1681
Segmenting comic book frames
This post is based on my project in my Computer Vision class last semester
New
The latest blog posts to take off on Hacker News. 3,091 posts.
1682
Gemini and Google's Culture
The Google Gemini fiasco shows that the biggest challenge for Google in AI is not business model but rather company culture; change is needed from the top down.
1683
Mamba Explained: The State Space Model Taking On Transformers
The State Space Model taking on Transformers
1684
Does offering ChatGPT a tip cause it to generate better text?
Modern AI rewards being very weird.
1685
GPT in 500 Lines of SQL
A complete GPT2 implementation as a single SQL query in PostgreSQL.
1686
Meta's new LLM-based test generator
Meta's TestGen-LLM is a sneak peek to the future of developer productivity: specialized, orchestrated, and rigorously filtered.
1687
Mamba: The Easy Way
An overview of the big ideas behind Mamba, a brand-new language model architecture.
1688
Building a deep learning rig
Blog post of Samsja
1689
The killer app of Gemini Pro 1.5 is using video as an input
Last week Google introduced Gemini Pro 1.5, an enormous upgrade to their Gemini series of AI models. Gemini Pro 1.5 has a 1,000,000 token context size. This is huge—previously that …
1690
Things I don't know about AI
The more I learn about AI markets, the less I think I know. I list questions and some thoughts.
1691
1692
Planner programming blows my mind
Picat is a research language intended to combine logic programming, imperative programming, and constraint solving. I originally learned it to help with vacation scheduling but soon discovered its planner module, which is one of the most fascinating programming models I’ve ever seen. First, a brief explanation of…
1693
Microsoft is spying on users of its AI tools
Microsoft announced that it caught Chinese, Russian, and Iranian hackers using its AI tools—presumably coding tools—to improve their hacking abilities. From their report: In collaboration with OpenAI, we are sharing threat intelligence showing detected state affiliated adversaries—tracked as Forest Blizzard, Emerald…
1694
AI your home on street view
Maps Mania is a blog dedicated to tracking the very best digital interactive maps on the internet and the tools used to create them.
1695
The AI bullshit singularity
I’m sure we are all familiar with the idea of a technological singularity. Humans create an AI that is smart enough to create an even smarter successor. That successor then creates an even sm…
1696
1697
Representation Engineering: Mistral-7B on Acid
Playing around with the Representation Engineering paper, I made some interesting control vectors, and a Python package to make your own.
1698
Altman's Ambition
Sam Altman is not playing around.
1699
I worry our Copilot is leaving some passengers behind
Why I worry about the effect GitHub Copilot is having and will continue to have on the accessibility of the web at scale.
1700
Raising Children on the Eve of AI
How do we prepare them for what we're not prepared for?
1701
I'm an Old Fart and AI Makes Me Sad
Previous major advancements in Computer Science and its applications always felt empowering to me, AI feels different.
1702
Building an LLM from Scratch: Automatic Differentiation (2023)
Automatic Differentiation
1703
Show me the prompt
Quickly understand inscrutable LLM frameworks by intercepting API calls.
1704
Neural network training makes beautiful fractals
This blog is intended to be a place to share ideas and results that are too weird, incomplete, or off-topic to turn into an academic paper, but that I think may be important. Let me know what you think! Contact links to the left.
1705
The story behind the NeuralRad organ and tumor segmentation cloud service
Late last month, I received a message.
1706
Kalman Filter Explained Simply
Tired of equations and matrices? Ready to learn the easy way? This post explains the Kalman Filter simply with pictures and examples!
1707
Thinking about high-quality human data
[Special thank you to Ian Kivlichan for many useful pointers (E.g. the 100+ year old Nature paper “Vox populi”) and nice feedback. 🙏 ] High-quality data is the fuel for modern data deep learning model training. Most of the task-specific labeled data comes from human annotation, such as classification task or RLHF…
1708
Demystifying GPU compute architectures
Getting 'low level' with Nvidia and AMD GPUs
1709
1710
Beyond self-attention: How a small language model predicts the next token
A deep dive into the internals of a small transformer model to learn how it turns self-attention calculations into accurate predictions for the next token.