0-_-0

4 ms·

1.
by 0-_-0 2mo ago
What if there are math problems that can only be brute forced? How do we know that Navier-Stokes even has a nice and elegant solution?
2.
by 0-_-0 2mo ago
Is there an example of foveated rendering implementation?
3.
by 0-_-0 2mo ago
Instruct the AI to output an .MD file that can be used to reconstruct the code
4.
by 0-_-0 3mo ago
Why didn't Russia just use local models?
5.
by 0-_-0 3mo ago
You're hallucinating
6.
by 0-_-0 3mo ago
Affected by what exactly?
7.
by 0-_-0 3mo ago
The kv cache is not discretised, only the output tokens.
8.
by 0-_-0 3mo ago
Reasoning is already taking place in latent space, the CoT is just a sideeffect
9.
by 0-_-0 3mo ago
I wonder if you could prune the kv cache as well
10.
by 0-_-0 3mo ago
Same is true for 1
11.
by 0-_-0 3mo ago
Right, so it was sealed only in the biological sense, which makes sense since there needs to be an inflow of energy
12.
by 0-_-0 3mo ago
No but they could over 5 million years
13.
by 0-_-0 3mo ago
The information that was missing from my life: For the Average Man, gaining an inch of height is worth losing roughly $25k in annual salary. Meanwhile, an inch of pecker is roughly offset by reducing annual salary by $128k.
14.
by 0-_-0 4mo ago
You would need to hide the existence of the original profile while in the decoy profile for this to work, which GrapheneOS considers too complex to implement
15.
by 0-_-0 4mo ago
It's not a benchmark, it's a meme
16.
by 0-_-0 4mo ago
They want to be power efficient, while Nvidia doesn't. That means a much higher cost (larger chip area) for the same performance.
17.
by 0-_-0 4mo ago
Yes, but all subsequent tokens do depend on them, so they are still being charged for
18.
by 0-_-0 4mo ago
Don't you jump the highest with com_maxfps 333 anyway? Or did Quake Live change this from Quake 3?
19.
by 0-_-0 4mo ago
You can only see satellites during twilight when they can reflect sunlight. Don't panic.
20.
by 0-_-0 4mo ago
In that article Grok shows highest neutrality
21.
by 0-_-0 4mo ago
This made me wonder about a 3D printer alternative that builds things by folding a thin sheet of metal into arbitrary shapes instead of extruding filament.
22.
by 0-_-0 4mo ago
Even if you had all that you would get completely different weights out at the end, and you also don't have the resources to "compile" an LLM because the compilation can cost $100M. If you were given the training data but not
23.
by 0-_-0 4mo ago
Source code makes it easier to modify a binary, training data doesn't make it easier to modify a trained network.
24.
by 0-_-0 4mo ago
If we can launch 1M satellites, how many telescopes can we launch?
25.
by 0-_-0 4mo ago
And what are they reflecting when they are in Earth's shadow?
26.
by 0-_-0 4mo ago
32 bits vs 4 bits it looks like
27.
by 0-_-0 4mo ago
Browser fingerprinting can already pinpoint you exactly. We should focus on that.
28.
by 0-_-0 4mo ago
That math makes no sense
29.
by 0-_-0 5mo ago
These never ever accomplish anything
30.
by 0-_-0 5mo ago
Not if it's mass produced

More ›