I knew that motion sickness is triggered by frequent starts and stops and frequent turns, but even I was not aware of how big a contribution the engine vibration makes until I got to experience a ride without it.

I was apprehensive about EVs but the first time I rode in one I immediately fell in love with it. I get carsick easily, and the super-smooth ride without the chug-chug-chug of an internal combustion engine made the experience surprisingly much more pleasant for me. I do not use a car, but if I had to buy one, I don’t think I could ever stomach an ICE again knowing that this alternative is available.

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.

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!

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

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.

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.

The trolley is like really slow.

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.

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

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.

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.

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

Through the headphones :P

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.

They incorrectly broke it because they were overzealous.

There was some scare in lemmy development circles recently about script injection vulnerabilities. The various apps and frontend developers “solved” the problem by peppering untrusted user input with escape sequences all over the place. User submits post? Escape title! Receive new post from a federated instance? Escape title!

Obviously if you escape the title twice and display once, it will show up weird. The problem is that the various devs haven’t agreed yet which parts of the messaging protocol are supposed to be already escaped and which are not. Ideally all user input should be stored and transmitted in raw form, and only escaped right before displaying. But due to various zealously-cautious devs we get this instead:

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.

Central Park is not New York’s largest park. It’s the 5th.

Found this paper from 2019 with open access, where they do double slit electron diffraction and then slide a shutter in to close the second slit.

They talk about how it was never actually possible to do this before, because it requires very fine “electron optics” and manufacturing of components, like slits and shutters, with nanometer precision. So while the thought experiment with electrons itself was proposed by Feynman in 1963 (which is probably what inspired the monkey meme and the like), it was not actually realized until 2019. I’m also now guessing that the electron quantum eraser paper from 2014 doesn’t use a double slit but some other electronic quantum circuit that is easier to work with.

The two-stripe photo to match the monkey meme, with electrons and measuring which-way information, probably doesn’t exist yet. So that’s why!