10xDev

6 ms·

1.
by 10xDev 6d ago
AI is more like an ongoing research project than it is a product. GPUs are currently the product that is profitable and Nvidia will keep funding labs to keep buying GPUs.
2.
by 10xDev 6d ago
Zugzwang is everywhere.
3.
by 10xDev 7d ago
Not exactly scientific but it is at least entertaining and some things do sound plausible https://www.youtube.com/watch?v=Gw_hnD7m00M
4.
by 10xDev 7d ago
Risk/reward. Isn't the reward worth the risk?
5.
by 10xDev 7d ago
The idea that they are entirely trained on human intelligence is already outdated. Yes, the earlier models relied heavily on RLHF and human curated data but we have since moved on to synthetic data produced by the models themselves and rein
6.
Muse, your personal AI agent.(auth.muse.ai)
7 points by 10xDev 8d ago|0 comments
7.
by 10xDev 8d ago
Eventually the bottleneck might be people themselves.
8.
by 10xDev 8d ago
Compute will always be the bottleneck even if this were true.
9.
by 10xDev 9d ago
If Astra can actually reason in 3D space, I would like to see it play The Witness.
10.
by 10xDev 12d ago
This statement is behind times, go and watch astra play Pokemon: https://www.twitch.tv/gpt_plays_pokemon
11.
by 10xDev 12d ago
You generally forget things you don't retrieve. It is not really a bug but a way to declutter for efficiency. That's not the same as it not fitting your context window because it was at the beginning.
12.
by 10xDev 13d ago
Unless you infinitely increase the context window, its memory will always be limited. And their latest ARC3 result with and without harness demonstrates how important not discarding memory is for learning.
13.
by 10xDev 13d ago
It will be AGI once it can update its own weights. It can't be "general" intelligence if its weights are frozen and requires to be updated manually.
14.
by 10xDev 13d ago
It is about memory retention. No heavy lifting done on the reasoning side so I hardly see anything misleading here. Edit: update from fchollet https://x.com/fchollet/status/2095598451115614371
15.
by 10xDev 15d ago
No, the unsolved problem of AI is continuous learning. We never stop learning, we don't have a "training phase". You are always updating your world model even when you sleep. Also more quality training data does lead to great
16.
by 10xDev 2mo ago
That’s why you have a private dataset.
17.
by 10xDev 2mo ago
It is remixing what the industry created, not advancing past it. OpenAI even ended Sora in under a year. Anthropic doesn't even bother.
18.
by 10xDev 2mo ago
Not sure where I implied they are "raw base models" and not sure what "various adjustments" means here or how "ton of RLHF" contradicts anything. If we look at research for open source models, "adjustments
19.
by 10xDev 2mo ago
Coding models are not the destination. Coding models are just part of the bootstrapping process towards general intelligence.
20.
by 10xDev 2mo ago
Seeing the dramatic differences in scores just going from high to xhigh is just another demonstration of the bitter lesson: Just keep scaling search and learning. We are probably going to need a lot more GPUs.
21.
by 10xDev 3mo ago
This is putting Nintendo on a pedestal. Major Japanese studios in general are consistently publishing great games and increasing hiring count / raising salaries. Switch 2 is also essentially just a spec bump from switch 1 which came ou
22.
by 10xDev 3mo ago
Your strength comes both from your nervous system and physiology. Training your nervous system without also building a proper physical foundation to handle tension is a fast track to injury.
23.
by 10xDev 3mo ago
Do you ever IP ban?
24.
by 10xDev 3mo ago
This is a bot. It has been commenting all day every day. Why has it still not been removed?
25.
by 10xDev 3mo ago
You are essentially calling for a civil war. Reason why Russia pushes these ideas under fake "patriot" accounts.
26.
by 10xDev 3mo ago
You certainly see a lot of tourists and expats. And it is easy to see the worst behaving ones especially in parts of Asia.
27.
by 10xDev 3mo ago
Nothing to see here of course. Luckily some us have savings and can see the tide shift. Good luck.
28.
by 10xDev 3mo ago
As economies shrink and jobs become scarce, we may reach pre-ww2 order.
29.
by 10xDev 3mo ago
Don't worry, data centres in space will solve this problem.
30.
by 10xDev 3mo ago
> 4,197,168 job applications submitted by 3,372,132 applicants to 1,746 positions. So roughly 2000 applicants per position which there may be 1-3 of available? And that's across multiple industries. The rest of the article doesn&#x2

More ›