ein0p

21 ms·

301.
by ein0p 2y ago
To be fair, there's plenty to lambaste him for. But that's not the question. The question is, what's to prevent my doctor from prescribing "treatments" that are based on recently (or not so recently) discredited res
302.
by ein0p 2y ago
Maybe not "execute" but some real-world accountability would be nice. As in, perhaps, a lengthy prison sentence.
303.
by ein0p 2y ago
What do doctors prescribe nowadays? Same discredited meds? How about SSRIs, for which much of the research does not provide supporting evidence? E.g. there's a 2022 meta-study by Moncrieff et al that shows there's no consistent ev
304.
by ein0p 2y ago
It never was safe in the first place. Storing sensitive data in a locale under the jurisdiction where it can be freely accessed without your knowledge has always been idiotic. That's why all proper, sovereign countries demand that thei
305.
by ein0p 2y ago
Most definitely an unsolved problem at this point, so no demos of that from anyone.
306.
by ein0p 2y ago
The federal government can control what constitutes a school shooting, for example, in order to push anti-2A agenda. Idk if that's what happened here, hence the question.
307.
by ein0p 2y ago
I wonder why the school shootings went up so much under Biden, and why we heard so little about it.
308.
by ein0p 2y ago
That's because your $8B figure itself is a lie as well: https://x.com/DOGE/status/1892318654827524297 . It was never used when calculating the totals.
309.
by ein0p 2y ago
How about 226M for DEI bullshit that's just incomprehensible word salad: https://x.com/doge/status/1892350446456500349 . Just for starters. Is that "building" or "fraud" in your opinion?
310.
by ein0p 2y ago
Could we maybe stop showing robots vacuuming with a hand vacuum at this point? $200 robovacs have LIDARs now. Nobody is going to buy a $100K robot to do this.
311.
by ein0p 2y ago
How do you like your "liberal democracy", UK-ians? Is that democratic enough for you yet? Do you feel in control?
312.
by ein0p 2y ago
What's funny is that if you're well-read in your field, all such bullshit is plainly apparent. You know what the good baseline is, for example, and you know why the author didn't choose it. You know the deficiencies of the be
313.
by ein0p 2y ago
This is way too outdated to be relevant in any way. Back in the day they had a board with a TPU on it, before everyone else did. That board ran object detection at a pretty good resolution at like 80fps in 2.5W power budget. I still have th
314.
by ein0p 2y ago
Beatings will continue until openness improves, apparently. Kudos to Deepseek, about time someone spilled some significant beans.
315.
by ein0p 2y ago
30 years into this career I'm still learning to code, just different things than the ones I have started with. And without a shadow of doubt, AI is a godsend for me. You can have it code up a well known algorithm you already know in a
316.
by ein0p 2y ago
Feed costs money. Unless you live in an area where feed is very cheap, or grow your own feed, this isn't going to be economically viable. Having said that, some people enjoy keeping chickens as pets, and in that case economic viability
317.
by ein0p 2y ago
Browse Instagram, apparently.
318.
by ein0p 2y ago
There's no way this is 100% real though. No startup demo ever is.
319.
by ein0p 2y ago
Don't let the facts get in the way of the "Musk is a midwit who just stumbles into founding trillion dollar companies" story. ;) It's an article of faith for these people.
320.
by ein0p 2y ago
It doesn't choke anything. Anyone can go to GCP or other cloud providers and get as many GPUs as they need, within reason.
321.
by ein0p 2y ago
It is probably reproducible, if you have the requisite million cores. That isn't even difficult today - a million cores is about 100 GPUs.
322.
by ein0p 2y ago
It's hardly Google's problem that nobody else has a million cores, wouldn't you agree? Should they not publish the result at all if it's using more than a handful of cores so that anyone in academia can reproduce it? Tha
323.
by ein0p 2y ago
The statement is true irrespective of the number of views. If you watch the news you are misinformed. You believe things that just aren't true at all. From time to time your view of the world clashes with reality, and you reject the re
324.
by ein0p 2y ago
Nadella is currently claiming on X that this opens up a "direct path to 1 million qubits". Based on my priors, I put the probability of this statement being horseshit at 99.9%. Could someone knowledgeable make it 100%?
325.
by ein0p 2y ago
That applies to absolutely everyone. Convenient results are highlighted, inconvenient are either not mentioned or de-emphasized. You do have to be well read in the field to see what the authors _aren't_ saying, that's one of the p
326.
by ein0p 2y ago
"Custom proprietary license" pretty much guarantees nobody will use your thing.
327.
by ein0p 2y ago
It is a very bad idea to handle the KV cache in Jax naively like that. Jax requires static shapes. You're creating dynamic shapes there, causing a ton of recompilation.
328.
by ein0p 2y ago
> It's informative. This is not supported by evidence. https://www.youtube.com/watch?v=Z4RYyYtD6i4
329.
by ein0p 2y ago
So basically it was a billion-dollar taxpayer funded gravy train. Nobody could predict this, absolutely nobody.
330.
by ein0p 2y ago
Who specifically? Could you name names? Or are you going to ask us to believe without evidence that the guy who got FIVE mega-Unicorns off the ground (3 of them "impossible") "understands nothing"?

More ›