ImprobableTruth

14 ms·

31.
by ImprobableTruth 2y ago
No, that's a different scenario. In the one I gave there's explicitly a dependency between requests. If you use gather, the network requests would be executed in parallel. If you have dependencies they're sequential by nature
32.
by ImprobableTruth 2y ago
The reason is that the usage is completely different from coroutine based async. With GPUs you want to queue _as many async operations as possible_ and only then synchronize. That is, you would have a program like this (pseudocode): b =
33.
by ImprobableTruth 2y ago
On their subscriptions, specifically the pro subscription, because it's a flatrate to their most expensive model. The API prices are all much more expensive. It's unclear whether they're losing money on the normal subscriptio
34.
by ImprobableTruth 2y ago
If you compare with e.g. Deepseek and other hosters, you'll find that OpenAI is actually almost certainly charging very high margins (Deepseek has an 80% profit margin and they're 10x cheaper than openai). The training/R&
35.
by ImprobableTruth 2y ago
Do you take issue with the 'purely empirical' approach (just trying out variants and seeing which sticks) or only with its insufficient documentation? I don't know how you'd improve on the former. For a lot of it there s
36.
by ImprobableTruth 2y ago
There are even much cheaper services that host it for only slightly more than deepseek itself [1]. I'm now very certain that deepseek is not offering the API at a loss, so either OpenAI has absurd margins or their model is much more ex
37.
by ImprobableTruth 2y ago
The "97.3%" match is probably just the confidence value - I don't think a frequentist interpretation makes sense for this. I'm not an expert in face recognition, but these systems are very accurate, typically like >99
38.
by ImprobableTruth 2y ago
Yes, that's how it works. I think in Zig for new types you'd use enums for ints and packed structs for more complex types.
39.
by ImprobableTruth 2y ago
Latency is magnitudes more critical when it's something where you have to react.
40.
by ImprobableTruth 3y ago
Transmission speeds aren't fast enough for this, unless you crank up the batch size ridiculously high.
41.
by ImprobableTruth 3y ago
Really sad to see. Emad pivoting to posting crypto nonsense on twitter made me think the writing is on the wall for Stability, but I still didn't expect it so soon. I expect they'll pivot to closed models and then fade away.
42.
by ImprobableTruth 3y ago
MS vs Sun was about MS doing their usual EEE thing and adding parts to MS Java that weren't part of the spec (which is why it was also a trademark lawsuit). Google vs Oracle is about whether APIs are copyrightable and Google won.
43.
by ImprobableTruth 3y ago
IME NVLink would be overkill for this. Model parallelism means you only need bandwidth to transfer the intermediate activations (/gradients + optimizer state) at the seams and inference speed is generally slow enough that even pcie x8
44.
by ImprobableTruth 3y ago
48 gb suffice for 4-bit inference and q-lora training of a 70b model. ~80 GB allows you to push it to 8-bit (which is nice of course), but full precision finetuning is completely out of reach either way. Though you're right of course t
45.
by ImprobableTruth 3y ago
Calling him "the person hired to build Stable Audio" seems a bit misleading? He was in a executive position (VP of product for Stability's audio group). An important position, but "person hired to build" to me evoke
46.
by ImprobableTruth 3y ago
The first post - which I responded to - is _explicitly_ about the government paying people to do art: > Games in this day and age ideally need generous government subsidy to be successful. > something like an artist's living wage
47.
by ImprobableTruth 3y ago
The issue isn't that it's not real-time/sped-up, it's that it doesn't actually take video as input, but multiple hand picked stills.
48.
by ImprobableTruth 3y ago
Because we don't live in a world with UBI, so we have to decide how to allocate finite resources. The alternative to using some arbitrary standard of commercial success is some other arbitrary standard (unless you introduce UBI). If we
49.
by ImprobableTruth 3y ago
Unless you go for UBI, art subsidies will be finite. You need a measure to decide who gets subsidized and who doesn't. We're not talking about people doing art as a hobby after all, but about the government paying people to make a
50.
by ImprobableTruth 3y ago
Commercially successful art is that which is "popular enough". The value of art is completely subjective, so what other measure do we have? If there's "bad art" that appeals to you and others enough, then it isn&#x2
51.
by ImprobableTruth 3y ago
Because what other good measure is there? You mention great artists who died destitute - but we consider them great because they are posthumously popular. We can't look into the future to see what stands the test of time, so current po
52.
by ImprobableTruth 3y ago
Why would you want it to be divorced from success? Seems like popularity/commercial success is the best measure of "being good" societally that we have for art. People making art/games only for themselves is fine of cour
53.
by ImprobableTruth 3y ago
I've definitely seen edge computing used in the context of IoT to refer to compute done on sensing devices. The less narrow meaning at least isn't really "new".
54.
by ImprobableTruth 3y ago
(I assume this question is about whether models need all those layers during training, even if they don't need them during inference) Yes. There's the so called "lottery ticket hypothesis". Essentially the idea is that l
55.
by ImprobableTruth 3y ago
Why would this only apply to open source models? Or are you assuming that a company like OAI could license all the material?
56.
by ImprobableTruth 3y ago
And what are these "better systems"?
57.
by ImprobableTruth 3y ago
Voices are uncopyrightable, but impersonation isn't legal (see Midler v. Ford, for a notable case), so I don't think the situation is totally clear.
58.
by ImprobableTruth 3y ago
Records are the way to go, but there are many situations where the ecosystem doesn't work with them e.g. if you're stuck using something like Hibernate.
59.
by ImprobableTruth 3y ago
Because as is C# is very noticeably a second class citizen. Something like Godot 4.0/4.1 missing C# support for android/ios/web shows that perfectly imo.
60.
by ImprobableTruth 3y ago
Take the state monad : state -> (a, state). I think one would only count that as a wrapper by squinting really, really hard.

More ›