Rendered at 19:09:31 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
ilaksh 18 hours ago [-]
Those are amazing accomplishments but I am more interested in research developments in things like In-Memory (Analog) or other different approaches.
Companies like EnCharge, Mythic, etc.
And much more efficient devices like RRAM, MRAM, and FETs. Like FE-FETs with AlScN.
The stuff just coming out of research or still in research is more exciting in terms of the potential for truly huge efficiency and performance boosts.
Taalas is interesting also because of it's efficiency and speed. Guess it was just purchased by AMD.
skew-aberration 11 hours ago [-]
I'm not sure that more efficient compute can win out, since it also has to beat out the competition on capital investment. There's an upper bound of 'just get 3 sets of GPUs and run them on solar in three different timezones' where the cost of electricity is negligible. Efficient digital architectures run about 1/5 or 1/10th the speed and require much more wafer area. Analog technologies seem unlikely to have the fidelity required for training, though inference sounds plausible. It's also not clear that analog technologies would necessarily use less power.
ilaksh 1 hours ago [-]
Some technologies will definitely be much faster and much more power efficient. They haven't come out of research or if they have they have not been scaled yet, so it's not 100% proven. But it looks likely for multiple options to get there.
alain94040 3 hours ago [-]
Agreed. One of the least efficient things to do with digital logic is floating point multiplication. The second least efficient is probably integer multiply (I'm just kidding a bit here).
Multiplication may make sense for the math, but it's just not practical in hardware. There has to be a better, much simpler operation, that can give you equivalent results. Yes, you may have to re-think and re-train everything, but eventually it should make current architectures obsolete.
Is anyone working on running neural networks on CPU architectures that are not limited by synchronous clock signals? Organic neural networks are inherently asynchronous and signals propagate through different parts of the network at their own pace.
jpecar 9 hours ago [-]
Yes, for over a decade now and it's already available in the cloud. It's a spinn-off of human brain project. Search for spinnaker neuromorphic computing.
giancarlostoro 8 hours ago [-]
Microsofts Bitnet is really solid on CPUs (though still better to train on a GPU for speedier training) but theres a lot of room for improvement in these spaces.
6keZbCECT2uB 3 hours ago [-]
I did not spot any errors in my skim of the architectures I'm familiar with where I would expect an LLM to go wrong, and it has a nicely scoped overview of topics that people in the space should familiarize themselves with.
I don't think I've seen a better primer.
hadlock 19 hours ago [-]
Maybe it's just me, but between the extremely thin font and layout design, I find this extremely difficult to parse. Overuse and improper use of italics is confusing as well.
monkeydust 4 hours ago [-]
Yea stresses the brain to much. I have built my own reader app, one of the simple things it does is very quickly convert article to highly readable font, with correct spacing on dark mode. Most things I read now are through that all, plus it saves it locally and remembers where I left off so useful for long form reads when I am on tube.
SomeHacker44 17 hours ago [-]
I didn't believe it till I tried to read it. Insane. Impossible to read.
pure_magic 18 hours ago [-]
This text is most likely AI-generated. Borderline unreadable.
alain94040 3 hours ago [-]
I'm surprised by this comment. It's probably mostly AI generated, but I have no problem reading the first 10 pages, and I found them clear and correct.
Might be that we come at this page with different experience and knowledge. I'm not seeing any specifically hard to read AI here.
iamflimflam1 15 hours ago [-]
My meatbag or robot scored it as likely a robot and most likely Claude.
I just used it on 4 texts. The 1st version was the output from my first (chatGPT) prompt. The 4th was the last prompts of "reduce, make it less AI, etc" of the same original request/output.
In all cases it was meatbag. In the 1st case gave chatGPT 0.35. The remaining ones were much lower.
EDIT: I didn't ask just to generate some text. The original prompt was very specific, which is most of the time the way I use LLMs anyways. Probably it catches better the "hey just generate this email...".
ghurtado 17 hours ago [-]
There's no borderline about it.
When the density of italicized words is about 1:1, I'm out.
postalrat 18 hours ago [-]
Most like AI generated? If you can't tell then what's the problem?
catlifeonmars 18 hours ago [-]
It’s shorthand for “it’s unreadable, but in the way an AI would write it”. I’m not OP, but that’s the issue. If it was well written (even if AI written), there would be no problem.
As it turns out, you still need to be a good writer in order to prompt an AI to write well.
ghurtado 17 hours ago [-]
Wise people tend not to talk in absolutes, because they always realize they might be wrong.
Ignorant people are always certain.
catlifeonmars 7 hours ago [-]
This is just a rephrasing of the Dunning-Krueger effect, but quantized to a binary domain (ignorant vs wise).
0x20cowboy 5 hours ago [-]
Strong opinions…
Zacharias030 16 hours ago [-]
always?
ghurtado 12 hours ago [-]
I am one of the ignorant ones, after all, so I'm probably wrong on this. Wait...
paaloeye 18 hours ago [-]
def. AI-assisted, both content and style style/nav.
Content-wise, it looks legit, but with that amount of stuff, it's nearly impossible to say for sure.
Not slope tho!
paaloeye 18 hours ago [-]
Not too shabby, but it shouldn't have been like 3 posts
geremiiah 6 hours ago [-]
It's an interesting topic, but the article is completely AI generated, which isn't surprising considering the startup, but still, if you don't even go over the output what's the point?
cman1444 2 hours ago [-]
How do you know they didn't go over it?
brcmthrowaway 18 hours ago [-]
I have a feeling the hardware architecture for LLMs are completely wrong. There's no way hundreds of kilowatts is required for intelligence.. just in terms of the physics. Is there someone out there in the analog/neuromorphic computing world that could make these power-hungry monsters completely redundant?
skyberrys 17 hours ago [-]
It's more the economics of what sells that lets the power hungry monsters thrive. If you can tape out one chip design, and one gives you 1 TOPS and the other is 5 TOPS, the sales will go to the 5 TOPS because power is a downstream cost. That being said, it's only a matter of time before both chips are maximum TOPS and the only differentiator is power. My latest simulations are showing a competitive design around 70W, but I'm still confirming it and it would really be beneficial to do some data gathering tapeouts at larger nodes to feel like it's worth the risk of more expensive nodes.
brcmthrowaway 17 hours ago [-]
Systolic hardware?
skyberrys 16 hours ago [-]
Race logic with pulses, it's a descent of systolic arrays. There is less of an array shape and more about time.
whaleofatw2022 15 hours ago [-]
Dumb question, wouldn't node size impact the logic latency itself on some level?
skyberrys 14 hours ago [-]
Yes smaller nodes give lower power operation with faster transistor switching speeds. You can also underpower a node to intentionally cause slower transistor switching speeds.
brcmthrowaway 13 hours ago [-]
Where can I find the whitepaper?
baybal2 10 hours ago [-]
[dead]
logicallee 18 hours ago [-]
human brain uses 20 watts.
jagged-chisel 18 hours ago [-]
I have a gut feeling the particular stochastic processes we’re currently using are the wrong ones to go beyond LLMs into actual intelligence.
imtringued 11 hours ago [-]
You're completely wrong about hardware being the bottleneck.
The software architecture LLMs run on was chosen precisely because hardware can brute force it and brute force has become the method of choice ever since, because it is easier to scale.
And here you're saying "what if everyone was wrong and what we need is just more brute force?", without looking at the reason that makes brute force necessary to begin with.
kzrdude 8 hours ago [-]
This is also the problem, we are continuing with LLMs because they fit hardware well, instead of experimenting with trying to model knowledge (how do I know what I know) and metacognition (what do I know).
Current architecture can't tell the difference between fact and fiction, so to speak. (Which I think is a better way to say it rather than truth and falsehood; facts have provenance and context, not just "true" or "false".)
brcmthrowaway 13 hours ago [-]
Can memristors help AI?
martheen 13 hours ago [-]
If they ever figure out how to make it in any usable scale, maybe.
peter_d_sherman 13 hours ago [-]
I personally found this article highly educational...
One of the best deep-dives on AI chip architectures that I've ever read.
Other posters are welcome to their opinions, but I (again, personally) thought it was great!
jhide 18 hours ago [-]
[dead]
klm127 4 hours ago [-]
AI chips feel like a gimmick. An excuse to add "powered by ai chips" to the marketing material. That sort of phrase in on just about every consumer laptop now.
My brother bought one. I challenged him to find something that actually used the AI chip on the laptop. He couldn't - it always read at 0% in task manager.
As a consumer, I'd rather have that much more graphics card, hard drive space, or RAM for the price than a specialized chip thats almost never used and, when it is used, only saves you a little electricity.
anentropic 4 hours ago [-]
FWIW the article is not talking about those kind of AI chips at all
Companies like EnCharge, Mythic, etc.
And much more efficient devices like RRAM, MRAM, and FETs. Like FE-FETs with AlScN.
The stuff just coming out of research or still in research is more exciting in terms of the potential for truly huge efficiency and performance boosts.
Taalas is interesting also because of it's efficiency and speed. Guess it was just purchased by AMD.
Multiplication may make sense for the math, but it's just not practical in hardware. There has to be a better, much simpler operation, that can give you equivalent results. Yes, you may have to re-think and re-train everything, but eventually it should make current architectures obsolete.
Recently on HN: https://news.ycombinator.com/item?id=49352470
Might be that we come at this page with different experience and knowledge. I'm not seeing any specifically hard to read AI here.
https://meatbag.atomic14.com/
In all cases it was meatbag. In the 1st case gave chatGPT 0.35. The remaining ones were much lower.
EDIT: I didn't ask just to generate some text. The original prompt was very specific, which is most of the time the way I use LLMs anyways. Probably it catches better the "hey just generate this email...".
When the density of italicized words is about 1:1, I'm out.
As it turns out, you still need to be a good writer in order to prompt an AI to write well.
Ignorant people are always certain.
Not slope tho!
The software architecture LLMs run on was chosen precisely because hardware can brute force it and brute force has become the method of choice ever since, because it is easier to scale.
And here you're saying "what if everyone was wrong and what we need is just more brute force?", without looking at the reason that makes brute force necessary to begin with.
Current architecture can't tell the difference between fact and fiction, so to speak. (Which I think is a better way to say it rather than truth and falsehood; facts have provenance and context, not just "true" or "false".)
One of the best deep-dives on AI chip architectures that I've ever read.
Other posters are welcome to their opinions, but I (again, personally) thought it was great!
My brother bought one. I challenged him to find something that actually used the AI chip on the laptop. He couldn't - it always read at 0% in task manager.
As a consumer, I'd rather have that much more graphics card, hard drive space, or RAM for the price than a specialized chip thats almost never used and, when it is used, only saves you a little electricity.