mquander

9 ms·

31.
by mquander 1y ago
That's just not true. If Telegram wanted it a small amount, and xAI wanted it a large amount, it would be normal for Telegram to negotiate to get paid for doing it.
32.
by mquander 1y ago
The obvious case in which you would be "left behind" is the one in which LLM usage is hard, and you cannot otherwise do the hard things that LLMs are doing (or you can do them, but much slower and/or to a lower standard of qu
33.
by mquander 1y ago
I earned that wage circa 2008 and saved half my salary.
34.
by mquander 1y ago
If we picked up that plaque from space with an illustration of some aliens and an alien solar system and someone Tweeted it, the correct hypothesis for the numbering would develop plurality consensus within one hour. I don't know if al
35.
by mquander 1y ago
Seeing Krebs link to this downgrades my impression of how trustworthy his assessments are.
36.
by mquander 1y ago
Is there no non-monetary value in the other things that someone might be doing between the ages of 18 and 22, which may also involve learning?
37.
by mquander 2y ago
In what sense are the bleeding edge models incremental improvements over GPT-3 (read his examples of GPT-3 output and imagine any of the top models today producing them!), GPT-3.5, or GPT-4? Look at any benchmark or use it yourself. It&#x27
38.
by mquander 2y ago
A lot of people arguing in this thread. I'm an adult who started learning piano about two years ago. I found that the following exercise was immediately easy for me with no special practice: - Learn a piano song, memorize it, play it m
39.
by mquander 2y ago
If there's an actual problem with Chinese courts corruptly defending Tesla, why is this article presenting a case where Tesla seems to be in the right, rather than one in which the court made a clearly corrupt decision?
40.
by mquander 2y ago
The technology also makes it easier and cheaper to make good things, so the direction of the outcome isn't guaranteed.
41.
by mquander 2y ago
I tried to tell my English teachers that all through high school but it never worked.
42.
by mquander 2y ago
It's possible for people who aren't currently the president to do things.
43.
by mquander 2y ago
Although I agree with you that there's no reason to expect an equal dollar amount to produce a balanced outcome, I disagree completely with the conclusion. The paying party is a random website saying an offensive opinion, and the recei
44.
by mquander 2y ago
This is inaccurate because if you take the high end of the Falcon 9 per-launch emissions estimates discussed in this thread (about 28k metric tons equivalent of CO2) and divide by the payload size (about 17 metric tons) then each gram of ma
45.
by mquander 2y ago
I think tax rates should be higher. So does that mean I should have to pay extra taxes by myself and the rates for everyone else should stay the same?
46.
by mquander 2y ago
Isn't it dramatically easier to provide more useful history to machines? If I'm providing history to a doctor I am pretty much trying to jam the history into a two minute explanation, and they are trying to remember our previous i
47.
by mquander 2y ago
I don't know if this calculator was good or bad, but the rationale sounds superficially ridiculous. > Visitors who didn't see the calculator were 16% more likely to sign up and 90% more likely to contact us than those who saw i
48.
by mquander 2y ago
I was curious whether this was true but it seems made up. The source is this part of his biography by Ashlee Vance. The actual quotes only support that he thought CA HSR was mediocre, which is what made him want to suggest something better.
49.
by mquander 2y ago
The "efficient journal hypothesis" -- if something is written in a paper in a journal, then it's impossible for anyone to know any better, since if they knew better, they would already have published the correction in a journ
50.
by mquander 2y ago
Many people lack the numeracy to understand basic ideas about money and finance, which directly results in them getting scammed by banks, brokerages, credit card companies, and various hucksters.
51.
by mquander 2y ago
According to the bill, a model has "hazardous capabilities" if and only if the existence of the model makes it "significantly" easier to cause the damages the bill covers. If Google is equally good at telling you how to
52.
by mquander 2y ago
The bill doesn't give special treatment to "EA" models, so what does it matter whether projects are tied to EA or whether EAs are good stewards? Either it's a good law or it isn't. At a glance it looks like it'
53.
by mquander 2y ago
The parent claimed that you "can die up to 30-40 years earlier than you would've" which is completely insane and unsupportable. The review I linked shows that short sleep is associated with a 1.01-1.07x increased risk -- clos
54.
by mquander 2y ago
No you can't, e.g. https://www.nature.com/articles/srep21480
55.
by mquander 3y ago
If you ran an RCT where you could directly make people to be male and then see what happened to them, you wouldn't need any further modelling to distinguish between which is causing which. The reason you need the model is if you can&#x
56.
by mquander 3y ago
If they're rich, they should be able to easily solve this problem as soon as it starts causing trouble for them, by paying the workers much more money.
57.
by mquander 3y ago
Picture of two little identical castles, towns, and armies, caption: Their barbarous "barriers", "status quo", "ossified ideas" vs. Our blessed "privacy", "treasured social norms", "dig
58.
by mquander 3y ago
This is totally wrong, GPT-4 knows what it knows and doesn't know to a good degree of precision. Look at the calibration graphs on this page: https://openai.com/research/gpt-4
59.
by mquander 3y ago
What do you think it should do? It gives a better response than I would, since it notices that you might be trying to brainstorm some kind of plot or roleplay: > It sounds like you're describing an imaginative or hypothetical scenar
60.
by mquander 3y ago
When ChatGPT sees '473936482738338373926', it tokenizes it into 7 tokens, not 20. You can play with it here: https://platform.openai.com/tokenizer So to reverse the string, it can't just learn a kind of gener

More ›