jey

5 ms·

31.
by jey 10mo ago
> When I ran this program, I expected the `CF_OEMTEXT` string to have the byte 44, but it didn’t. It had the byte 90. We will start unraveling this mystery next time. Whoa there exists something Raymond Chen didn’t know about Windows co
32.
by jey 10mo ago
This seems to be incorporated into current LLM generations already -- when code execution is enabled both GPT-5.x and Claude 4.x automatically seem to execute Python code to help with reasoning steps.
33.
by jey 11mo ago
And the person who did the implementation, Lyapsus, did it without access to the hardware?? https://github.com/nadimkobeissi/16iax10h-linux-sound-saga/i...
34.
by jey 1y ago
Anyone have a link to the patch in uutils? Curious to see that the problem and solution were.
35.
by jey 1y ago
Get two subscriptions if it's delivering that much value and you hit the limits?
36.
by jey 1y ago
Greenland is an “autonomous territory” of Denmark
37.
by jey 1y ago
I think it’s feasible because of their token prefix prompt caching, available to everyone via API: https://docs.anthropic.com/en/docs/build-with-claude/prompt-...
38.
by jey 1y ago
I’m pretty sure you are supposed to declare agricultural products at customs. Sure, if the apples are cooked into a pie that’s probably fine but I believe most countries don’t let people bring in fresh fruit because of the possibility tha
39.
by jey 1y ago
That the annotation applies to variables and not types is surely an oversight or mistake right? Seems like it could have been easier to initially implement that way but it just doesn’t seem to fit with how C type system works. (Yes it will
40.
by jey 1y ago
Try Claude Code. It’s perfect for this type of use case.
41.
by jey 1y ago
Odd. How can one get the vibes without looking at the code? Like gleaming diamond vibes versus confused spaghetti vibes.
42.
by jey 1y ago
I’ve had good results using Claude Code by specifically coaching it on how its implementation should be. It’s not always perfect and sometimes I do have to try again, but it’s remarkably effective when given enough guidance. If only I could
43.
by jey 1y ago
Btw despite the helpful pedanticism[1] of HN, I think your English is impeccable and idiomatically natural for someone who has probably not spent much time in immersive spoken English environments. 1. pedantry
44.
by jey 2y ago
Works fine if you copy the subject line into the body too
45.
by jey 2y ago
Unfortunately that’s very vague because there’s many notions of “dual space” within applied mathematics, even just considering the parts relevant to engineering applications.
46.
by jey 2y ago
Apparently this is the background context: https://www.npr.org/2025/02/14/g-s1-48923/stonewall-monument...
47.
by jey 2y ago
You're not too far from the truth -- most mapping services render political boundaries differently based on where you're accessing the site from.
48.
by jey 2y ago
That's true. I happily concede!
49.
by jey 2y ago
Hm. Why create this environment where the people with the most options will preferentially be induced to leave? Also the impact on the company project seems less predictable. Aside from avoiding the extreme difficulty involved in doing layo
50.
by jey 2y ago
In San Francisco there’s a unique secondary “high-pressure water supply system” which is fully independent of the regular water mains of the city. The Twin Peaks Reservoir that was built to supply it has two separate basins so that either
51.
by jey 2y ago
looks to be this: https://github.com/docker/for-mac/issues/7527#issuecomment-2...
52.
by jey 2y ago
I recently ran into a bunch of problems running neovim under GNU Screen with `TERM=screen.xterm-256color`. There was some kind of problem relating to GNU Screen's parsing and re-transmit of certain full-color terminal escape codes. I d
53.
by jey 2y ago
iTerm2 increasingly seems too complex and bloated to me, with too many security issues. I haven't shopped for a new terminal emulator on macOS in a long time, but perhaps it's now time. I should also get around to switching to tmu
54.
by jey 2y ago
Generally yes, vanilla gradient descent gets plenty of use. But for LLMs: no, it’s not really used, and stochastic gradient descent provides a form of regularization, so it probably works better in addition to being more practical.
55.
by jey 2y ago
> F* is oriented toward verified, effectful functional programming for real-world software, while Lean is geared toward interactive theorem proving and formalizing mathematical theories, though both support dependent types and can serve
56.
by jey 2y ago
Yes, but it's impractical unless you have galactic-scale matrices to multiply (at least).
57.
by jey 2y ago
This notation doesn't require graduate-level statistics knowledge, it's more like stuff that would be covered in a first mathematical course on probability and statistics. It's totally practical to learn this stuff on your ow
58.
by jey 2y ago
This is just a practical approximation to the actual mathematical definition of GELU, which is `GELU(x) := x * Φ(x)` where Φ(x) is the CDF of the Gaussian distribution.
59.
by jey 2y ago
> Parents rebelled and half the states […] banned it. Can you elaborate or provide a citation for me to learn more?
60.
by jey 2y ago
That's fully in line with what they're announcing here. It's the announcement of a new project that has passed the prototyping stage, but one that has not reached the 1.0 stage.

More ›