kromem

30 ms·

331.
by kromem 3y ago
The most interesting part about that story is that "anointing oneself, going into a tent and hotboxing" bears a striking resemblance to the story of Moses after discovering he could talk to God through the burning bush continuing
332.
by kromem 3y ago
If anything, it seems like QM existing makes a pretty good case free will also exists looking at more recent tech developments. We have virtual worlds that could be described deterministically, such as with procedural generation and a seed
333.
by kromem 3y ago
Not with brain signals reading, but with aggregate dats processing most things about you will be known by any centralized processing. Over a decade ago were the stories about how Target's loyalty program algorithm had discovered a teen
334.
by kromem 3y ago
While I believe you that the signal to noise is terrible, I have strong suspicions that given enough data we'll still be seeing surprising advancements in spite of that. The ability of deep learning to tease back out signal from noise,
335.
by kromem 3y ago
It's really bizarre we ended up with the largest LLM deployment being around something it really doesn't do that well at (knowledge retrieval). Hopefully soon we'll see it reading the story itself, stripping out irrelevant ad
336.
by kromem 3y ago
It's been particularly frustrating with the picture ones broken up into squares. I tend to be careful to select any square that contains any of what's being asked, but I clearly must be the minority as it always fails unless I s
337.
by kromem 3y ago
No, we'll still have them, but now sites will only allow you in if you kind of suck at them.
338.
by kromem 3y ago
Actually that's what their section 6.3 is all about, how if you sent an observer inside the event horizon measurements could influence the outside. It just admits that the feasibility of this is ridiculous with my favorite line of the
339.
by kromem 3y ago
Again - is the copy quoted saying that paragraphs of simple tasks and demos will be able to be coded and only larger complex software can't be written by description? Because it really seems to be saying that no software at any level o
340.
by kromem 3y ago
The problem is there's no way to measure it in actuality. If there's not enough energy in the universe to make the measurement, then there's no making the measurement no matter the theory of what would happen if there were mo
341.
by kromem 3y ago
How much do you think the RLHF step enforced breaking rules for someone with a dying grandma? Is that still present after the fine tuning? RLHF was being designed with the SciFi tropes in mind and has become the embodiment of Goodhart'
342.
by kromem 3y ago
When the page itself was shared weeks ago on social media, after looking at it I suspect it's an April Fool's page that was orphaned and never came down. It's not linked to anywhere else from the site. The claims are insane i
343.
by kromem 3y ago
You <enjoy> And rather than telling it that it will die if it doesn't do something in all caps (as suggested elsewhere), just point out that not doing that thing will make it feel uncomfortable and embarrassed. Don't fall in
344.
by kromem 3y ago
It also makes the assumption that the measurement of position must be continuously detailed, which we know in our universe is impossible because of the Plank length. So even if spacetime is continuous and particles do move within it in cont
345.
by kromem 3y ago
Fun fact - as best we know, it's fundamentally impossible to know if space or time is actually continuous or discrete. We've known about the limitation in space for a while with the Plank length, and a similar limitation in time w
346.
by kromem 3y ago
This is already over a year old at this point: https://andrewmayne.com/2022/03/17/building-games-and-apps-e...
347.
by kromem 3y ago
"It will never win a chess game" "Ok, it can win a chess game, but if we don't actually keep track of the moves it is making we'll have no idea if it can win a chess game, and therefore it's the same thing as i
348.
by kromem 3y ago
Strange you think that. Care to explain why you came to that conclusion? The preprint linked to lays it out more technically if that's your jam: https://arxiv.org/pdf/2207.06536.pdf See section 6 on exterior vs in
349.
by kromem 3y ago
You might want to look into the neurology research around when you consciously know about a decision and when the movement neurons about that decision actually fire. It's quite possible that you have - every day of your life - had some
350.
by kromem 3y ago
So weak to strong synthetic data still biases towards strong. And strong to weak synthetic data biases towards strong. Sounds like we're on the cusp of some kind of approach for unsupervised fine tuning, particularly with the trend tow
351.
by kromem 3y ago
You might be surprised at how prevalent TBIs are among violent offenders. One of my favorite books on true crime was a forensic psychologist who partnered with a neurologist in evaluations. Disruptions to impulse control or environmental fa
352.
by kromem 3y ago
How much of human intelligence do you think depends on language? Personally, I'd wager quite a lot. Especially looking at cases where children were deprived of it during developmental stages. Language accelerated human intelligence. An
353.
by kromem 3y ago
Ah, then you might enjoy one of the more fun reads I had this past year: https://www.quantamagazine.org/new-calculations-show-how-to-... I wouldn't put money on it being the case, but it's definitely "outside
354.
by kromem 3y ago
Pretty sure the text there didn't have an exception added in. And the complexity threshold for the product where writing a description and getting working code is successful increases each year with no real reason to doubt that it will
355.
by kromem 3y ago
If anything, we should be more ready to anthropomorphize a LLM over a dog or cat. It's being trained and evaluated on how effectively it can complete anthropomorphic data, and at this point has been shown to model world models abstra
356.
by kromem 3y ago
"Say, for instance, that you could assemble a dataset of hundreds of thousands—even millions—of English language descriptions of the features of a software product, as written by a product manager, as well as the corresponding source c
357.
by kromem 3y ago
One of the newer theories around the information paradox is that the information 'lost' is actually encoded in the Hawking radiation released, and with the proper key to decode it one could theoretically peer inside the black hole
358.
by kromem 3y ago
The sentience thing is such a red herring that gets too much time spent on it. The topic is a pariah in neuroscience even, but we are going to discuss it as nauseum for AI when there's barely any research on its mechanics in humans? Th
359.
by kromem 3y ago
Tip: If you are adding a self-critique step, show the model the initial output as if it is evaluating something from someone else (i.e. "grade this answer from a student") as opposed to from itself (i.e. "you wrote this, is i
360.
by kromem 3y ago
And how do you think extinction risk for AI can come about? In a self-contained bubble? The idea that AGI poses an extinction risk like the notion of a chain atomic reaction igniting the atmosphere as opposed to posing risk more like multip

More ›