TauZero

@TauZero@mander.xyz

• ⁨Comment⁩ on ⁨Inside the 'arms race' between YouTube and ad blockers / Against all odds, open source hackers keep outfoxing one of the wealthiest companies.⁩ ⁨⁨11⁩ ⁨months⁩ ago⁩:

  By some argument, section 103 of the DMCA (which is what grandparent post is referring to) does make it illegal to even talk about DRM circumvention methods.

  illegal to: (2) “manufacture, import, offer to the public, provide, or otherwise traffic in” a device, service or component which is primarily intended to circumvent “a technological measure that effectively controls access to a work,” and which either has limited commercially significant other uses or is marketed for the anti-circumvention purpose.

  If youtube implements an “access control measure” by splicing the ads with the video and disabling the fast-forward button during the ad, and you go on a forum and say “Oh yeah, you can write a script that detects the parts that are ads because the button is disabled, and force-fast-forwards through those”, some lawyer would argue that you have offered to the public a method to circumvent an access control measure, and therefore your speech is illegal. If you actually write the greasemonkey script and post it online, that would definitely be illegal.

  This is abhorrent to the types among us for whom “code IS free speech”, but this scenario is not just a hypothetical. DMCA has been controversial for a long time. Digg collapsed in part because of the user revolt over the admins deleting any post containing the leaked AACS decryption key, which is just a 32-digit number. Yet “speaking” the number alone, aloud, on an online platform (and nothing else!) was enough for MPAA to send cease and desist letters to Digg under DMCA, and Digg folded.

• ⁨Comment⁩ on ⁨Browse through 4096 concepts identified within a 512-node toy neural network⁩ ⁨⁨11⁩ ⁨months⁩ ago⁩:

  Some notes for my use. As I understand it, there are 3 layers of “AI” involved:

  The 1st is a “transformer”, a type of neural network invented in 2017, which led to the greatly successful “generative pre-trained transformers” of recent years like GPT-4 and ChatGPT. The one used here is a toy model, with only a single hidden layer (“MLP” = “multilayer perceptron”) of 512 nodes (also referred to as “neurons” or “dimensionality”). The model is trained on the dataset called “Pile”, a collection of 886GB text from all kinds of sources. The dataset is “tokenized” (pre-processed) into 100 billion tokens by converting words or word fragments into numbers for easier calculation. You can see an example of what the text data looks like here. The transformer learns from this data.

  In the paper, the researchers do cajole the transformer into generating text to help understand its workings. I am not quite sure yet whether every transformer is automatically a generator, like ChatGPT, or whether it needs something extra done to it. I would have enjoyed to see more sample text that the toy model can generate! It looks surprisingly capable despite only having 512 nodes in the hidden layer. There is probably a way to download the model and execute it locally. Would it have been possible to add the generative model as a javascript toy to supplement the visualizer?

  The main transformer they use is “model A”, and they also trained a twin transformer “model B” using same text but a different random initialization number, to see whether they would develop equivalent semantic features (they did).

  The 2nd AI is an “autoencoder”, a different type of neural network which is good at converting data fed to it into a “more efficient representation”, like a lossy compressor/zip archiver, or maybe in this case a “decompressor” would be a more apt term. Encoding is also called “changing the dimensionality” of the data. The researchers trained/tuned the 2nd AI to decompose the AI models of the 1st kind into a number of semantic features in a way which both captures a good chunk of the model’s information content and also keeps the features sensible to humans. The target number of features is tunable anywhere from 512 (1-to-1) to 131072 (1-to-256). The number they found most useful in this case was 4096.

  The 3rd AI is a “large language model” nicknamed Claude, similar to GPT-4, that they have developed for their own use at the Anthropic company. They’ve told it to annotate and interpret the features found by the 2nd AI. They had one researcher slowly annotate 412 features manually to compare. Claude did as well or better than the human, so they let it finish all the rest on its own. These are the descriptions the visualization shows in OP link.

  Pretty cool how they use one AI to disassemble another AI and then use a 3rd AI to describe it in human terms!

• Browse through 4096 concepts identified within a 512-node toy neural networktransformer-circuits.pub ↗
  Submitted ⁨⁨11⁩ ⁨months⁩ ago⁩ to ⁨science@mander.xyz⁩ | ⁨2⁩ ⁨comments⁩
• ⁨Comment⁩ on ⁨Weird 🤔⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  Oh I was well aware what community I was in 😁. I hate cars and exclusively ride bikes myself and here I was making a joke how I managed to get !fuck_cars of all places to downvote me for not watching fox news. All because my groupthink is not exactly identical to their groupthink (I am not the grandparent comment btw).

  The secret is that karma does not matter anywhere! However, as long as the comment sorting algorithm is the way it is, I will keep believing that the downvote button is for posts that are non-constructive contributions, not for disagreement. Burying discussion is not constructive, but that’s what the algorithm will do. Maybe this is a hopeless task, but I wish that after a conversation I have learned something new, or taught someone something, not just made myself feel better.

• ⁨Comment⁩ on ⁨Astronauts dropped a tool bag during an ISS spacewalk, and you can see it with binoculars⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  The joke is that the orbit was clearly originally reported in kilometers, but the article editor “helpfully” converted it to miles and reported it in miles as default, but it makes no sense now because the same “miles” number now equals two different “kilometers” numbers.

• ⁨Comment⁩ on ⁨First live birth of a chimeric monkey using embryonic stem cell lines⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  To be clear, human chimeras already exist naturally, from the fusion of twin embryos in utero. Most of them go entire lives without even realizing it. Only occasionally it pops up in the news when someone receives a negative paternity test and after lots of stress and hairpulling and doctor’s visits it turns out that their blood comes from a different cell line than their balls.

  Human-ape chimeras are the stuff of bioethicists’ nightmares and thankfully illegal everywhere civilized.

• ⁨Comment⁩ on ⁨Weird 🤔⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  Fox the broadcast TV channel is different from Fox News the cable channel. Broadcast TV is operated by one of ~100 local affiliates and shows the Simpsons and local news. Fox News Channel is the Murdoch personal project to produce 24-hour conservative propaganda to shift the national discourse. Or at least it was this way 20 years ago, haven’t seen what the TV branding looks like nowadays.

• ⁨Comment⁩ on ⁨Weird 🤔⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  My last comment got downvoted for saying I have never watched Fox news 🤷🏾.

• ⁨Comment⁩ on ⁨Sleeping Beauty Trolley Problem⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  Exactly! Trying to think outside the box in a trolley problem is like wishing you could wish for more wishes in a genie problem.

• ⁨Comment⁩ on ⁨Sleeping Beauty Trolley Problem⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  Excellent excellent!

  If 6 is rolled, then P(X|R=6) = (N-1 choose 9)/(N choose 10)

  Might as well reduce that to 10/N to make the rest of the lines easier to read.

  If you don’t flip it, you have a 2/3 chance of dying.

  There is also a chance that your switch is not connected and someone else has control of the real one. So there is an implicit assumption that everyone else is equally logical as you and equally selfish/altruistic as you, such that whatever logic you use to arrive at a decision, they must have arrived at the same decision.

  No matter what your goal is, given the information you have, flipping the switch is always the better choice.

  That is my conclusion too! I was surprised to learn though in the comment thread with @pancake that the decision may be different depending on the percentage of altruism in the population. E.g. if you are the only selfish one in an altruistic society, you’d benefit from deliberately not flipping the switch. Being a selfish one in a selfish society reduces to the prisoner’s dilemma.

• ⁨Comment⁩ on ⁨Sleeping Beauty Trolley Problem⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  there’s no way to know which track the trolley is on

  It’s a standard trolley meme problem, the trolley will keep going on the main track unless the lever is switched 😁. I thought !science_memes would be familiar with trolley problems, but I guess I get to introduce some of you! You might want to start off on some easier trolley memes first, this is advanced level stuff.

  where the real lever sends it

  There is not usually ambiguity with the lever. If you wish, you can have an announcement in the headphones “main track… side track…” every time you flip the lever. Your only uncertainty is which track you yourself are bound to, given how you’re blindfolded.

  there’s a 0.017% chance

  1/6 * 10% = 1/60 = 0.01666… = 1.666…% ~= 1.7%! Careful there!

  It’s not really a trolley problem, because in both scenarios a track is empty,

  Everything is a trolley problem.

• ⁨Comment⁩ on ⁨Sleeping Beauty Trolley Problem⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  The trolley is like really slow.

• ⁨Comment⁩ on ⁨Sleeping Beauty Trolley Problem⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  You are on the side track in scenario A. You die if you pull. Ironically, you’d be killing yourself. The dice are to make the two scenarios not equally likely.

• ⁨Comment⁩ on ⁨Sleeping Beauty Trolley Problem⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  Half the fun of trolley problems is adapting them to puzzles for which they are utterly unsuitable:

  Image

• ⁨Comment⁩ on ⁨Sleeping Beauty Trolley Problem⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  It’s a standard runaway trolley problem. The trolley is traveling down the main track unless the switch is flipped to send it down the side track. The lever is labeled such that there is no ambiguity which way it is set, the blindfolds notwithstanding. The villain is pernicious and will be equally (though not exceedingly so) delighted to see you die by your own action where inaction would have had saved you. You can somehow trust that the announcement in the headphones is true and not a lie. Such as, for example, you have seen this exact situation happen many times before on TV and survivors/witnesses have described the villain to be truthful every time.

• ⁨Comment⁩ on ⁨Sleeping Beauty Trolley Problem⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  Good question to ask, since specifics of selection process may affect the decision outcome! Other variants include growing humans in a vat from scratch on demand, using Star Trek transporter clones, or abducting the necessary number of people from a pre-selected list where your name happens to be the first one. For now, imagine the potential population as the 5 billion living cognizant adults.

  as X approaches infinity, the difference becomes negligible

  It may be negligible to the 4.999… billion adults sleeping comfortably and securely in their beds tonight, but the problem presupposed that you have already been abducted. It remains underdefined whether you refers to you the specific person reading this meme, or a more general you-the-unfortunate who has been chosen and is now listening to this on the headphones.

• ⁨Comment⁩ on ⁨Sleeping Beauty Trolley Problem⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  Or spit, or blow air at your potential neighbor, or fart in their general direction!

• ⁨Comment⁩ on ⁨Sleeping Beauty Trolley Problem⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  Through the headphones :P

• Sleeping Beauty Trolley Problemmander.xyz ↗
  Submitted ⁨⁨1⁩ ⁨year⁩ ago⁩ to ⁨science_memes@mander.xyz⁩ | ⁨41⁩ ⁨comments⁩
• ⁨Comment⁩ on ⁨The real double-slit quantum eraser they don't want you to know about!⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  cannot be accurately measured

  I want to clarify that the “cannot” here refers not to the inadequacy of our tools (which hypothetically could have been fixed in the future by building better tools), but by a fundamental prohibition of the quantum mechanics theory. Practically, the single-photon lasers and detectors used here are like 90%+ efficient - plenty good enough to distinguish between the two monkey scenarios. But some observables in quantum mechanics are “orthogonal” - you can measure one or the other, but not both at the same time - the math will not allow it. The typical example of that is “position” and “momentum” of a particle.

  The math is quite beautiful actually, the analogy I’d use is something like asking “Which way is east at the North Pole?” In your head you can either know “This direction is east.” or “I am standing at the North Pole.” but you cannot hold both pieces of knowledge in your head at the same time.

  The orthogonal observables in this experiment are the “which-way top/bottom slit” information and the “which-interference-category Pattern 4/Pattern 5” information. It’s even more beautiful in the delayed choice quantum eraser experiment that I was ranting about here. There, both pieces of information are stored orthogonally in a single photon. You can choose at a later time to either measure it one way, which will tell you the which-way info, or in a different way, which will tell you the interference category info, but there is no hypothetical way to measure it in both. The only way you can get the category info out to allow your computer to draw the interference pattern is if you guarantee that the which-way information has been irrecoverably erased. It is as if the whole universe conspires to censor this information from you! But it’s just the consequence of the math rules in use.

• ⁨Comment⁩ on ⁨Black holes could come in 'perfect pairs' in an ever expanding universe⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  Oh! They don’t mean that black holes must come in perfect pairs! The headline makes it sound like it’s about wormholes across vast distances. No! What they’ve found is a stable “orbit” solution for the two-body problem. Normally when you place two bodies anywhere in an empty universe, they will gravitate towards each other until they collide. But in a universe with dark energy, there is some perfect distance between them, where the accelerating expansion perfectly counterbalances the accelerating attraction. They’ve used general relativity math to actually calculate such an arrangement.

  The “stable” orbit in this case is the same kind of stable as a pencil balanced on its sharp tip - if it tilts even slightly one way it will fall out of control. Although they tantalize the idea that they might be able to make it truly stable against small perturbations once they finish their spinning black hole solution.

  I would like to have known some specific numbers examples! Like if you have as much dark energy as our universe, and two 10-solar-masses stellar black holes, how far apart would that be? Is it like 1Ly or 1MLy? How far for two 10-million-solar masses supermassive black holes? The formulas they created should give the exact answer but I am not skilled enough to substitute the correct numbers for the letters.

• ⁨Comment⁩ on ⁨youtube getting more agressive⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  Piped/invidious work by scrapping the video chunks directly from google and proxying them through volunteer servers. They will stop working as soon as google gets around to locking down the APIs that they are abusing, or blocks their server IPs.

• ⁨Comment⁩ on ⁨Thousands of IT workers contracting with U.S. companies have for years secretly sent millions of dollars of their wages to North Korea for use in its ballistic missile program, FBI says⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  This is like that Key and Peele sketch where they rob the bank by working in there for 20 years.

• ⁨Comment⁩ on ⁨youtube getting more agressive⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  We are totally gonna get DVRs back, aren’t we 🤣?

• ⁨Comment⁩ on ⁨The real double-slit quantum eraser they don't want you to know about!⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  In this experiment, they didn’t even bother measuring which slit each photon passes through. The 3D glasses don’t measure or observe the photons, they merely polarize them (although they do block 50% of light). The detector D_S doesn’t measure which-path information either. The researchers could have placed a circular polarizer in front of D_S, and when they get a hit they could have said with confidence “this photon came through the top slit!” but they didn’t even bother doing it this time. The fact that the 3D glasses alter the light in a manner which makes the which-path information theoretically measurable (even if not actually measured), alone is sufficient to destroy the interference pattern.

• ⁨Comment⁩ on ⁨The real double-slit quantum eraser they don't want you to know about!⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  The paper doesn’t use an actual screen, they only have the detector D_S that they move up and down to record the coincidences. I simulated what the monkey would see had there been a screen in place for the purpose of the meme. I copied down the datapoints from the graph and simulated 100,000 photons hitting the screen with the probabilities indicated by those points. My javascript pastie is available here: html.cafe/xcd2a5ed3?k=19f51bff26c65bcf253ee5257a5… Importantly, the monkey can never see images 4 and 5 on the physical screen - those can only be displayed on the computer. The monkey will only ever see image 3, which is the sum of 4 and 5.

• The real double-slit quantum eraser they don't want you to know about!mander.xyz ↗
  Submitted ⁨⁨1⁩ ⁨year⁩ ago⁩ to ⁨science_memes@mander.xyz⁩ | ⁨36⁩ ⁨comments⁩
• ⁨Comment⁩ on ⁨City-size comet racing toward Earth regrows 'horns' after massive volcanic eruption⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  The poster is just submitting the article as titled, and the scientists are just making cool observations not bothering anyone. The fault for clickbait lies solely with the journalists/editors. I wish instead of continuing the practice from that other site of “submission title must match article title”, we normalize the culture of writing our own plain language de-clickbait-ified titles, the way HackerNews does it.

• ⁨Comment⁩ on ⁨City-size comet racing toward Earth regrows 'horns' after massive volcanic eruption⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  What a disgusting clickbait title 🤮. The comet is not “racing towards Earth” any more than say Mars is (and Mars comes closer to Earth than this comet will). Maybe some day we’ll live in a world where journalists do not sensationalize headlines with a mix of Deep Impact and The Ninth Gate, but today is not that day.

• ⁨Comment⁩ on ⁨What is the name of this type of image⁩ ⁨⁨1⁩ ⁨year⁩ ago⁩:

  The picture is clearly at the very least a composite, because there are zero clouds anywhere. I was skeptical whether it can be called a “photo”. Given how clear the unlit terrain is, even in the ocean around the Bahamas for example, I thought it must have been a visualization, or a photo of daytime terrain shaded blue and overlaid with a map of nighttime lights. But I found the actual source:
  …nasa.gov/…/night-lights-2012-map
  …nasa.gov/…/dnb_land_ocean_ice.2012.13500x13500.B…
  It really is a (composite) photo taken by the Suomi National Polar-orbiting Partnership satellite, whose cameras are so sensitive they can see reflected moonlight and “the nocturnal glow produced by Earth’s atmosphere”, albeit partially in the infrared.

  This new image of the Earth at night is a composite assembled from data acquired by the Suomi National Polar-orbiting Partnership (Suomi NPP) satellite over nine days in April 2012 and thirteen days in October 2012. It took 312 orbits and 2.5 terabytes of data to get a clear shot of every parcel of Earth’s land surface and islands.

  The nighttime view of Earth was made possible by the “day-night band” of the Visible Infrared Imaging Radiometer Suite. VIIRS detects light in a range of wavelengths from green to near-infrared and uses filtering techniques to observe dim signals such as gas flares, auroras, wildfires, city lights, and reflected moonlight.

  I’m unsure though what “assembled from data” means exactly. At the very least the colors are artificial, shifted from the infrared-to-green range of the camera into human visual range. This page describes some more how the sensor functions, along with raw photos:
  earthobservatory.nasa.gov/features/IntotheBlack