derefr

50 ms·

481.
by derefr 1y ago
> Most amusingly insomnia is a known risk of atomoxetine. Given that atomoxetine is an sNRI, and norepinephrine modulates both alertness and muscle tone, I would bet dollars to doughnuts that what this treatment is doing is mostly helpin
482.
by derefr 1y ago
> if I boot with clearcpuid=smap (not for prod use!), larger reads go significantly faster. On both Intel and AMD CPUs interestingly. Is there a page anywhere that collects these sorts of "turn the whole hardware security layer off&
483.
by derefr 1y ago
> Remember, you can get blood from humans in your own country, and prepare it there. So, why do you need it from the US? Random hypothesis: • equipment needs for uranium enrichment for Manhattan project in 1940s => US cornering the ma
484.
by derefr 1y ago
Not really. Presume that everything is time-reversed, and the game kills you if you do anything not possible when your playback is flipped around and played "forward." Then: - Obviously, you "emit" dots and fruit and p
485.
by derefr 1y ago
You missed it; the article does answer this. > By selecting ‘Get Book’ within the Kindle for iOS app, customers can now complete their purchase through their mobile web browser.
486.
by derefr 1y ago
There is one other very useful form of "expansion" that LLMs do. If you aren't aware: (high-parameter-count) LLMs can be used pretty reliably to teach yourself things. LLM base models "know things" to about the same
487.
by derefr 1y ago
That was not a good attempt at changing the style. You can't just say "don't sound like an LLM." The LLM does not in fact know that it is "speaking like an LLM"; it just thinks that it's speaking the way
488.
by derefr 1y ago
> They are invariably verbose, interminably waffly, and insipidly fixated on the bullet-points-with-bold style. No, this is just the de-facto "house style" of ChatGPT / GPT models, in much the same way that that that parti
489.
by derefr 1y ago
I take this not as people being dumb, but as a clear conflict of interest: people want to be able to rate the logistics provider separately from the product, but marketplaces don't want to give people the option to do that — as that wo
490.
by derefr 2y ago
What about convention centers, subway platforms, and other places where you have a lot of people packed together outside the reach of exterior towers? They stick microcells on the ceilings of these — wouldn't it make sense for those to
491.
by derefr 2y ago
It is this sort of case that makes me think that criminal justice systems should expect to output balanced-ternary outcomes by default: not “guilty or innocent”, but rather “defendant is provably at fault / no one is probably at fault
492.
by derefr 2y ago
Is it us, or is it corporations? Look at a specific microcosm: dating. Dating is "awful" now (according to people both young and old), in a particular way that it wasn't even ten years ago. And sure, this is in part because w
493.
by derefr 2y ago
> The theory here is largely mechanical; and it’s not just psychiatrists like Scott who are weak at mechanical explanations; it’s doctors in general as well as medical researchers and biologists. A tangent here, about not just "mech
494.
by derefr 2y ago
For all of these reasons, audio snooping is much more likely to be something done by wired, stationary devices that maybe have a decent amount of RAM + a fair bit of usually-idle processing capacity (to run the transcription model locally a
495.
by derefr 2y ago
A finer distinction might be between MLMs (where the multi-level part is very literal — everyone is being garnished by their upline and profiting from their downline), vs. flat marketing organizations that just give (centralized, corporat
496.
by derefr 2y ago
> If your intention is to learn, you are in luck! It's never been easier to teach yourself some skill for free. I'll emphasize this: for generally well-understood subjects, LLMs make incredibly good tutors. Talking to ChatGPT o
497.
by derefr 2y ago
> You either want to have global ordering of all messages on a given topic, or (more commonly) ordering of all messages with the same key. In contrast, defined ordering of otherwise unrelated messages whose key happens to yield the same
498.
by derefr 2y ago
Try this analogy out: it's no wonder that people are interested in / have demand for generic reproductions of licensed cultivars of a plant (e.g. buying generic "grape tomatoes" rather than specific, expensive "cher
499.
by derefr 2y ago
I feel like, despite the apochrycality of it, there’s a theologically-valid interpretation of the quote: namely, the same principle behind putting on your own breath mask first on a plane in an emergency. The Bible-compatible spin on this m
500.
by derefr 2y ago
> 2. I've been programming and hosting websites for a decade+ and I would have no idea where to start with any of the things that you propose they "just" require. That's because OV/EV certification are (currently
501.
by derefr 2y ago
Given usernames-as-domains, is there a reason to not just piggyback this on the X.509 web of trust? After all, we already have an established and highly-monitored set of sibling "trust roots" — we call them Certificate Authorities
502.
by derefr 2y ago
Bluesky's moderators, apparently. > For example, the New York Times can now issue blue checks to its journalists directly in the app. Bluesky’s moderation team reviews each verification to ensure authenticity.
503.
by derefr 2y ago
> AI is therefore designed to convince you it's right, which is a different optimization than actually being right. I get the impression, when talking to conversational AIs, that they're more tuned to convince you that you&#x2
504.
by derefr 2y ago
I'm the opposite — and I think this might be the "4D chess"† interpretation of this move as well. In peacetime, I think everyone is generally alright with something centralized like the CVE database. But in what increasingly
505.
by derefr 2y ago
> Good psychiatrists absolutely “call themselves doctors” and definitely seek to exclude or treat organic causes of psychiatric symptoms. Well, yeah, you're just stating the contrapositive corollary to my assertion: that psychiatris
506.
by derefr 2y ago
> I do not know why doctors are so hesitant to link the immune system and mood disorders. People who call themselves doctors — e.g. neurologists — generally aren't hesitant to do this. But psychiatrists — and even moreso, therapis
507.
by derefr 2y ago
What about believing that he's a particularly-easily-manipulated patse (esp. when it comes to things he doesn't care about), and so this is someone else playing 4D chess through him? For all the accusations of fascism, nobody se
508.
by derefr 2y ago
Exactly — one thing LLMs are great at, is basically acting as a coworker who happens to have a very wide breadth of knowledge (i.e. to know at least a little about a lot) — who you can thus ask to "point you in a direction" any ti
509.
by derefr 2y ago
Well, yes and no. You can in-context-learn an LLM into being a domain expert in a specific domain — at which point it'll start challenging you within that domain . But — AFAIK — you can't get current LLMs to do the thing that e
510.
by derefr 2y ago
An LLM is a very effective human-solution-description / pseudocode to "the ten programming languages we use at work, where I'm only really fluent in three of them, and have to use language references for the others each tim

More ›