coolvision

6 ms·

1.
by coolvision 3mo ago
this is brilliant! I just went through a few years of this, very relatable
2.
by coolvision 4mo ago
thanks for the Xdrive pointer, I just got samples of tinymovr and odrive controllers (with FOC), they seem nice but are more expensive.
3.
by coolvision 7mo ago
it's funny how your forecast reaches such a similar results as "AI 2027"
4.
by coolvision 7mo ago
to me, programming has always been about problem solving, not s/w in itself. I love that with more progress in computing I can solve harder challenges, faster. I'm in my 40s and am more excited than ever about my work.
5.
by coolvision 8mo ago
https://grgv.xyz/
6.
Exploring vocabulary alignment of neurons in Llama-3.2-1B(grgv.xyz)
1 points by coolvision 1y ago|0 comments
7.
by coolvision 1y ago
This is very misleading, just look at the data: https://en.wikipedia.org/wiki/List_of_unicorn_startup_compan... Yes, it's different for different countries, and some EU countries are less startup friently. But cou
8.
by coolvision 1y ago
Might it be because of more relaxed culture, and less pressure on researchers and engineers? Innovation likes freedom and exploration, not deadlines.
9.
by coolvision 2y ago
This stopped being a territorial dispute long ago. Russia is only interested in a ceasefire as a pause before occupying most of Ukraine. Not a single time Russia has even hinted to accepting presence of any sort of third party force in Ukra
10.
by coolvision 2y ago
This is rich. Russia occupied and annexed Crimea, and a part of eastern Ukraine. Of course Ukraine was seeking NATO membership.
11.
by coolvision 2y ago
Is it better to invest into TSMC or NVIDIA stocks, in context of AI growth?
12.
by coolvision 2y ago
nice! did you use cloud GPUs or built your own machine?
13.
by coolvision 2y ago
PX4 with Pixhawk FC would be much better in terms of working out of the box. But of you use it with a random "supported" FC then still lots of debugging is needed
14.
by coolvision 2y ago
how does it compare to 8-bit/4-bit quantization in terms of speed/accuracy?
15.
by coolvision 2y ago
I've worked a bit on inductive syntax-guided synthesis, but did it at the worst time -- right before it was made obsolete by LLMs: https://grgv.xyz/inductive_program_synthesis/
16.
by coolvision 3y ago
I tried to make a robot simulator with rapier, with a mixed sucess: https://grgv.xyz/blog/simulator4/ The performance is great but realism of joints and motors is lacking, and not suitable realistic simulation.
17.
by coolvision 3y ago
i moved to 4 day work week with a small pay reduction. I spend Fridays and some evening on learning and side projects. might earn a bit less now, but then i can be sure that my skills and knowledge will stay relevant
18.
by coolvision 3y ago
bookmarking extension, not much traction though https://chromewebstore.google.com/detail/autolicious/jbmpoml...
19.
by coolvision 3y ago
You need therapy. It's not a good idea to explain/justify mental health issues with some external factors. It's all in your head. The worst that can happen if you can't find a job, is that you just wait for a short bit u
20.
by coolvision 3y ago
https://exrx.net/WeightTraining https://exrx.net/WeightTraining/Research
21.
Task vectors and analogy making in LLMs(grgv.xyz)
2 points by coolvision 3y ago|0 comments
22.
by coolvision 3y ago
are we discussing here "intelligence" or high IQ? those problems sit on high IQ level for humans, and are not a good fit for testing intelligence in general. if one claims that passing those tests is a requirement for intelligence
23.
by coolvision 3y ago
>They can't reason logically, draw meaningful conclusions, or grasp the nuances of context and intent. Can you come up with some examples that would demonstrate this on ChatGPT (GPT-4)? I'm really curious, as last time I tried
24.
by coolvision 3y ago
It was written before ChatGPT, GPT-4 and maybe even before GPT-3. So his impressions of limitations of LLMs should not be taken as facts and should be re-evaluated. If you check yourself, I'm sure you will find that ChatGPT is quite ca
25.
Reflections on 2023(grgv.xyz)
3 points by coolvision 3y ago|0 comments
26.
Llama.MIA – fork of Llama.cpp with interpretability features(grgv.xyz)
5 points by coolvision 3y ago|0 comments
27.
by coolvision 3y ago
wrote few posts for my site/blog and that increased the traffic 20x! from 25/month to 500/month ) well, small steps...
28.
by coolvision 3y ago
This might be a very US thing, so one could move to a different country for few years. F.ex. in EU, experience can be proven by internships, freelance jobs, opensource projects etc... I've never seen years of job experience required.
29.
by coolvision 3y ago
Moore's law is not about transistor sizes, it's for number of transistors on a chip. This is still going strong, at least for the next few years.
30.
Reverse engineering of LLM analogy-making(grgv.xyz)
1 points by coolvision 3y ago|0 comments

More ›