saretup

5 ms·

31.
by saretup 11mo ago
The top navigation definitely put me off, I immediately came back without even scrolling down.
32.
by saretup 11mo ago
histrionic and meretricious
33.
by saretup 11mo ago
Counterpoint: Humans visualize stuff in their minds before trying new things or when learning new concepts. An AI system with LLM based language center and a world model to visualize during training and inference would help it overlap mor
34.
by saretup 11mo ago
Seems reductive. Some applications require higher context length or fast tokens/s. Consider it a multidimensional Pareto frontier you can optimize for.
35.
by saretup 1y ago
There can be hundreds of safe, possibly effective, and cheap interventions. Can’t try them all. Higher quality of evidence helps narrow it down.
36.
by saretup 1y ago
The entirety of the internet vs government-sanctioned homegrown EU data.
37.
by saretup 1y ago
The math actually checks out. 5% of 800 million is 40 million. 40 million thoughts per year divided by 52 weeks per year approximately equals around 1 million thoughts per week.
38.
by saretup 1y ago
I would assume the next iterations/fine-tuned variants of current models would reach similar accuracy for TOON as they do for JSON. The current models unfortunately do not have TOON in their training set, so they would probably require
39.
by saretup 1y ago
On the other hand, for future employers you might wanna simply showcase the creation. They don’t have all the knowledge to appreciate the technicals or even the time to sit through it.
40.
by saretup 1y ago
> android (in an economic sense) is dead weight Well it depends on your business model. Android has much smaller user LTV in most cases (especially in apps with no ads and only IAPs/Subscriptions), but the CPI is also smaller, so th
41.
by saretup 1y ago
Duh? It just launched. That’s like saying there are way more people using vscode than cursor on the day cursor launched.
42.
by saretup 1y ago
Can’t believe the top comment to this post is an ad of a competitor. No offense, your extension seems fine, but realistically how are you planning on surviving given OpenAI has moved to this space?
43.
by saretup 1y ago
Yeah so probably one or more of the pins are bad in your cable, since the usb c pin design the same from left to right in either orientation: https://www.delock.com/infothek/USB-3.2_USB-Type-C/usb-3.2-u...
44.
by saretup 1y ago
That’s what she said
45.
by saretup 1y ago
Why is this old post from 2022 (Updated Nov 2024) with no scientific information at the top of HN?
46.
by saretup 1y ago
You clearly haven’t played my idle game.
47.
by saretup 1y ago
What does 113% slowing mean? I thought if the speed is x then 80% slowing means the speed is now 0.2x
48.
by saretup 1y ago
Didn’t you know? Unbiased means what aligns with my point of view.
49.
by saretup 1y ago
Surprised to see no one here is worried about privacy. Z.ai's privacy policy pretty much allows them to store and use prompts/code in a perpetual, irrevocable, worldwide, sublicensable license to use and train their models.
50.
by saretup 1y ago
Interesting, although how hard is it to add a sorting functionality to the table?
51.
by saretup 1y ago
Exactly my thought. Just make it a desk toy.
52.
by saretup 1y ago
Well, this one is a bit too jarring to the ears.
53.
by saretup 1y ago
I thought he gave up on that when he couldn’t get AMD chips to run for long enough without bugs.
54.
by saretup 1y ago
What do you mean? An African or a European swallow?
55.
by saretup 1y ago
Doesn’t it just use the primary language you select in your account settings? Unless you’re talking about using it in incognito, in which case it does get annoying when it assumes a language based on region without asking.
56.
by saretup 1y ago
And while we’re at it, let’s move away from Python altogether. In the long run it doesn’t make sense just because it’s the language ML engineers are familiar with.
57.
by saretup 1y ago
As a non-US citizen, I heard it in the TV show ‘The Good Place’ as the official Bad Place song: https://youtu.be/iDjX4-LKqCA
58.
by saretup 1y ago
There are some really good open source TTS models out there now. Dia 1.6B or OpenAudio S1 are good options, and you can always check whichever models are trending on huggingface: https://huggingface.co/models?pipeline_tag=t
59.
by saretup 1y ago
I get why you added the static noise sound between clips but that gets annoying real quick.
60.
by saretup 1y ago
https://youtu.be/1NBfZcNU4O0

More ›