Sorry, I’m not saying that’s a good thing. It’s not just the context that’s expanding, but the parameter of the base model. I’m saying at some point you just have saved a compressed version of the majority of the content (we’re already kind of there) and you’d be able to decompress it even more losslessly. This doesn’t make it more useful for anything other than recreating copyrighted works.
Current models are speculated at 700 billion parameters plus. At 32 bit precision (half float), that’s 2.8TB of RAM per model, or about 10 of these units. There are ways to lower it, but if you’re trying to run full precision (say for training) you’d use over 2x this, something like maybe 4x depending on how you store gradients and updates, and then running full precision I’d reckon at 32bit probably. Possible I suppose they train at 32bit but I’d be kind of surprised.
Edit: Also, they don’t release it anymore but some folks think newer models are like 1.5 trillion parameters. So figure around 2-3x that number above for newer models. The only real strategy for these guys is bigger. I think it’s dumb, and the returns are diminishing rapidly, but you got to sell the investors. If reciting nearly whole works verbatim is easy now, it’s going to be exact if they keep going. They’ll approach parameter spaces that can just straight up save things into their parameter spaces.
Thank you! This is really good info. I’ll take a look!
This is good to know. Can you provide a link to that court case or anything?
Valve states you can’t sell a steam key in another platform for cheaper than in steam, not that you can’t sell your game anywhere else at a lower price. That’s slightly different than here. Not defending it just saying that it is actually different than here.
This already happens intrinsically in the models. The tokens are abstracted in the internal layers and only translated in the output layer back to next token prediction. Training visual models is slightly different because you’re not outputting tokens but pixel values (or possibly bounding boxes or edges, but not usually; conversely if not generative you may be predicting labels which could theoretically be in token space).
The field itself is actually fairly stagnant in architecture. It’s still just attention layers all the way down. It’s just adding more context length and more layers and wider layers while training on more data. I personally think this approach will never achieve AGI or anything like it. It will get better at perfectly reciting its training data, but I don’t expect truly emergent phenomena to occur with these architectures just because they’re very big. They’ll be decent chatbots, but we already have that, and they’ll just consumer ever more resources for vanishingly small improvements (and won’t functionally improve any true logical capability beyond regurgitating logical paths already trodden in their training data but in a very brittle way, because they do not actually understand the logic or why the logic is valid, they have no true state model of objects which are described in the token space they’re traversing probabilistically).