Alt. Profile @Th4tGuyII
To be fair, there’s a point to be made that someone who’s overly trigger-happy on dislike should be shamed for it. Just like you would be if you kept being snide to everyone in real life.
I agree that transparency would do much more good than harm, plus compared to the info that people already put in their profiles/comments, it’s not likely to make them anymore identifiable.
Votes should absolutely be public. They were on KBin, and it made people more civil for it because you could be shamed if you were dislike trolling or liking all of your own posts/comments to make them look better (which is something you actively have to do on here, unlike Reddit).
Given this place is pseudo-anonymous anyways, and people comment far more personal and identifiable info here anyways (which tbf you should be careful about), I think public votes would do much more good than harm.
Because that’s comparing oranges to apples.
In terms of pure image quality, real objects would win every time because they only have to be filtered by our eyes - digital images are filtered through the GPU and screen before ever reaching our eyes.
As such, the real contest is the ability of displays to make digital images look comparable to those real objects - because that’s harder to do vs. ust looking at the real life object, it’s more impressive to us.
So providing a fine-tuned model shouldn’t either.
I didn’t mean in terms of providing. I meant that if someone provided a base model, someone took that and but on of it, then used it for a harmful purpose - of course the person modified it should be liable, not the base provider.
It’s like if someone took a version of Linux, modified it, then used that modified version for a similar person - you wouldn’t go after the person who made the unmodified version.
SB 1047 is a California state bill that would make large AI model providers – such as Meta, OpenAI, Anthropic, and Mistral – liable for the potentially catastrophic dangers of their AI systems.
Now this sounds like a complicated debate - but it seems to me like everyone against this bill are people who would benefit monetarily from not having to deal with the safety aspect of AI, and that does sound suspicious to me.
Another technical piece of this bill relates to open-source AI models. […] There’s a caveat that if a developer spends more than 25% of the cost to train Llama 3 on fine-tuning, that developer is now responsible. That said, opponents of the bill still find this unfair and not the right approach.
In regards to the open source models, while it makes sense that if a developer takes the model and does a significant portion of the fine tuning, they should be liable for the result of that…
But should the main developer still be liable if a bad actor does less than 25% fine tuning and uses exploits in the base model?
One could argue that developers should be trying to examine their black-boxes for vunerabilities, rather than shrugging and saying it can’t be done then demanding they not be held liable.
Win11 becomes a less and less appealing switch day by day… When I can no longer hold into Win10, I think I’ll just have to jump ship to Linux.
Win10 is already quite privacy poor, but Win11 is straight up intolerable.
The Internet Archive is currently fighting in the courts to maintain free digital library access to over 500,000 books they own from their own collection, yet Meta uses a pirated dataset of nearly 200,000 books to train their proprietary AI and is just allowed to get away with that??
Publishers will go after a charity making fair use of their content, but not the corporation outright stealing from them. What utter bollocks.
The TL;DR for the article is that the headline isn’t exactly true. At this moment in time their PPU can potentially double a CPU’s performance - the 100x claim comes with the caveat of “further software optimisation”.
Tbh, I’m sceptical of the caveat. It feels like me telling someone I can only draw a stickman right now, but I could paint the Mona Lisa with some training.
Of course that could happen, but it’s not very likely to - so I’ll believe it when I see it.
Having said that they’re not wrong about CPU bottlenecks and the slowed rate of CPU performance improvements - so a doubling of performance would be huge in this current market.
Just having read the README, even if it’s a little faffy, the idea of being able to converse with someone and have it be translated back in near real-time is awesome!
This is the kind of application I was looking forward to seeing come out of open source AI - I think it’d still be a little too clunky to save you in an emergency, but probably less clunky for most people than learning a new language or having a human translator on retainer haha
Hugging and play fighting, sure, but can’t say I’ve snuggled with the lads (or girl friends either for that matter) - feels more intimate, like something you’d do with a loved one (or a pet).
Drunk me might lean a bit more towards leaning or pushing against the lads, but that’s always playful and jokesy rather than sincere as your post implies
If you have access to the BIOS, you could always just underclock and undervolt the system to give more power efficiency.
Though a small server with a retroconsole attached is also not a bad idea for the excess juice.
True, but if you’re already replacing the mobo (say for an upgrade, which Framework laptops are designed to let you do), then you can use your old one for this rather than having to buy a Pi clone or some such
That’s why I specified the community, as in the more tech savy folks that would care about this, because I know that the wider public is surprisingly tech illiterate
I’ll be curious to see where this ends up going, as I doubt the community will take this lying down.
The few times I’ve had to go without an Ad blocker, I’ve seen just how bad the Ads have gotten - they’re almost the same as regular TV Ad breaks now! … And then YouTube Premium is just not a good deal in my eyes, £12.99 a month is an awful lot to pay just to not see Ads.
Yeah. If you’re on a public forum accessible to anyone, which the whole fediverse is, then you should never assume privacy.
Honestly transparency in this regard would be better - they’re already visible to much of the community, so they might as well be visible to everyone.