jszymborski

7 ms·

91.
by jszymborski 4mo ago
> wtf even is a "default value"? The article mentions the default value is a null, which would be the cue to run whatever computationally expensive op or query the db or hit the disk etc... that you would normally run if you ha
92.
by jszymborski 4mo ago
I think the Qwen 0.6B is so cool. It is super fast and as illustrated here it has a clear niche, esp. when fine-tuned. I'm also interested in it as a student for distillation.
93.
by jszymborski 4mo ago
I run local models and they work fine for me, but specifically for use in coding harnesses, I'm having a hard time. Tools tend to end up in the same loop, trying to `ls` the same folder or `grep` the same file, over and over and eating
94.
by jszymborski 4mo ago
> The phenomenon of "correlated RNG" (or "CRNG") This is a pretty funny abbreviation since CRNG is sometimes "cryptographic random number generator", which would not be susceptible to this correlation. Albei
95.
by jszymborski 4mo ago
Totally true that what is and is not sustainable is more complicated than it first appears. I think what I'm honing in on is the idea that hand cranks produce very limited, often interrupted power and are relative low-tech, both of whi
96.
by jszymborski 4mo ago
There's a bit of a difference between population control and reaching a sustainable equilibrium. One can also argue the death of all life on earth is a pretty evil logical conclusion of infinite growth on a finite planet.
97.
by jszymborski 4mo ago
As a bit of an aside, I really like the idea of trying to design things with the constraint of it having to be able to run off a hand-crank. I feel like it is not only an interesting engineering challenge but one that might lead to a more e
98.
by jszymborski 4mo ago
That's totally fair, and I'm often reminded not to paint all "Frontiers" journals with the same brush (although it's difficult for me sometimes).
99.
by jszymborski 4mo ago
So, I'm not making any accusations or decelerations of "bad science", I've only read the abstract and this isn't my field. That said, some relevant context here is that: (1) Case studies are some of the most easily
100.
by jszymborski 5mo ago
Under first-last-the-post, we actually don't :P
101.
by jszymborski 5mo ago
I'd have voted for it as a judge! I stopped participating in hackathons because they became just about integrating sponsor APIs and half-baked start-up ideas based on quarter-baked demos... All entirely corporate and loveless. This, ho
102.
by jszymborski 5mo ago
I think we're hugging it to death.
103.
by jszymborski 5mo ago
I guess we'll just have to agree to disagree!
104.
by jszymborski 5mo ago
Wouldn't the Esperanto of Flags be the flag of Esperanto?
105.
by jszymborski 5mo ago
Esperanto actually already has a flag https://en.wikipedia.org/wiki/Esperanto_symbols#/media/File%...
106.
by jszymborski 5mo ago
> But if you’re talking to a performer, and they have a fake, glassy-eyed smile, and go through all the correct motions, while obviously being totally checked out, you’re not asking the right questions. Or it's an indication you are
107.
by jszymborski 5mo ago
There was a viral Medium post that was about LLMs but then there was a reveal at the end was that the whole thing was a ChatGPT post. That was my first "wow" moment. It was on hackernews... anyone know what I'm talking about?
108.
by jszymborski 5mo ago
and later in that thread https://bsky.app/profile/neocities.org/post/3mnkqgxt3kn2k . The post acknowledging the site is up was posted 7 mins ago though :)
109.
by jszymborski 5mo ago
https://neocities.org seems to up right now...
110.
by jszymborski 5mo ago
There's a lot of stuff that looks officious enough that will trick folks, especially those distracted or not well-versed in the attack vector.
111.
by jszymborski 5mo ago
I hate shoving LLMs everywhere, but honestly this is probably a good use case for tiny models like the 0.6B Qwen model to flag account names for human review.
112.
by jszymborski 5mo ago
I can't begin to imagine the pain and stress caused by those symptoms, but I am so very happy to hear the prognosis is quite good. A linked scientific article makes the case that this is a very new diagnosis, and increased awareness mi
113.
by jszymborski 5mo ago
Totally agree that it is "encoding" in the general sense, but I think they are referring to the lack of an "encoder" neural network.
114.
by jszymborski 5mo ago
Guides as an astrolab or compass might
115.
by jszymborski 5mo ago
Landed?
116.
by jszymborski 5mo ago
Notably, if someone is "cooked", it's bad. If someone is "cooking", it's often (but not necessarily) positive, most commonly in the form "let him/her cook" or "he's/she's cook
117.
by jszymborski 5mo ago
I don't really understand how being scared/traumatized by videos of bike accidents will increase that child's visibility. The onus here is on municipal and federal governments to make roads and cars safer.
118.
by jszymborski 5mo ago
It's a result of the second Quebec referendum. The Clarity Act may appear like it facilitates leaving the federation, but many critics (among them federalist and sovereigntists) believe that the law is too vague as it give the House of
119.
by jszymborski 5mo ago
Important context, this referendum isn't binding, but rather a referendum on whether a binding referendum should be held. Separation is deeply unpopular, but Smith has been putting her thumb on the scale every step of the way, and this
120.
by jszymborski 5mo ago
"Stream fraud" is already a thing, so if anything this would make botting more profitable. Great synergy to be had here by fraudsters

More ›