I talk with have used the words "mumble" and "babble" to describe LLM reasoning. Sort of like human babble, see https://www.lesswrong.com/posts/i42Dfoh4HtsCAfXxL/babble