lsb

7 ms·

31.
Vibe Scraping / Vibe Coding a schedule app on a phone(simonwillison.net)
2 points by lsb 1y ago|0 comments
32.
by lsb 2y ago
I’ve been running half-a-billion parameter models comfortably in a web browser, especially with WebGPU, and you can definitely run billion parameter LLMs in the browser. It becomes a heavyweight browser app, but if the main costs are runnin
33.
by lsb 2y ago
I made an AI Art clock, using Stable Diffusion to render a 24-hour clock face into a landscape: https://leebutterman.com/diffusion-local-time/
34.
by lsb 2y ago
Clojure, for example, uses Hash Array Mapped Tries as its associative data structure, and those work well
35.
by lsb 2y ago
All New Yorkers pay 80% of the cost to run the bus before they even think about boarding. (“Farebox recovery ratio”) This is an excuse to fund more cops. Transit should be free, like sidewalks and parks.
36.
by lsb 2y ago
The history is still there, in archives, and modern usage is what we’re talking about
37.
by lsb 2y ago
The model rescinded her consent to be used as a technical benchmark photo, a while ago
38.
by lsb 2y ago
Doesn’t Google Cloud’s AI infrastructure include Colab? That’s useful for so many things
39.
The six dumbest ideas in computer security (2005)(ranum.com)
267 points by lsb 2y ago|284 comments
40.
by lsb 2y ago
Default permit, enumerating badness, penetrate and patch, hacking is cool, educating users, action is better than inaction
41.
by lsb 2y ago
Of the “six famous bad ideas in computer security”, the first and second are “default permit” and “enumerating badness”. http://www.ranum.com/security/computer_security/editorials/d...
42.
by lsb 2y ago
Dasung has eink HDMI monitors that have quite a high refresh rate. For a 25” monitor it runs about $1700. https://shop.dasung.com/
43.
by lsb 2y ago
Looks similar to the Dasung monitor latency. Would be interested to see a comparison, both for features and performance and price
44.
by lsb 2y ago
Exactly, like how a “bitcoin mining” chip would implement the SHA in hardware. And, CPUs prioritize hiding latency with all sorts of caches, and GPUs prioritize cores and bandwidth to hide latency, so there’s different tradeoffs about memor
45.
by lsb 2y ago
The earliest Intel chips (8086) had an extra add-on to do floating point math in hardware, instead of software (8087). There’s a lot of “AI” workloads that include a lot of matrix multiplication, which the Google TPU and the Nvidia tensor c
46.
by lsb 2y ago
Would this implement indexing strategies like HNSW? Linear scan is obviously great to start out with, and can definitely be performant, especially if your data is in a reasonable order and under (say) 10MB, so this shouldn't block a be
47.
by lsb 2y ago
It depends on your data and your embedding model. For example, I was able to quantize embeddings of English Wikipedia from 384-dimensions down to 48 7-bit dimensions, and the search works great: https://www.leebutterman.com/
48.
by lsb 2y ago
Binary, quantize each dimension to +1 or -1 You can try out binary vectors, in comparison to quantize every pair of vectors to one of four values, and a lot more, by using a FAISS index on your data, and using Product Quantization (like PQ7
49.
by lsb 3y ago
Yes, these are “adversarial patches” in image classification, like https://arxiv.org/abs/1712.09665 . Similarly you can take these adversaries and add them to your own larger model, in an arms race.
50.
by lsb 3y ago
You’re looking for what’s called upscaling, like with Stable Diffusion: https://huggingface.co/stabilityai/stable-diffusion-x4-upsca...
51.
by lsb 3y ago
It’s an absolute scandal, as if France only sold wine grapes, and no one in France had ever made or tasted wine. Most of the value add is in the processing.
52.
Diffusion Local Time, a timepiece made of generative AI art, at Art Hack Day(leebutterman.com)
1 points by lsb 3y ago|0 comments
53.
by lsb 3y ago
This is an example of the Trust On First Use policy, like when you SSH to a machine whose cert you don't have and you are invited to trust it. https://en.wikipedia.org/wiki/Trust_on_first_use
54.
by lsb 3y ago
The iOS app “MLCChat”
55.
by lsb 3y ago
I can run Mistral 7B, in 3-bit quantization, on my phone, and I get helpful answers for (just today) seeing environment variables in Python.
56.
by lsb 3y ago
That's not _nearly_ the most unusual unit of measurement one could find https://en.wikipedia.org/wiki/List_of_unusual_units_of_measu...
57.
by lsb 3y ago
That’s why I made the only-urban clock face, or the only-train clock face :)
58.
by lsb 3y ago
I can easily generate one every few minutes with a few CPU cores on my desktop, could be a cute dynamic background :) it’s a delicate balance to run on stable enough hardware that has low latency. The Ampere card that’s running this is tak
59.
by lsb 3y ago
All the models are available on Huggingface, happy to answer any questions
60.
Show HN: Stable Diffusion Clock(leebutterman.com)
4 points by lsb 3y ago|6 comments

More ›