_davide_

6 ms·

1.
by _davide_ 5d ago
Buoooo, boooring; I wanna see explosions
2.
by _davide_ 6d ago
I understand the appeal of saying fuck you to billionaires and pathological liars CEOs and the annoying usages of AI, but i don't understand the wider "fuck you AI". AI are here, is it going to demolish and (maybe) reconstruc
3.
by _davide_ 6d ago
You are thinking in reverse, think it more like a kickstarter, you sponsor the author, at that point the author would probably just release as public domain and you could buy a physical "official" copy on kindle or whatever, but y
4.
by _davide_ 6d ago
I switched between several local and remote providers and models and over different API (anthropic/openai) and it worked fine, just some minor issues but they were fixed within an hour. And the system prompt worked great regardless, s
5.
by _davide_ 6d ago
100% every professional developer at some point should build its own harness as daily driver
6.
by _davide_ 6d ago
> > building your own harness is a huge undertaking, a deep rabbit hole. > I eventually gave up on this task. It's not trivial, but cmon, i did during weekends from my phone and FOR ME it's so much better than the codex o
7.
by _davide_ 7d ago
CoT, was being studied using GPT-2, so...who invented hot water first?
8.
by _davide_ 8d ago
Unless you were paid during those five years
9.
by _davide_ 8d ago
I agree, copyright is a just yet another monopolistic tool and it isn't making the market any favor. Copyright could be gradually be reduced to a few years, the market will adapt, for example collecting money before producing the co
10.
by _davide_ 12d ago
Oh no, again
11.
by _davide_ 24d ago
Not a layer but a bit nerdy on the law but specially on the copyright side. Copyright is a framework to protect creative work. One could easily argue that an LLM model is just weights automatically produced from data and there is no creati
12.
by _davide_ 24d ago
if someone knows this guy, please provide him a good therapist, he really needs one.
13.
by _davide_ 27d ago
wouldn't it make that much more dangerous?
14.
by _davide_ 27d ago
Under this interpretation models themselves do not have copyright protection either
15.
by _davide_ 29d ago
the fact that sometimes someone finds a way to workaround the rules doesn't make them a ""suggestion"" they are very solid, just not absolute
16.
by _davide_ 1mo ago
> No one is seriously going to leave GitHub because of reliability had a discussion with the team yesterday, literally every one wants to leave and looking for the best alternative
17.
by _davide_ 1mo ago
in reality even the mention of a prohibition is enough to make the model reject that no matter what
18.
by _davide_ 1mo ago
i beg to differ, in an ideal world a system possibly is a binding law and high end models are starting to be really aligned to the exact system prompt. The instructions must be simple to follow, if you start doing complex rules it'll c
19.
by _davide_ 1mo ago
or completely deleting the app and use the browser version for desperation
20.
by _davide_ 1mo ago
Disagree with the conclusion, even without carefully curated context every high end LLM perform just as well, maybe with an extra detour. In contrast if even one of the learnings is not up to date or doesn't apply to the current situat
21.
by _davide_ 1mo ago
Yes, given a direction, they are pretty good at "fuzzing", trying out everything and eventually find something. Super useful, but it doesn't have the same level of precise targeting you could expect from a high end model.
22.
by _davide_ 1mo ago
Just go online rent a few servers, download K3, ablate it, run the thing, shut it down. Will probably cost a few hundreds bucks to ablate, but f* this non-sense paternalistic sh*. I wish i had the time to do it and blog it, "in your fa
23.
by _davide_ 1mo ago
this is a silly oversimplification
24.
by _davide_ 2mo ago
there is a simpler way, make a complete mental model of the changes and ask questions to confirm your understanding. so much faster.
25.
by _davide_ 2mo ago
feeding an existing conversion with output and reasoning from model x to model y will have many side effects, most likely a degradation of perfomance and alignment. should be done at least within the same family and generation of models...
26.
by _davide_ 2mo ago
php is still a thing?
27.
by _davide_ 2mo ago
What's the point of "releasing it"? It only makes sense in the labs, and what would calling your APIs give "me" as a researcher, other than a baseline to beat? mah the model is real, but it feels 100% internal
28.
by _davide_ 2mo ago
Seriously ready to fully drop google emails and services, i'm using vivaldi + personal email for a while now, the migration was slow a bit painful but complete. Only the phone is still attached to play store and google, but i'm 10
29.
by _davide_ 2mo ago
pls..? more like `if you don't we'll all die, so do it`
30.
by _davide_ 2mo ago
This post seems to be created out of thin air rather than real experience and data

More ›