neodypsis

6 ms·

1.
by neodypsis 8mo ago
I looked at the trails (lists of passages) and wondered whether it would be cheaper to build these lists by clustering passage embeddings, rather than having an LLM construct them.
2.
by neodypsis 1y ago
What is the steam cloud for after landing?
3.
by neodypsis 1y ago
Try one of the models with good vision capabilities and ask it to output code using build123d.
4.
by neodypsis 1y ago
Not true. It depends on the pathogen.
5.
by neodypsis 1y ago
You need to add sanitizer to the wash cycle, not just detergent.
6.
by neodypsis 1y ago
I like Gemini for "architect" roles, it has very good code recall (almost no hallucinations, or none lately), so it can successfully review code edits by Claude. I also find it useful to ground it with Google Search.
7.
by neodypsis 1y ago
Gemini's context is very long, so I can feed it full files. I do the same with Claude, but I may need to start from scratch various times, so Gemini serves as memory (and is also good that Gemini has almost no hallucinations, so it&#x2
8.
by neodypsis 1y ago
A combination that works nicely to solve bugs is: 1) have Gemini analyze the code and the problem, 2) ask it to create a prompt for Claude to fix the problem, 3) give Claude the markdown prompt and the code, 4) give Gemini the output from C
9.
by neodypsis 1y ago
Were you able to find the tracks using Shazam? It is having trouble matching the one from the 28:00 mark here: https://archive.org/details/KmartFebruary1990
10.
by neodypsis 1y ago
lol, never expected that to happen here
11.
by neodypsis 1y ago
Thanks for the reply, yeah, my interest was to test compatibility when we add support for Play Integrity to our app.
12.
by neodypsis 1y ago
Is there a GrapheneOS image for use with Android Emulator? I want to test an app on it but got no physical Pixel device.
13.
by neodypsis 1y ago
I've noticed 4o uses a lot of emojis, and, in general, is very enthusiastic. I find it funny. If I want a more formal bot, I switch to one of the o3 family.
14.
by neodypsis 2y ago
Perhaps I disagree on the following: > There’s no customer support, chatbot, or extensive documentation. I've seen many vending machines with customer support channels. If the machine takes the money but merchandise is not delivered
15.
by neodypsis 2y ago
so this "vending machine" is a postcard drop shipping website?
16.
by neodypsis 2y ago
It is mentioned on section "Limitations and Bloopers" of the page [0]: > Combining evolutionary optimization with LLMs is powerful but can also find ways to trick the verification sandbox. We are fortunate to have Twitter user
17.
by neodypsis 2y ago
Interestingly, o3-mini-high was correct when first thinking about it: > Okay, we're asked how to get exactly 6 liters of water using an 12-liter and a 6-liter jug. The immediate thought is to just fill the 6-liter jug, but that seem
18.
by neodypsis 2y ago
How does it compare to Jina V3 [0], which also has 8192 context length? 0. https://arxiv.org/abs/2409.10173
19.
by neodypsis 2y ago
Isn't that what happened to one of these guys? It is reported that one of them caught the fungus due to inspecting his attic (hadn't used the guano yet).
20.
by neodypsis 2y ago
I assume must feces require prior treatment before they can be used as fertilizer.
21.
by neodypsis 2y ago
Reportedly, one of them had a bat infestation that produced a thick layer of guano in his attic, which he planned to use as fertilizer. He probably should have inspected his attic wearing a N95 respirator.
22.
by neodypsis 2y ago
> Schneier’s book? Yes
23.
by neodypsis 2y ago
> I mind that resources about how to use crypto in software applications are often inscrutable, all the way down to library design, for no good reason. I haven't read it, but I plan to, eventually. There's a book titled "C
24.
by neodypsis 2y ago
Can a ramp pump be used as a booster pump?
25.
Electricity Cuts to Affect Costa Rica Next Week, Confirms ICE(ticotimes.net)
7 points by neodypsis 2y ago|1 comments
26.
by neodypsis 2y ago
Someone should finetune an LLM to create a "patentese" assistant writer.
27.
by neodypsis 2y ago
Reportedly, the ADC's of the ESP32 are unreliable.
28.
by neodypsis 2y ago
> Seems like we still have a long way to go after Adam... A preprint in arxiv suggests that Adam works better than SGD for training LLMs due to the issue of class-imbalance [0]. It appears that scaling the gradient step helps with the t
29.
by neodypsis 2y ago
> I know the one I make isn’t going to be as precise or accurate, the build quality won’t be as good, but it’ll be good enough for my purposes. Have you had issues with the ESP32's ADCs?
30.
by neodypsis 2y ago
Can I read this on Kindle?

More ›