Can you explain please, where I made a mistake?

Your mistake is thinking Earth is 6km in radius! :D 6km is how far you walk in an hour. Either you think Earth 1000 times too small or kilometer 1000 times too big.

For an object heavier than the Earth, 1g radius will be greater than the radius of Earth. For 56 Earth masses that’s sqrt(56) times bigger = 48000km.

A 56 Earth mass black hole will take 5.5e55 years to evaporate according to this calculator. A 100kg black hole (more close to what Richard used to be) is much smaller than the nucleus of an atom and will evaporate in 0.05 nanoseconds.

Curiously there was a paper recently that calculated that even if there was a small black hole in the center of the Sun, it would take millions of years for it to grow, because the aperture is so small not much can fit through, and the infalling gas heats up so much as to repel the rest, creating an internal hot bubble.

The credit companies do not insure against fraud, they simply take the money out of the merchant account and put it back into yours. Now it’s the merchant who has no recourse, if they have already shipped the product. So the only difference between CC and crypto is who is typically left holding an empty bag in case of theft - the payer or the payee. Certainly not the banks!

I’d argue in terms of assigning responsibility, it seems more fair to expect you the customer to keep your digital wallet secure from thieves, than to expect the merchant to try guess every time whether the visitor to their online store happens to be using a stolen credit card.

Question: do the Japanese actually care about privacy? I know I do, but if you were to ask a Japanese person why does their country use cash, would they say “We have considered a system of payment cards and decided against it for privacy reasons” or would they just shrug and say “I dunno, I’m not in charge of payment systems, I use what I have”?

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!

Can’t access the article, but wasn’t China the one most vulnerable from the Malacca Strait being a chokepoint? As in, their trade towards Europe and fuel from the Middle East being potentially threatened? How does Thailand pitching to the US make sense then? How would a Thai bypass even increase security, since both routes are in the same area and can be equally blockaded? There aren’t any problems with throughput capacity at Malacca, unlike say at the Panama Canal. Maybe it will make the travel distance slightly shorter, but is there really any way it could ever be cost-effective to offload and reload ships for a few hundred kilometers savings?

Thank you for your detailed input!

It’s not even a platonic ideal - it’s drawing a supply/demand curve and thinking you understand how prices work in a market economy.

You got me 😁. I love drawing supply-and-demand curves. Seems pretty hopeless then if to even begin to understand how to vote “correctly” you need 5 years of game theory PhD. Hearing someone say “just trust me bro, the optimal strategy is that one” is not good enough. Voting was supposed to be for the masses…

drop everything to just start suing states and protesting for voting rights

I could get onboard with ranked-choice voting. My city used IRV for our latest mayoral primary election, and even though none of my ranked candidates won, I felt extremely satisfied that at least my voice was finally being heard. When a literal police-mayor got elected (winning primary by only 7000 votes), I had the comfort of full knowledge that this was not due to any spoiler effect on my part, but solely simply due to more people voting for him. If we’d campaign for ranked-choice voting in federal elections - presidential primaries and general - we can eliminate all the above hand-wringing. The Democratic party should be totally on board with this since they could finally get the Green protest vote.

So I am proposing that the Democratic party is acting irrationally and suboptimally, but you claim that the Democrats are acting most optimally, and it is the fringe left that is acting irrationally instead by refusing to accept a unfair split against all game theory guidance, causing all of us to eat shit (despite them making up only low single digits). Yet if the Democrats are so rational, how come they keep losing? Shouldn’t they have found an optimal strategy to get around the irrational ultimatum of the left? Yet here we are.

I want people to be able to report bugs without any trouble.

Thank you for being aware! I’ve experienced this on github.com. I’ve tried to submit issues several times to open source projects, complete with proposed code to solve a bug, but github shadowbans my account 6 hours after creating it (because I use a VPN? a third-party email provider? do not provide a phone number? who knows). I can see the issue and pull request when logged in, but they only see a 404 on their project page even if I give them a direct link. I ended up sending them a screenshot of the issue page just to convince them this was even possible. Sad to hear gitlab does it even worse now by making phone mandatory.

the most a third party is going to do is shave off a few percentage points, resulting in the main party losing

If the third party can force the main party to lose, then it holds ultimatum power and game theory rules apply. The main party irrationally keeps rejecting the ultimatum and as a result keeps losing. To execute the threat of the ultimatum even after the unfair split has already been offered is the paradox of game theory. You have to appear credible enough to carry out such a threat, but the only reliable way to appear credible is to actually follow through on such threats every time.

The Democratic party keeps losing and shifting right because it acts irrationally and fails to execute optimal game theory strategy. It could have offered the left a fair split and we could have all had guaranteed single-payer medical care, food, and housing, but instead none of us will have women’s rights, and the immigrants and gays among us will be herded into cages.

Am I the only one for whom "open"subtitles.org hasn’t worked in years? I literally cannot find the download button, like in those okboomer memes. Never used the API. Switched to subscene.com and haven’t had problems since.

more evidence we need Civics back in our schools

Maybe we need more math as well - have you heard of the Ultimatum Game? Sometimes the rational strategy is to reject unfair split offers, even if that makes it a guarantee that you both get nothing.

I know traditionally the dream fantasy of book readers has been to own an expansive physical library, with shelf after shelf full of book spines, but I just could never get into it. I’m a data hoarder, not an object hoarder! All my books are digital, mandatory in plaintext DRM-free format, sorted and backed up. I find joy in the knowledge that everything I have ever read is instantly grep’able, ageless, and can fit in my pocket (on a thumbdrive) wherever I go.

I do prefer to read on e-ink as well, because the device is lighter than any book, guaranteed to fit in my pocket, can hold multiple books, and gives me control over font size. The only downside is when the battery gets old it needs more frequent recharging. A paper book will not refuse to work for lack of power!

German is perfect. Everyone will agree how to spell Schifffahrkarte the moment they hear it.

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

Solution: answer “yes” whenever asked but don’t do anything further. This will stop the questions since the “mystery” has been solved. If someone actually propositions, say “flattered but you are not my type”. End up drowning in pussy from all the gay-chaser girls who think they “can change you”.

I think it’s precisely because there is no governing body for English and all the rules are colloquial, developed through usage, that people do get grumpy! They are the only ones who can create and enforce the rules! Each English speaker feels personally responsible and compelled to correct use they perceive is in violation of the rules the way they want them to be. If they don’t do it right then and there, no one else can.

Btw, old reddit used to display the exact upvote/downvote counts (“x upvoted, y downvoted”) for posts AND comments in the “title” attribute of the total number, visible as a tooltip when hovering the mouse over it. All that “x% upvoted it” for posts and removing the counts entirely for comments was nonsense pushed through by the previous reddit jerk CEO on 2014-06-18 against all community opposition (what is it with reddit and dumping unpopular changes in June?).

If you gonna recreate old reddit, please do not include these later CEO meddlings in it! One of the best features of lemmy is the display of exact counts (which was also a major advertised draw of voat.com and raddle.me before it). Let’s keep that in!