joshcryer

7 ms·

31.
by joshcryer 4y ago
Brave is using IPFS for file storage but once the content address (CID) is known anyone can access the file you're looking for. So it remains to be seen how they will leverage IPFS to create scarcity of digital items for their merchand
32.
by joshcryer 4y ago
Don't really know that, either. They said they didn't do an empirical analysis on it. For example, it may show a few male nurses for hundreds of prompts or it may show none for thousands. They don't give examples. Hopefully t
33.
by joshcryer 4y ago
Who does it serve for plebs to be shown the approach openly? I don't know that it does a disservice to anyone by showing the approach. But in general it is likely more due in part to the fact that it's going to happen anyway, if w
34.
by joshcryer 4y ago
It's just a different asterisk to distinguish, in this case, in the paper, they are "core contributors."
35.
by joshcryer 4y ago
The bulk of the trained data is from western technology, images, books, television, movies, photography, media. That's where the very real and recognized biases come from. They're the result of a gap in data nothing more. Look a
36.
by joshcryer 4y ago
There are people working on reproducing the models, see here for Dall-E 2 for example: https://github.com/lucidrains/DALLE2-pytorch It's often not worth it to decentralize the computation of the trained model thou
37.
by joshcryer 4y ago
They're withholding the API, code, and trained data because they don't want it to affect their corporate image. The good thing is they released their paper which will allow easy reproduction. T5-XXL looks on par with CLIP so we ma
38.
by joshcryer 4y ago
Gato is a model of models, if you read the paper they trained it on multiple datasets, and is actually what I was referring to in my comment.
39.
by joshcryer 4y ago
You can get a ton of old, abandoned, or unsupported, software from Archive.org. I have no idea if the original authors are aware of it but if they are they don't seem to care that much.
40.
by joshcryer 4y ago
Not your keys, not your coins.
41.
by joshcryer 4y ago
But every specialized model like this is getting us closer to "doing a zillion other things." By logic it is exactly one step closer. The general AI agent will be composed of many such models.
42.
by joshcryer 4y ago
Just document the fuck out of everything you did dude, no matter what you chose to do.
43.
by joshcryer 4y ago
You just told the world about a compromised site that has something to do with crypto and stocks. That was your first mistake. If you want to white hat this you should just contact the admin and mass mail everyone affected and wash your han
44.
by joshcryer 4y ago
What Gato shows is with the proper datasets and training we can achieve general purpose like results. It doesn't show any sort of learning behavior. So your personal home robot may be able to dice up an onion, but it is not going to
45.
by joshcryer 4y ago
This result is unsurprising. "Give a model a bunch of unique datasets and it can do a bunch of unique things." There's nothing showing any sort of generalized learning or capability here.
46.
by joshcryer 4y ago
I suspect it's more bland than that, they need to be able to have a booting/working system otherwise the customer is going to call up support and say their computer isn't booting or that it's saying "no OS found&quo
47.
by joshcryer 4y ago
Just to clarify your number, an average Bitcoin block (currently) has about 2000 transactions, each of which, be it a transaction for half a cent, or a billion dollars, cost around $3 each. What you described is what happened when Bitcoin w
48.
by joshcryer 4y ago
You can read the paper here: https://www.nature.com/articles/s41467-022-29252-1 I can see this technology being made into a super computer type setup one day, but as far as home computing, I have my doubts.
49.
by joshcryer 4y ago
Honestly, it's no wonder The Pile produced models have to be "bias adjusted." Please, pick, at random, any date on that page and you can say the same thing about any of the subject matters. Here. I just did it. 2003. Middlese
50.
by joshcryer 4y ago
Here's the video that likely got them the Pulitzer: https://www.youtube.com/watch?v=ibWJO02nNsY
51.
by joshcryer 4y ago
One of the draws to use Hugging Face is because the open source community is giving those trained models away. What they are doing is permitting people to use those models affordably (something like a million words would be $10 if I rec
52.
by joshcryer 4y ago
Hugging Face is selling CPU cycles. They're also letting you upload your own datasets that aren't "limited" like others. I'm not quite sure where you think their approach of "open models" is wrong, they s
53.
by joshcryer 4y ago
A lot of people come up with an idea and then have no product. And investors are just happy to throw money at them. It's that whole idea that if you want to start an AI company, hire a bunch of people in India to do your work, then onc
54.
by joshcryer 4y ago
Yeah, I know the argument, but how many people in industrial society know how to do anything? The engineers build the systems that the factory workers use. In fact, the workers generally know only one subset of some problem that is being so
55.
by joshcryer 4y ago
I sold my RX 580 for $400 (which I got for $120) during peak hysteria. eBay locked my account. I went through the process of explaining to support that I was selling it because gfx card prices were so high.
56.
by joshcryer 4y ago
The lovely thing about this "old knowledge" or "old ways" fetishism is that the knowledge economy is a superset of all previous paradigms. If you were to ask a blacksmith 400 years ago what he thought about people li
57.
by joshcryer 4y ago
The entire article hinges on the principle that everything collapses because there is "on-paper upward mobility of functional industrial workers to dysfunctional knowledge economy workers." Translation: the knowledge economy isn&#
58.
by joshcryer 4y ago
The entire ETH VM is running at about 0.8 MIPS right now. Yes. That is not an exaggeration, current gas limit is 30 million / 3 gas for simpliest add instruction / 12 seconds per block. The calculation power of the network is abso
59.
by joshcryer 4y ago
The gas fees don't really correlate to any energy usage. They're actually a fee for computation on the block. It's no different than the Bitcoin miner fee, except it's based upon how much code is being included and the c
60.
by joshcryer 4y ago
Yikes, that Twitter feed: "Thoughts and prayers for the deep learning fanboys," as if ones admiration for DALL-E 2's achievement is something to be mocked... I feel like this sentiment is the way we're going to get I Hav

More ›