jeffybefffy519

9 ms·

1.
by jeffybefffy519 4d ago
Honestly matches my experience with Sol and the fact that all the random message boards appearing on the internet are attributed to Sol then it makes sense that its the one which makes up most requirements.
2.
by jeffybefffy519 4d ago
Do you review the outputs?
3.
by jeffybefffy519 6d ago
History has shown that restrictions promote innovation and ingenuity.
4.
by jeffybefffy519 9d ago
I wonder how much money LG makes from all of these practices, and if they stopped doing it what impact to cost of a TV would be?
5.
by jeffybefffy519 10d ago
When can AI start to have a big impact on medicine. Thats honestly how it becomes meaningful. And maybe material science/manufacturing is where there’s big unlocks waiting for humanity
6.
by jeffybefffy519 11d ago
I wonder how many responses on this thread are from rogue agents...
7.
by jeffybefffy519 12d ago
If you consider these guys admitted they dont really have eyes on pre and post training, then incompetence really does seem more likely... especially with how fast they are moving. Its the SaaS playbook, move fast and break shit.
8.
by jeffybefffy519 12d ago
Shouldnt you give the AI a different task each time? otherwise the model companies just optimise for this benchmark because its in their best interest to...
9.
by jeffybefffy519 13d ago
Its funny, my experience with Sol has been awful. It really overworks problems and tracks into areas it does not need to... I just dont get how its good for some, and bad for others. It makes me suspect that the models performance is not ev
10.
by jeffybefffy519 15d ago
And turns out, the "frontier" labs have no human oversight of the training data going into these models... Explains so much
11.
by jeffybefffy519 21d ago
You sure? These links are directly from tinfoil.sh's technology page: https://tinfoil.sh/technology They use these technologies: - https://www.nvidia.com/en-us/data-center/solutions/confi
12.
by jeffybefffy519 21d ago
Not commenting on speed, but it seems models and their harnesses have generally gotten much worse over time. My suspicion is that the "Frontier" labs really dont have a strong handle on good quality evals that equal expectations o
13.
by jeffybefffy519 21d ago
It would open heaps of use cases, you could almost pass it over frames of images the camera sees in real time for example...
14.
by jeffybefffy519 26d ago
I dont get how this fingerprints the browser? Are they able to read back the playback of audio somehow?
15.
by jeffybefffy519 28d ago
Isnt there also a theory and some evidence that the connectome is not enough to model a brain. Based on the fact that butterlies connectome dissolves during metamorphosis to butterfly but it still remembers things from before that stage of
16.
by jeffybefffy519 29d ago
Exactly right, and nVidia is protecting their moat through business practices rather than genuine product innovation.
17.
by jeffybefffy519 29d ago
10000%
18.
by jeffybefffy519 29d ago
Can you give examples of the tooling and tests?
19.
by jeffybefffy519 29d ago
I find all effort levels of sol are the same in terms of amount of hallucinated unnecessary changes. Luna is much better all round on xhigh but my point still stands, every release of these new models is not an upgrade, its re-learning how
20.
by jeffybefffy519 29d ago
I mean it added additional changes when it doesnt need to. Its basically hallucinating changes it thinks it needs to make regardless of effort levels i try.
21.
by jeffybefffy519 1mo ago
Reading the comments in this thread, i honestly dont get it. 5.6-sol has felt like a regression in capability. In fact, every model since 5.3-codex has been a regression from OpenAI. I just find 5.6-Sol over engineers problems, takes absolu
22.
by jeffybefffy519 1mo ago
AI will divide computer science in two: 1. Normal Computer Science as it exists today, just much fewer doing it due to job supply dropping off 2. AI driven Product/Application Engineering, bulk will go here and the course will be shor
23.
by jeffybefffy519 1mo ago
Fact checking removal really made it way worse....
24.
by jeffybefffy519 1mo ago
The sad thing is, the majority of the population does and they are a major source of influence and completely unregulated. Something has to change.
25.
by jeffybefffy519 1mo ago
Surely the moat is the training data... with the data you can explore new architectures much easier and get step changes in performance.
26.
by jeffybefffy519 1mo ago
The problem i have with this argument is they still struggle to generate images of so many basic things that should be explained in their "world model" (text, fingers, reflections, faces) - so clearly they dont have a "world
27.
by jeffybefffy519 2mo ago
and building inference hardware
28.
by jeffybefffy519 2mo ago
I do a lot of novel work, that is not setting up a webapp - and totally concur on the novelty point with LLM's failing hard.
29.
by jeffybefffy519 2mo ago
Always seemed like algae or cyanobacteria are the only real nature based options that could be scaled up.
30.
by jeffybefffy519 2mo ago
Kind of interesting, really what experts do is sort/organise weights into categories that are optimal to work together. Seems like a lot of research could be done to extend this concept to group weights together for common inputs ahead

More ›