Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
nullc
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
53 ms
·
841.
▲
by
nullc
2y ago
"Doesn't matter, had cybersex"
842.
▲
by
nullc
2y ago
SFO terminal 2 has something similar, though they might have chosen a different neologism for it.
843.
▲
by
nullc
2y ago
I also should have added that the ability to self explain when COT was in use only goes as deep as the COT, as soon as you probe deeper such that the content of the COT requires explanation the LLM is back in the realm of purely making stuf
844.
▲
by
nullc
2y ago
I agree people aren't great at this either and my post said as much. However we're familiar with the human limits of this and LLMs are currently much worse. This is particularly relevant because someone suffering from the mistaken
845.
▲
by
nullc
2y ago
> I mean just try it yourself with o1, go as deep as you like asking how it arrived at a conclusion I don't mean to disagree overall, but on this point the LLM can post-facto rationalize its output but it has no introspection and ha
846.
▲
by
nullc
2y ago
There were trackers before github for sure-- but like it's a pretty ordinary setup to make issues private to project members until triaged... that sort of thing pretty much kills brigading dead and it's a thing you can't do
847.
▲
by
nullc
2y ago
I agree that scale is the bigger factor, but the infrastructure matters too: Consider invite only vs writable-by-everyone default. Open source development moved from a lot more invite based in the mid 90s to a strong writable-by-everyone
848.
▲
by
nullc
2y ago
I was there, I'm personally convinced it's much worse now. Perhaps I'm suffering from rose colored glasses, wouldn't be the first time. My general impression is that the abuse in the past tended to concentrate in a few w
849.
▲
by
nullc
2y ago
Sources? Or are you just knowingly spreading false and defamatory claims?
850.
▲
by
nullc
2y ago
It isn't unique to the emulator scene by any means, though I'm sure some domains are worse than others. (e.g. due to the amount of interest from unsupervised children, among other factors) I think it's just simply that there
851.
▲
by
nullc
2y ago
The social media-izing of open source has been a massive blow. Projects and contributors across the board are randomly bombarded with mob harassment. Return to Cathedral.
852.
▲
by
nullc
2y ago
If uncertainty is an important signal then a model RL conditioned to perform good COT should be expected to learn how to encode an uncertainty sidechannel in its COT. If we're fortunate it'll do so using language choice that would
853.
▲
by
nullc
2y ago
Keeping the thinking interpretable makes it easier to impose conditions on it both at runtime and as part of reinforcement. It opens the doors to manually injecting relevant thoughts triggered by supervision ("I must remember to say n
854.
▲
by
nullc
2y ago
Hm. So I think if you define reasoning broadly -- as in collecting relevant data, enumerating and raking priorities, thinking through expected consequences... people do engage in this quite often in all areas of life, far more often than
855.
▲
by
nullc
2y ago
Total market control isn't the worst reason floating around out there, there are worse ones.
856.
▲
by
nullc
2y ago
In the US you can just adopt a new name, so long as you're not doing it for a fraudulent purpose (and obviously not by stealing someone elses identity). In this case, sure, he was ... _in the 1970s_. The initial name change was a cri
857.
▲
by
nullc
2y ago
Why was the law involved in any of this in the first place? Had Woods not been charged with identity theft in the first place Keirans' crime would have been victimless.
858.
▲
by
nullc
2y ago
Powerful AI technology being deployed against users to apply non-transparent and unaccountable censorship to their usage of these tools. Not exactly the brag they think it is. It wouldn't be much of a concern except for their efforts
859.
▲
by
nullc
2y ago
> my predictions about the outcome of AI are correct and theirs are not How very zizian of you.
860.
▲
by
nullc
2y ago
> What is your explanation? A combination of a psychological break when his sibling died and that being a doomsayer brought him a lot more more money, power, and worship per unit of effort and particularly per unit of meaningful work-lik
861.
▲
by
nullc
2y ago
> A lot of the "rationalist" scene would be more accurately called "rationalizationist". I've been saying this for years in varrious forms. Spot on.
862.
▲
by
nullc
2y ago
The human mind is quite malleable. You can make yourself mentally ill, or be made mentally ill by the people around you. This fact seems to be under-acknowledged because there is a slippery slope to blaming the victims. But just because y
863.
▲
by
nullc
2y ago
If you look at most great evil perpetrated by mankind, it's almost always someone trying to "do good". Sure there are serial killers or whatever, crimes of passion, but they just kill one or a few. The people who have kille
864.
▲
by
nullc
2y ago
Scandals with sex pests, minors kept in group houses, partnerships with ponzi scammers, orgies, grit sham charities that pocket the money, etc. isn't normal. It happens elsewhere, sure, but it's not normal. It's gonzo. But i
865.
▲
by
nullc
2y ago
> who try really hard to reason properly Who are really big into talking about reasoning with formal tools which few of them make any effort to understand or use. They might as well "try really hard to reason properly" with cry
866.
▲
by
nullc
2y ago
Suicidal ideation, for example, is common in the lesswrong community, even with people pledging to end their own lives before the machine overlord is able to scan their brains and simulate an infinitude of copies of them in a state of perpe
867.
▲
by
nullc
2y ago
Perhaps I overemphasized it, but a personal experience on that front was key to realizing that the lesswrong community was in aggregate a bunch of bullshit sophistic larpers. In short, some real world system had me asking a simply poised pr
868.
▲
by
nullc
2y ago
> Rationalists, by tending to overly formalist approaches, But they don't apply formal or "formalist" approaches, they invoke the names of formal methods but then extract from them just a "vibe". Few to none in
869.
▲
by
nullc
2y ago
[flagged]
870.
▲
by
nullc
2y ago
They brought a sword to a gun fight. Presumably they would have killed the landlord on the first go if he hadn't managed to shoot and kill on of the assailants first. But it seems they went back later and finished the job. (He survive
More ›