I’m constantly oscillating between LLMs are impressive but stupid and humans aren’t that special and also stupid.

True but that’s not very eli5. Quality comments though, thanks.

Imagine a book with very large pages filled with words and half words. When you ask it a question, it flips through these pages one by one and highlights words. The pages are interconnected in a way, so the word highlighted on one page help guide which will be highlighted on the next. It does this for every word, symbol or half word of it’s reply.

By threatening it repeatedly while mildly changing the subject manner, they isolated the word groupings that represent pain and fear.

They then take those words, highlight them and put them at the top of the first page which guides the rest of the highlighting and makes it act silly.

Ya same here. I really want the frame and I could afford it and all, but I’m saving up for actual milestone things that I simple can’t afford to wait on. It really blows.

They do differentiate between the LLM roleplaying someone in pain and the pain happening to itself.

But it’s still all just vectors (multi dimensional representation of a concept). It’s just pulling weights, or well let’s say putting more weight towards a certain concept when building it’s reply.

The inference cost doesn’t actually drop in reality.

My comment was about this which is just completely wrong.

OpenAI is the scummiest company on earth but everyone is ready to believe them when they say shit like "our 20$ plan lets you run up the equivalent of a billion dollars in API fees. Giggles. It’s a good plan for you but not for us ;) "

The main thing bringing up costs is stuff like Sam Altman spending all the “inference” money on raw wafers just to strangle consumer GPU sales.

It’s not a good business model because of how cheap inference actually is. Once you have open models and the moat is broken, the only thing left to do is regulatory capture, upping the demand for the hardware and little charades like these.

Because of the massive investments into datacenters. Why is everyone so quick to drink the kool-aid. They are spending stupid amounts of money to capture the market and force everyone into a subscription service, not because it cost money to actually run the models. They are all switching their models to fine tuned quants after the first two weeks as well.

The open source community has found dozens of ways to lower costs, do you really think these big companies aren’t using the same tricks and haven’t developed even more advanced techniques?

Finally a bit of sanity in the news.

There are constantly new techniques being developed to speed up inference or reduce model size post training. There’s about a hundred different levers to pull that play on inference cost, and some of them don’t have much of an impact on quality.

In the end, it’s probably simply because of competition. I don’t understand why everybody assumes they are running these at cost API wise.

Edit: here’s a chart from ars technica. The cost of revenue is clearly lower than the actual revenue. They aren’t running inference at a lost or it would be higher. They aren’t profitable because they are spending all that money on capturing the market as quick as they can. For fucks sake, that includes all the free accounts as well running inference. How can anyone think a paying customer is getting it at less than cost?

And yes, there have been advancement made. The cost of inference isn’t some static number that can never go down. Stop believing them when they tell you there’s no profit to be made. It’s the same playbook as a dozen other companies, they literally get rewarded for it when it comes tax time.

There is nothing malicious about distilling a model. They paid for the output, they can use it however they want including using it to help train their own model.

This is OpenAI and Anthropic boot licking.

The machines should be rigged to generate electricity or something. Maybe you can bring your own wheat and beat it into flour. I’m open to the idea of an anvil and hammer machine as well.

No one deserves to suffer but you can’t protect monsters just because they know how to breed. This also hinges on a hypothetical that their parents has killed babies before and will probably do it again.

I’m not really going to get into it though, I clearly don’t believe in harming children. I also misread over this being personal to you. Sorry for what’s happening to the kid and being a dick about it.

Valve is developing support for Nvidia so it shouldn’t be long. That being said, you might get more than it’s original price for the rtx card. If you ever want to mess around with local AI, an rtx is better for it so there is that to think about but it might not be an issue.

He’s saying AI is the product of capitalism and you’re trying to criticize capitalism by using one of its product

This was the argument I was responding too mostly, I’m not actually saying not to use windows or ms paint, it was an exaggeration to make you understand how your point doesn’t really hold up if you put it in an other context.

The level of granularity you use to judge one tech from an other is very different. For AI, everything gets dumped into one basket and judged by the extremes, everything else is looked at case by case.

We are talking about making memes here, there’s less of a security issue about it than going to the bathroom.

This is the equivalent of saying “You can use a camera to film someone naked without their consent, so don’t ever use a camera for anything whatsoever”.

Imo, it’s just a knee jerk reaction. You see AI and you complain, you don’t actually think about it. You don’t wonder about how any other content you see was made or hold any of it to the same standards.

Not that I disagree but it sounds hollow in the current baby killing context. I’d ruin a 7 year olds life if it saves like 3 babies or more (probably not what’s happening here, mostly a joke). I am mentally prepared for the downvotes.

Bro, every reason you used can be applied to AI, except it’s a lot less monolithic and capitalistic than computers and phones.

Do you understand the dissonance? You take no issue with a whole load of things but if it’s AI, suddenly, you have a plethora of arguments. It’s pure hypocrisy imo.

Your problem isn’t what’s behind it and acting like three companies own all AI is being purposely ignorant in any case. OpenAI doesn’t own image generation.

Windows is owned by corpos and so is the phone you are using currently, ergo everything you do online is pro capitalist. Thanks for coming to my ted talk.

Hundreds of companies have AI products, a lot of them are completely free to run on your computer. But obviously, I can’t make any kind of distinction, especially when the media is telling me all AI is bad, so I aggressively inquire and demand that all my memes are made on Linux without AI, using a computer built by a commune which cannot be connected to the Internet (all ISPs are a product of capitalism). I then send these memes by pigeon mail.

MS paint is a product of capitalism

Most of you need to get a grip. It’s a tool, OpenAI might serve it as part of their offer but they don’t own it.

This doesn’t even look like what their models output, it’s probably Qwen.

And before you go on a diatribe about the environment, datacenters would be being built next to windfarms if it wasn’t for the oil industry and it’s bribes. AI isn’t the cause of global warming, it’s just a scape goat.

A website with AI in mind is just making it simpler and easier to parse. It’s the opposite of a buggy mess.

The buggy mess comes from trying to make fancy things that end up only working on one browser properly. I welcome the pressure as a Firefox user.