Hacker News Reader

Comment by ot | Hacker News Reader

In case it's not clear what's happening here (and from the comments it doesn't seem like it is), someone (not Meta) leaked the models and had the brilliant idea of advertising the magnet link through a GitHub pull request. The part about saving bandwidth is a joke. Meta employees may have not noticed or are still figuring out how to react, so the PR is still up.

(Disclaimer: I work at Meta, but have no relationship with the team that owns the models and have no internal information on this)

View on HN

> Meta employees may have not noticed or are still figuring out how to react

Given that the cat is out of the bag, if I were them, I would say that it is now publicly downloadable under the terms listed in the form. It is great PR, which if this was unintentional, is a positive outcome out of a bad situation.

Facebook fumbling it's way into being the better open source AI than OpenAI would be amusing.

Considering the opt-models that they've previously released publicly, intentionally, I'd say that facebook is better open source AI than OpenAI even without fumbling.

How likely is it that there is a larger model that they haven't discussed?

It's not even clear someone has leaked the models. A random person has put a download link on a PR, it could be anything.

Yes you're absolutely right. I went by another comment that seemed to confirm the contents, but that could be trolling too.

So do I understand correctly — those that tell don’t know, and those that know don’t tell?

I had to look up the legality of downloading torrents of copyrighted IP - and let me just say, don’t do it.

You had to look this up? Who verified it for you?

HN user (with >3k) karma seems to confirm the leak. Take it for what it's worth.

The folder structure definitely looks like model weights, I didn't download or run it though so for all I know it only generates the words to "Never Gonna Give You Up".

Not going to lie but an LLM that generates Fresh Prince pasta sounds like one of the most amazing things ever.

It's fairly easy to obtain the weights. I've had two of my friends downloading these weights and sharing them with me, so it's probably not surprising that the weights got leaked.

Was every Meta employee able to download them or why did so many people have access?

Anyone can submit a request form and enter their email address to request access to the weights. People who have a .edu email and are involved in deep learning are likely to succeed and get a download link sent to their email.

> and have no internal information on this

But you just said it was not Meta. Is that based on internal information? ;-)

As I explained in another comment, if the author was a Meta employee, they wouldn't have had to sign the CLA. It's all in the PR.

I wondered why I didn't see that comment of yours. Turns out the comment you replied to got flagged.

someone (not Meta)

Did you mean to write "someone (possibly Meta)" ?

Reply on news.ycombinator.com