aktenlage

5 ms·

1.
by aktenlage 8d ago
Would it speed up prompt processing if you increased the -ub (and -b) parameters.
2.
by aktenlage 8d ago
> Just wish we have a little bit of context See the comment of vocx2tx
3.
by aktenlage 8d ago
They resumed development on jemalloc only recently, after years of no releases.
4.
by aktenlage 9d ago
So does duckduckgo with their duck.ai. No outrage there.
5.
by aktenlage 11d ago
Enlighten me
6.
by aktenlage 11d ago
I use Ag integrated in the editor all the time. But while it is indispensable for me, I wouldn't say that this helps in all situations, nor is it the best tool to find connections in many situations.
7.
by aktenlage 25d ago
My take on the self correction is, that later research usually cannot be build on wrong results. So over the decades or centuries the correct results prevail.
8.
by aktenlage 27d ago
Wow, setting the "ubatch" size to 2048 (default is 512), like the "batch" size roughly doubled prompt processing speed.
9.
by aktenlage 28d ago
Hm, maybe. Though I thought that's a combination of what Corona did to the society, late-stage capitalism and me getting older and more sceptical.
10.
by aktenlage 28d ago
Good point on the systemic forces. Yet I feel your second point falls broadly in with what I meant with "choosing the right places". E.g., I also use YouTube, but don't watch shorts, avoid clickbait, ragebait, etc, and filter
11.
by aktenlage 28d ago
"Online" is not one place. And what the author is complaining about is not happening in the places I spend 99% of my online time on. That's just the saying as I know it. Regarding the post it would be along the lines of &quot
12.
by aktenlage 28d ago
This so exaggerated. If you think some of the people you know are assholes, that's fine. If you think all of them are assholes, you are either choosing the wrong places to be, or (more likely) the problem is with you.
13.
by aktenlage 29d ago
Thank you, very interesting to read. I haven't considered batch sizes at all up to now. Seems I should look into it. How did you get the prediction that good? The default spec-draft-max-n is 3 and if I go beyond, the output rate stays
14.
by aktenlage 1mo ago
Have you tried a mixture of experts model? Dense models have been quite slow for me, as I have only 6 GB VRAM. But with llama.cpp and --cpu-moe I get 200 t/s input and almost 30 t/s output with Gemma 4 26B A3B, which feels ok to u
15.
by aktenlage 2mo ago
Not OP, but it seems to be the one most likely not following the silicon valley playbook (aka the road to enshitification).
16.
by aktenlage 3mo ago
I'm impressed, because it is so minimalistic. On the other hand it reminds me of 2010ish robotic demo videos from academic researchers and willow garage. Problem with robotics is usually that it's easy to produce a good-enough loo
17.
by aktenlage 3mo ago
> completely unrelated They are both chess-inspired and roguelike. IMHO this is a completely normal "hey I've been doing something similar" response. Why are you so aggressive about it?
18.
by aktenlage 3mo ago
What's the difference of a "robot game engine" to a simulator like < https://gazebosim.org >
19.
by aktenlage 6mo ago
I work in robotics and with quaternions (mainly 6DoF SLAM and used to do robot arm kinematics), but I don't get the use case for this. Maybe provide some example use cases?
20.
by aktenlage 6mo ago
> most people know how to use their own money to further their own life if given capital and opportunity, thus capitalism is the solution I didn't know that "giving capital and opportunity" to people is inherent to capital
21.
by aktenlage 6mo ago
Very cool. Reminds me of 1D Pacman: https://abagames.itch.io/paku-paku
22.
by aktenlage 6mo ago
Where is that? I found "Return the simplest working solution. No over-engineering." which sounds more like the simplest fix.
23.
by aktenlage 6mo ago
Freezing assets is simple. Seizing them is a huge pain. The EU has a hard time agreeing how to do it and who takes the liability for the Russian claims.
24.
by aktenlage 6mo ago
Woa, that's a weird @ in the screenshot in line 6.
25.
by aktenlage 6mo ago
Can you elaborate on this? My guess would be, that because of their status as a government backed research institute, they invent a lot, but let others do the commercialisation. So patent fees seem like a natural choice for them, to recover
26.
by aktenlage 7mo ago
It does sound interesting and I signed up for the wait list. But I actually don't like pure chronological order. It feels like I have to look at everything to find the good stuff. Here I probably won't miss out too much if I check
27.
by aktenlage 7mo ago
I have a Pixel 6a with GrapheneOS. Runs great for years, except for one or two apps that require an "official" Android. Anyway, I now need to get the battery replaced, because apparently they are dangerous and Google pays for the
28.
by aktenlage 8mo ago
> Is it, though? Yes it is a Milchmädchenrechnung. I do not want to argue whether fusing states makes sense or not. This isn't even a moral condemnation of the original poster. All I say is that the equation `efficiency_gain == fede
29.
by aktenlage 8mo ago
Thinking the number of federal states is equivalent with an efficiency factor is utterly unsubstantiated. There may be a correlation (or not, I don't claim anything here), but `efficiency_gain == federal states before / federa
30.
by aktenlage 8mo ago
I tried it to review some C++ code. It actually found minor bugs, but the signal to noise ratio is too high (maybe 10% of the found issues were real issues)

More ›