I think it’s all made up. The most power hungry data center in the world consumes 150MW of power, and that’s from a massive 11 million square foot facility in China that’s significantly larger than any other data center.
EDIT: an hour of heavy thinking does consume approximately 12 watt hours though, so that figure seems reasonable.
I love LibreOffice, but I wish there was an Android app. I’ve even considered learning more app development to try and help, but it’s such a daunting task.
I’ve been writing all my college papers in LaTeX and it’s been great. They look so professional, and it’s easier to work on a collection of text files than one monolithic document.
People say this, but almost every time the time interval is left off it’s hours.
Either way, the numbers in this meme are clearly made up. Most image generation uses fewer than 10 watt hours.
So like the size of a horse?
The average horse is about half the height and weight of the average giraffe. Giraffes are just a really bad unit of measurement, males weight about 400kg more than females and there is a wide height difference over their global population, they are technically four different species we just all call giraffe 🦒
And the coast guard was there too.
P cores give them better single core performance. But in parallel computing AMD has the advantage and has defended it for a long time now.
Why would you use a large language model to examine a biopsy?
These should be specialized models trained off structured data sets, not the unbridled chaos of an LLM. They’re both called “AI”, but they’re wildly different technologies.
It’s like criticizing a doctor for relying on an air conditioner to keep samples cool when I fact they used a freezer, simply because the mechanism of refrigeration is similar.
Yes they did. It says so in the article.
Yes, but also because they’re just better chips and you probably should have only been getting them to begin with. Way more power efficient, smaller process, less heat, easier to upgrade, better multi core performance, lower price; you just get a better CPU.
Reading comprehension is a skill. It seems like some people just glance at the words and make up what they expect to see. I have no idea how your comment could have been interpreted the way they did without a total disregard for the substance. Maybe they were reading it through an auto translation? Still, seems like the emoji is pretty obvious.
deleted by creator
Dang, what a mean cow.
Yeah, banning it is wild. I could understand setting it to opt-in, like a lot of websites do for adult content, but removing it entirely is just backwards and hateful.
True. Though in what tank vs tank combat there was, the advantages of modern armor were stark.
I was talking about the Gulf War in the 90s: https://youtu.be/b5EeKsEFpHI
I think the Iraqi tanks were mostly blown up by the time Bush Jr did his invasion.
Mixture of experts has been in use since 1991, and it’s essentially just a way to split up the same process as a dense model.
Tanks are an odd comparison, because not only have they changed radically since WW2, to the point that many crew positions have been entirely automated, but also because the role of tanks in modern combat has been radically altered since then (e.g. by the proliferation of drone warfare). They just look sort of similar because of basic geometry.
Consider the current crop of LLMs as the armor that was deployed in WW1, we can see the promise and potential, but it has not yet been fully realized. If you tried to match a WW1 tank against a WW2 tank it would be no contest, and modern armor could destroy both of them with pinpoint accuracy while moving full speed over rough terrain outside of radar range (e.g. what happened in the invasion of Iraq).
It will take many generational leaps across many diverse technologies to get from where we are now to realizing the full potential of large language models, and we can’t get there through simple linear progression any more than tanks could just keep adding thicker armor and bigger guns, it requires new technologies.
The gains in AI have been almost entirely in compute power and training, and those gains have run into powerful diminishing returns. At the core it’s all still running the same Markov chains as the machine learning experiments from the dawn of computing; the math is over a hundred years old and basically unchanged.
For us to see another leap in progress we’ll need to pioneer new calculations and formulate different types of thought, then find a way to integrate that with large transformer networks.
We’re just talking about the filename, the exact creation time is tracked by the OS. Plus I’d imagine most documents also have a time and date inside. The file name is mostly for sorting and human readability.
The national guard of the 1600s was organized around fighting american indians to assist with the colonization of North America.
The modern national guard wasn’t formed until the Dick Act of 1903, after which they were almost immediately used to suppress labor strikes. You can look into the Colorado Labor Wars or the Ludlow Massacre for some contemporary examples.