Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Facebook LLAMA is being openly distributed via torrents

github.com

Post reply on HN
Print view View original

331–340 of 719 posts

Re: Facebook LLAMA is being openly distributed via torrents

#331

Post by im3w1l »

narrator wrote: post #151
Tiberium wrote: post #99

It seems that the leak originated from 4chan [1]. Two people in the same thread had access to the weights and verified that their hashes match [2][3] to make sure that the model isn't watermarked. However, the leaker made a mistake of adding the original download script which had his unique download URL to the torrent [4], so Meta can easily find them if they want to. [1]: https://boards.4channel.org/g/thread/9184826…

It's funny that part of the 4chan excitement over this is that they think they'll get back the AI girlfriend experience of when character.ai was hooked up to uncensored GPT-3. All that has been thoroughly shut down by character.ai and Replika and they just want their girlfriends back.

I think it's funny that out of all the scifis I know, Chobits of all things is looking to be the most accurate.

Re: Facebook LLAMA is being openly distributed via torrents

#332

Post by KMag »

shadowgovt wrote: post #276
bentcorner wrote: post #224

Earlier quoted context omitted.

I posit to a friend that: a) As these AI constructs become more advanced (especially around memory and personalization), we will eventually be able to treat them as people b) Some business will eventually sell an off-the-shelf product (hardware and/or software) that is an AI you can bring into your home, that you can treat as a friend, confidant and partner c) Someone will eventually lose their AI friend of many mont…

At the end of the day, the Turing Test for establishment of AI personhood is weak for two reasons. 1. We're seeing more and more systems that get very close to passing the Turing Test but fundamentally don't register to people as "People." When I was younger and learned of Searle's Chinese Room argument, I naively assumed it wasn't a thought experiment we would literally build in my lifetime. 2. Humanity has a histor…

On the flip side of subhuman treatment of humans, we have useful legal fictions like corporate personhood. It's going to be pretty rough for a while, particularly for nontechnical judges, to sort all of this out.

We're almost definitely going to see multiple rulings far more bizarre than Citizens United ruling that limiting corporate donations limits the free-speech rights of the corporation as a person.

I'm not a lawyer, and I don't particularly follow court rulings, but it seems pretty obvious we need to buckle up for a wild ride.

Re: Facebook LLAMA is being openly distributed via torrents

#333

Post by jakear »

DharmaPolice wrote: post #228
bentcorner wrote: post #224

Earlier quoted context omitted.

I posit to a friend that: a) As these AI constructs become more advanced (especially around memory and personalization), we will eventually be able to treat them as people b) Some business will eventually sell an off-the-shelf product (hardware and/or software) that is an AI you can bring into your home, that you can treat as a friend, confidant and partner c) Someone will eventually lose their AI friend of many mont…

It became something of a meme but there are huge numbers of guys out there that would pay good money for Joi from Blade Runner 2049. https://bladerunner.fandom.com/wiki/Joi

That has more to do with Ana de Armas looking how she does than anything else. I'd have dated her as a Harlan Thrombey's nurse too.

Re: Facebook LLAMA is being openly distributed via torrents

#335

Post by randomdestructn »

bentcorner wrote: post #242
JKCalhoun wrote: post #234

Earlier quoted context omitted.

> b) Some business will eventually sell an off-the-shelf product And by sell you mean a monthly subscription, ha ha.

Yeah, that's probably the most dystopian thing. This is almost a guaranteed outcome - someone pays a high subscription cost and cultivates a model with their personal details for years, and then loses all of it when they can't keep up the subscription cost. Cue a month or two later - they buy back in and their model has been wiped and their AI friend now knows nothing about them. It's easy to poke fun at people who u…

Or maybe they sell that data to another company that operates kind of like a collections agency, which takes on the 'risk' of storing the data, then repeatedly calls and offers to give them their AI friend back at an extortionate rate.

The data privacy side of this is an interesting conversation as well. Think of the information an employee or hacker could leak about a person after they spent some time with such an instance.

Re: Facebook LLAMA is being openly distributed via torrents

#336

Post by adam_arthur »

pavlov wrote: post #41

It’s interesting that these models are both massively expensive to produce and self-contained to a degree that you can distribute the end product in a torrent. This has not been the case for most commercial software for the past 20 years, during the cloud era. If you could steal a dump of random Facebook source code, it would be 99% useless because it’s so closely tied to the infrastructure. There’s almost nothing yo…

Which is why it will be very difficult to monetize them with similar margins to SaaS era businesses.

The bar to competition is far lower, as already evidenced by the plethora of AI products being put forward. Its a race to the bottom on pricing

Re: Facebook LLAMA is being openly distributed via torrents

#337

Post by EamonnMR »

retinaros wrote: post #302
colinmorelli wrote: post #88

Earlier quoted context omitted.

I am quite confident that using this model for commercial purposes will, if detected, land you in quite a legal quagmire that almost certainly sides in favor of Meta. And even if it did not, Meta certainly has a more capable legal team with more cash to spend than the average HN user.

didnt they train this on copyrighted content?

They might have trained it on everyone's DMs.

Re: Facebook LLAMA is being openly distributed via torrents

#338

Post by EamonnMR »

sebzim4500 wrote: post #24

Is there anything stopping anyone from using this for commercial purposes? I know that when you fill in the google form you need to agree to noncommercial use, but someone downloading this will never have agreed to that licence agreement.

Considering ML's tenuous relationship with IP, I can't help but find this situation amusing.

Re: Facebook LLAMA is being openly distributed via torrents

#339

Post by jandrese »

JKCalhoun wrote: post #234
bentcorner wrote: post #224

Earlier quoted context omitted.

I posit to a friend that: a) As these AI constructs become more advanced (especially around memory and personalization), we will eventually be able to treat them as people b) Some business will eventually sell an off-the-shelf product (hardware and/or software) that is an AI you can bring into your home, that you can treat as a friend, confidant and partner c) Someone will eventually lose their AI friend of many mont…

> b) Some business will eventually sell an off-the-shelf product And by sell you mean a monthly subscription, ha ha.

I am reminded of a virtual girlfriend service that used to exist in Japan where you could buy virtual clothes and other presents for your virtual girlfriend using real life money. The more you spent on her the friendlier she was. I think it was all on the phone, although my memory of the articles has become fuzzy over the years.

Re: Facebook LLAMA is being openly distributed via torrents

#340

Post by terafo »

joker99 wrote: post #318
underlines wrote: post #280

- how much vRAM needed to run each model parameter size? - any inference optimization we can use similar to StableDiffusion, to bring down the vRAM requirements? I only know about these: - use 8bit precision - https://github.com/bigscience-workshop/petals - https://github.com/FMInference/FlexGen - https://github.com/microsoft/DeepSpeed Anything that could bring this to a 10GB 3080 or 24GB 3090 without 60s/it per toke…

If I may tack on a question as someone with zero clue of ML: when, if ever, will someone like me be able to run this on a Mac Studio with a M1 Ultra and 128GB of ram?

As far as I can tell you can do it right now, at least for small 13B model, not sure about bigger models.
Post reply on HN