Imagine you are the mainstream media and you see that on a website interested in ai people are sharing “calmness videos”. Would your takeaway be that everything is perfectly fine? :D
Throwaway2367
I think “erasure” in the “<noun> erasure” construct (like “trans erasure”) is meant to convey a similar concept, though I am not a native speaker so I can’t say for sure.
Why would you not title your post “Significantly Enhancing Adult Intelligence By Banging Head On The Wall May Be Possible”? This is just below the other on the front page, it would have been perfect!
On a more serious note: It sure is hard to fake those talents, much harder at least than faking not having them. I think they probably practiced before in secret then said this story. One relevant and similar situation is when people can talk fluidly in a foreign language after a coma.
Would it be fair to describe this concept abstractly as follows:
“Two process can theoretically generate exactly the same set of objects, but with different probabilities. Therefore, when presented with a given object, one can use Bayes-theorem to infer which process generated it.”
?
The second part of the second rootclaim debate (90 minutes) https://youtu.be/FLnXVflOjMo?si=dPAi1BsZTATxEglP
How is it that bad at codeforces? I competed a few years ago, but in my time div 2 a and b were extremely simple, basically just “implement the described algorithm in code” and if you submitted them quickly (which I expect gpt-4 would excel in) it was easy to reach a significantly better rating than the one reported by this paper.
I hope they didn’t make a mistake by misunderstanding the codeforces rating system (codeforces only awards a fraction of the “estimated rating-current rating” after a competition, but it is possible to exactly calculate the rating equivalent to the given performance from the data provided if you know the details (which I forgot))
When searching the paper for the exact methodology (by ctrl-f’ing “codeforces”), I haven’t found anything.
Anthropic should let Claude be used in the EU.
Ah yes, there is no way situational awareness might emerge in LLMs, just no way at all..
I know. I skimmed the paper, and in it there is a table above the chart showing the results in the tasks for all models (as every model’s performance is below 5% in codeforces, on the chart they overlap). I replied to the comment I replied to because thematically it seemed the most appropriate (asking about task performance), sorry if my choice of where to comment was confusing.
From the table:
GPT-3.5′s codeforces rating is “260 (below 5%)”
GPT-4′s codeforces rating is “392 (below 5%)”
Cases clustering at wetmarket, proline at fcs, otherwise suboptimal fcs, out of frame insertion, WIV scientists’ behavior after leak (talking about adding fcs to coronavirus in december, going to dinner, publishing ratg13), secret backbone virus not known (for some reason sars not used like in other fcs insertion studies), 2 lineages at market just off the top of my head
- 10 Feb 2024 3:25 UTC; 3 points) 's comment on Brute Force Manufactured Consensus is Hiding the Crime of the Century by (
But I’m not likely to regularly check your blog.
Just noting in case you (or others reading) are not familiar that Substack provides an RSS feed for every blog.
Happy Birthday!
Embarrassingly minor nitpick I’m too neurotic to not mention: It’s the ceil of N/2 instead of floor.
Whats your dictionary? Google says: “the action or practice of inflicting severe pain or suffering on someone as a punishment or in order to force them to do or say something.” which feels closer to the word’s meaning (as I use it) This definition technically also doesn’t apply. It fails at least the “someone” part as animals are not someones.
However, and more importantly, both this objection and yours aren’t really relevant to the broader discussion as the people who “avoid thinking about whether categories like ‘torture’ apply” would only care about the “extreme suffering” part and not the “purposeful” or “human” parts (imo).
In this respect this is an inverse non-central fallacy. In a non-central fallacy you use a word for somthing to evoke an associated emotional response which in the first place got associated to the word for an aspect not present in the specific case you want to use it for. Here you are objecting to the usage of a word even though the emotional response bearing aspect of the word is present and the word’s definition does not apply only because of a part not central to the associated emotional response.
I have not seen the Simulation Trilemma and anthropic reasoning mentioned in any of the other comments, yet I think those topics are pretty interesting.
Also +1 for FDT.
Unfortunately also the most likely person to be in charge of an AGI company..
Had it got it right, that would have probably meant that it memorized this specific, very common question. Memorising things isn’t that impressive and memorising one specific thing does not say anything about capabilties as a one line program could “memorize” this one sentence. This way, however, we can be sure that it thinks for itself, incorrectly in this case sure, but still.
I hope I’m not the only one who sees an elephant..
Anthropic should let Claude be used in the EU.
I will give a simple argument:
-
Given that it is already deployed, increasing the region of its legal availability is very unlikely to increase AI risk.
-
The EU also has people who might be able to contribute to AI safety research who will contribute less if the frontier models are not available legally in the EU.
-
Therefore, the action has net benefits.
-
I believe the action’s costs are much less.
-
You did the same thing Peter Miller did in the first rootclaim debate just for the opposite side: you multiplied the probability estimates of every unlikely evidence under your disfavored hypothesis, observed that it is a small number then said a mere paragraph about how this number isn’t that small under your favored hypothesis.
To spell it out explicitly: When calculating the probability for your favored hypothesis you should similarly consider the pieces of evidence which are unlikely under that hypothesis!! Generally, some pieces of evidence will be unlikely for one side and likely for the other, you can’t just select the evidence favorable for your side!