Zuiii

25 ms·

301.
by Zuiii 4y ago
Like it or not, they do offer a compelling argument though I personally don't care either way. What I'd care more about is whether this issue had harmed the project in a technical way, which, unfortunately, it apparently has.
302.
by Zuiii 4y ago
I'd also add: - People living abroad who are completely debanked because local banks refuse to open banks accounts for people who are merely suspected of being US citizns, despite them not being US citizns, with no path to resolve thei
303.
by Zuiii 4y ago
Break them up! The potential for manufacturers to abuse their customers when they control the full stack. Belllabs both Google and Apple. Hardware and software should developed by separate entities anyway. If it works for normal computers,
304.
by Zuiii 4y ago
Thank you. I previously relied exclusively on samsung to for phones, tablets, and smartwatches but decided to stop dealing with them after they decided to silently enable sending everything I typed to grammarly in an update. (Literally 2 de
305.
by Zuiii 4y ago
The US is not in any position to start a shooting war with anyone that has the ability to push this field forward and come out of it intact. AI training will continue and Yudkowsky will be proven wrong.
306.
by Zuiii 4y ago
They absolutely did not spend millions to train it. Credible estimates place the cost for an entity like Meta at about 30-100k, probably less since Meta likely owns the 256 A100x8s needed to train it. Even as an individual, it wouldn't
307.
by Zuiii 4y ago
LLaMA weights are effectively public domain data according to the criteria outlined in statement issued by the US copyright office two weeks ago [1]. Those weights are an original work that was created by a mechanical process that involved
308.
by Zuiii 4y ago
Unless you sourced the training dataset and trained it on that model, the US copyright office disagrees: https://www.federalregister.gov/documents/2023/03/16/2023-05...
309.
by Zuiii 4y ago
> The unfiltered model is completely amoral and will do its best to give you the answer you requested. Like how cars do not object to driving over the speed limit, or wikipedia permitting you to research explosives, this is how the model
310.
by Zuiii 4y ago
> who thought splitting ownership and consequences of not updating software was a good idea? You can't because companies refuse to let customers own and control their devices. Because we can't control them, we can't patch
311.
by Zuiii 4y ago
Seeing the geopolitical condition of those who have them and those who don't, I can't say that idea is without merit.
312.
by Zuiii 4y ago
Sure but it wont be completely one-sided.
313.
by Zuiii 4y ago
The ads typically shown this way rarely related to security. It's usually something that Microsoft wants the user to "agree" to. It's basically physiological coercion.
314.
by Zuiii 4y ago
The code to train and run the model perhaps, but model weights lack two fundamental requirements available in normal code: human authorship and creativity. Because these models were the result of a purely mechanical process and thus lack th
315.
by Zuiii 4y ago
While they'll release useless minor models here and there, I think these companies will keep this tech behind closed doors regardless of what the community does. Humanity needs to act quick in order to avoid yet another extreme consoli
316.
by Zuiii 4y ago
The position of the US copyright office is based in large part on prior case law (from your first link).
317.
by Zuiii 4y ago
This was my main question when I read the description. Fortunately, the authors released the data set so it should be a simple matter of grep -VF "but as an AI language model".
318.
by Zuiii 4y ago
This statement nicely describes my experience with LLMs. They may have a hard time staying on topic (especially with larger untuned models like LLaMA 13B+), but if you help them stay on track, they become very useful.
319.
by Zuiii 4y ago
> "Model weights aren't part of the release for now, to respect OpenAI TOS and LLaMA license." Makes sense if they originally licensed the model weights from Meta. Fortunately you can get the weights via torrent without a
320.
by Zuiii 4y ago
lol what are you talking about? The weights are being linked to on Meta's repo right now: https://github.com/facebookresearch/llama/pull/73/files The idea that Meta is frantically running around try
321.
by Zuiii 4y ago
Did he fine tune the model or was all the required information contained in the foundational LLaMA model itself? If he did, and he fine-tuned it on an exploits database then I can see how the model could be used this way. This is a good thi
322.
by Zuiii 4y ago
Yeah, I'm sure microsoft (the entity showing the ads) has some presence in Australia. Or better yet, sue microsoft proper, get a judgement, and secure and injunction until they comply. If all goes well, the entire government will be fo
323.
by Zuiii 4y ago
What do windows developers do now when they need to test multiple domains on their local machine?
324.
by Zuiii 4y ago
What are the other ones?
325.
by Zuiii 4y ago
Sure you can but what about cmdlets? Do we have an open source "coreutil" that offers reasonable compatibility with the standard powershell environment?
326.
by Zuiii 4y ago
> avoiding to fund the organizations responsible is a moral imperative I never thought about it that way. I avoided getty and other stock media publishers out of principle. In their deranged quest to subjugate google, they made image sea
327.
by Zuiii 4y ago
> You might be a rarity here. And I might not be, hence my call for actual metrics.
328.
by Zuiii 4y ago
> Seems gone. I can't say I've personally ever used it I rely on it heavily and I'm sure many others do to. I wish framework included the key in their study as well. Ether way, no menu key == no buy from me.
329.
by Zuiii 4y ago
Why do companies insist on messing with laptop keyboard layouts? Where's the menu key? Where are the Home, PgDn, PgUp, and End keys? I've used cheap sub 300-dollar laptops with keyboards more complete than this. The silver lining
330.
by Zuiii 4y ago
Carious timing given that the US copyright office itself does not consider these weights to be copyrightable per its statements just a few days ago. I hope someone takes Facebook to court.

More ›