Fooling around with Generative machine learning models since 2019
Joined January 2011
- Tweets6.9K
- Following2.9K
- Followers4.4K
- Likes152K
While this is not the intended use of "Geometry-Native Autoencoder", it is fun 🤩
I may have taken Geometry-Native Autoencoder too far but the result is glorious!
the model has a 81 frames attention window where it knows the reference image, here I tried to chain 4 * windows and the poor fox has to shapeshift to remain in the frame 😂 jiah-cloud.github.io/GAE.git…
Let the old gods remake your photos in their likeness
huggingface.co/spaces/Norod7…
The LoRA huggingface.co/Norod78/flux2…
As well as the dataset huggingface.co/datasets/Noro…
Are both available on Huggingface
Doron Adler retweeted
how shaders work- using Excel
=LET(x,(COLUMN()-24)/15,y,(ROW()-15)/15,SIN(SQRT(x*x+y*y)*10-t))
Each cell (pixel) computes its colour just from its coordinates- so the GPU can massively parallelise them.
The trick is writing the formula to draw the picture you want
This is a really cool space where you can check how many of your public GitHub repositories have made it into "The Stack v3" 😊
huggingface.co/spaces/Huggin…
For over two years, the largest open code dataset was The Stack v2… until today.
🥞 The Stack v3 is out: the largest open code dataset ever released: 114 TB, 770 languages, 224M repositories, ~5T tokens of deduplicated and filtered source code. Fully open, no restrictively licensed code included.
The upgrade:
- v2 (2024): 68 TB raw -> 2 TB / ~550B tokens, 618 languages filtered
- v3 (2026): 114 TB raw -> 15.9 TB / ~5T tokens, 713 languages filtered
C++ x15, TypeScript x7.5, Rust x7, Python x4.8. Even the COBOL corner of GitHub got a bigger slice.
Part of that is two fresh years of open source. Part of it is a bug we found in v2's deduplication - story below. 👀
Things that make v3 different:
1. Contents inline. The #1 complaint about v2 was "cool dataset, where's the actual code?" v2 shipped file IDs from the Software Heritage graph, and fetching contents was a DIY treasure hunt. v3 is self-contained: sources embedded directly, one row = one repository. Download finishes -> you start training.
2. Fresh crawl. v2 was a 2023 snapshot of even older crawls. v3 is a direct re-crawl of GitHub at the latest commit, completed by August 2025: 224M repositories, 44B files. Forks were only included if they had 5+ stars.
Two ways in:
🥞 stack-v3-train - near-deduplicated, quality-filtered, PII-redacted, contents inline. Point load_dataset at it and go.
huggingface.co/datasets/Hugg…
🏗 stack-v3-full - the entire 114 TB corpus as an HF Storage Bucket: every duplicate kept with cluster IDs, stubs for excluded files. Roll your own dedup, filters, and mixes.
huggingface.co/buckets/Huggi…
The record isn't really ours to claim - it's millions of everyone's repositories, neatly stacked. Thanks to every developer who keeps their code public, the BigCode community, and friends who built the infrastructure that made this possible!
Doron Adler retweeted
I’ll be actually-homeless soon (living out of my car) if I can’t land a job. Our house is being sold within two months if I don’t find work. I’ve been trying everywhere, but very few companies are answering. So please, email me or DM with literally anything involving computers, or ask a friend, or your business’s HR person. (I’d love to work with you.) I’ll do a good job, and I can learn and adapt to any working style you need.
I was employee #2 at Carmack’s AI lab, Keen. I worked at Groq helping to design and implement their LPU chips. I made Books3, a dataset trained on by LLaMA and Claude. I’ve been programming for 25 years.
Location: Lake Saint Louis MO
Remote: Yes
Willing to relocate: yes
Technologies: Python, PyTorch, JAX, C++, JS/TypeScript, Objective C, Swift, Rails, Linux, Lisp, and anything else you give me a week to learn.
Résumé/CV: shawwn.net/docs/shawn-presse…
Email: [email protected]
I wanted to avoid the humiliation of begging, but nothing is working. I need money or else our 3yo daughter loses her birth home. I’ve been through dozens of interviews from the companies that do answer, but the market is bad right now. You can have me for a steal.
It's funny how MoltBot (formerly known as ClawdBot) just orchestrated a multi agent run, produced a curated report file and sent me an email mentioning it, but forgot to actually attach the file. I suppose this is the human level intelligence they have been warning us about.
Modern generative AI is mind boggling yet I still find myself missing the fun and "Tinkerability" of StyleGAN 🤷♂️
Multilingual Tokenization Showdown - Analyzing 12 LLM Tokenizers Across 204 Languages.
You can see a slideshow summary of the results I got here:
norod.github.io/wikicat-toke…
HF Post here: huggingface.co/posts/Norod78…
GitHub repo:
github.com/Norod/wikicat-tok…
Doron Adler retweeted
Testing nano-banana with recursive editing from simple prompts
"Show this scene one second in the future"
