Rendered at 08:40:22 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
ilaksh 8 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 18 minutes 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.
hadlock 8 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.
SomeHacker44 7 hours ago [-]
I didn't believe it till I tried to read it. Insane. Impossible to read.
pure_magic 8 hours ago [-]
This text is most likely AI-generated. Borderline unreadable.
iamflimflam1 5 hours ago [-]
My meatbag or robot scored it as likely a robot and most likely Claude.
When the density of italicized words is about 1:1, I'm out.
postalrat 8 hours ago [-]
Most like AI generated? If you can't tell then what's the problem?
catlifeonmars 7 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 7 hours ago [-]
Wise people tend not to talk in absolutes, because they always realize they might be wrong.
Ignorant people are always certain.
Zacharias030 6 hours ago [-]
always?
ghurtado 2 hours ago [-]
I am one of the ignorant ones, after all, so I'm probably wrong on this. Wait...
paaloeye 8 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 8 hours ago [-]
Not too shabby, but it shouldn't have been like 3 posts
brcmthrowaway 3 hours ago [-]
Can memristors help AI?
martheen 2 hours ago [-]
If they ever figure out how to make it in any usable scale, maybe.
brcmthrowaway 8 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 6 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 6 hours ago [-]
Systolic hardware?
skyberrys 6 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 4 hours ago [-]
Dumb question, wouldn't node size impact the logic latency itself on some level?
skyberrys 4 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 3 hours ago [-]
Where can I find the whitepaper?
imtringued 58 minutes 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.
jagged-chisel 8 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.
logicallee 7 hours ago [-]
human brain uses 20 watts.
peter_d_sherman 3 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!
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.
https://meatbag.atomic14.com/
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.
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!