JustFinishedBSG

18 ms·

1.
by JustFinishedBSG 10d ago
They still run the same thirdparty firmware. Seeed Studio has the reTerminal Sticky too. M5 device are cool but they do have that, as ugly as it is useful, functional diagram printed on the back...
2.
by JustFinishedBSG 17d ago
That's a whole lot of text to say "it's markdown".
3.
by JustFinishedBSG 1mo ago
> almost as if it was trained to spit out as complex of sentences as it can. It probably was, at least inadvertently. Very easy to imagine that one of the post-training step (RLHF, DPO etc) reinforced the "sounds clever" behav
4.
by JustFinishedBSG 2mo ago
This doesn't (and cannot) solve what the post is talking about
5.
by JustFinishedBSG 2mo ago
At a very high level mathematics is basically 100% text/symbolic rewriting. You start from some set of postulate assumed true and you do your thing to get a new different set of equivalent assertions in a form that is more useful. I do
6.
by JustFinishedBSG 2mo ago
My experience may not be entirely representative because to be entirely honest I’m not exactly a great researcher and there are brilliant PhD students. That said it indeed was my experience that in the pre-PhD / early PhD period ( or
7.
by JustFinishedBSG 3mo ago
It's very hard those days to think of companies/people more arrogant than Anthropic/Dario, which is quite the achievement as the bar is very high. If that arrogance was well placed at least you could somewhat excuse it, but
8.
by JustFinishedBSG 5mo ago
Interesting, it doesn't seem intuitive at all to me. My (wrong?) understanding was that there was a positive correlation between how "good" a tokenizer is in terms of compression and the downstream model performance. Guess no
9.
by JustFinishedBSG 5mo ago
Most of the cost (to the government) for Windows is "support" (in a very general sense) and that cost isn't disappearing with Linux. Especially since it is easier to find badly underpaid (and not particularly competent) Windo
10.
by JustFinishedBSG 7mo ago
> for the same reason math problems require one to show their working. We don't put our transitional proofs in papers, only the final best one we have. So that analogy doesn't work. For every proof in a paper there is probably
11.
by JustFinishedBSG 7mo ago
> While it's noisy and complicated for humans to read through, this session info is primarily for future AI to read and use as additional input for their tasks. Context rot is very much a thing. May still be for future agents. Dumpi
12.
by JustFinishedBSG 7mo ago
I understand the idea but the way I work, a commit isn't "a" session, it's potentially tens of sessions with branching in each session. I honestly don't know if I'm doing something very wrong or if I have a ver
13.
by JustFinishedBSG 7mo ago
What problem were you trying to solve ? ( not that you need to solve one. I’m just curious )
14.
by JustFinishedBSG 7mo ago
https://github.com/edk2-porting/edk2-rk3588
15.
by JustFinishedBSG 8mo ago
Well according to the website you cannot buy a fully mechanical lens anymore so it doesn’t appear to be true that they sell cameras with zero electronic anymore
16.
by JustFinishedBSG 8mo ago
It's useless for LLMs and it's actually slower than Hailo 8H for standard vision tasks, so, why ?
17.
by JustFinishedBSG 8mo ago
How, when foundries are a natural monopoly? How can you realistically create a startup to build a multibillion / multi year fab ?
18.
by JustFinishedBSG 9mo ago
It could be 1 dollar a month, I'm still not paying to use my own ressources
19.
by JustFinishedBSG 10mo ago
I wouldn't trust LMArena results much. They measure user preference and users are highly skewed by style, tone etc. You can litteraly "improve" your model on LMArena by just adding a bunch of emojis.
20.
by JustFinishedBSG 10mo ago
There’s just no way to stop cheating client side despite what devs love to think. But server side anti cheat is much harder and requires more work; it’s much simpler to just install spyware / rootkits on the client and call it a day.
21.
by JustFinishedBSG 10mo ago
CTRL-F: "Fokker-Planck" > 97 matches Ok I'll read it :)
22.
by JustFinishedBSG 11mo ago
Exactly my plan yes :)
23.
by JustFinishedBSG 11mo ago
Pandoc isn't a typesetter, it's more a translation engine between various typesetting languages (and it can then call the typesetter). Pandoc is a ridiculously underrated and powerful tool but it solves a different problem. Some
24.
by JustFinishedBSG 11mo ago
I think I'm going to subscribe without any intention of using the app just as a "financing" donation. I love, and hate, LaTeX and the idea of a LaTeX successor / alternative is incredibly appealing. And the fact that the
25.
by JustFinishedBSG 11mo ago
> Has the situation changed on AI code legally speaking? I think the position has shifted to "let's pretend this problem doesn't exist because the AI market is too big to fail"
26.
by JustFinishedBSG 11mo ago
While your example is not AI related (should be handled by the LSP integration with "Rename Symbol") I agree that Zed's Next Edit Prediction model is *extremely* subpar. Imho they should either scrap it and just work on havin
27.
by JustFinishedBSG 11mo ago
> This makes sense, right? It's a relatively novel thing to be writing. It's really not though? Honestly I'm surprised coding agents fail hard at this task apparently
28.
by JustFinishedBSG 1y ago
> Let's not kid ourselves that maintaining 10TB with resiliency handling and other controls built in is something that is trivial. It is though. People in tech need to stop pretending everything they are doing is super complicated.
29.
by JustFinishedBSG 1y ago
"FastThingJS: A blazing fast thing library for humans . Made with on "
30.
by JustFinishedBSG 1y ago
I actually disagree. Modern encoding formats can, and do, hallucinate blocks. It’s a lot less visible and I guess dramatic than LLMs but it happens frequently enough that I feel like at every major event there are false conspiracies based o

More ›