61
62
Anthropic Claude 3.5 can create icalendar files, so I did this
Generate calendar entries from an image using Anthropic Claude 3.5. Convert lesson dates in an image to a calendar ICS file
63
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…
64
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!
65
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.
66
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.
67
A ChatGPT mistake cost us $10k
Edit: I want to preface this by saying yes the practices here are very bad and embarrassing (and we've since added robust unit/integration tests and alerting...
68
Building an LLM from Scratch: Automatic Differentiation (2023)
Automatic Differentiation
69
Google's First Tensor Processing Unit: Architecture
Looking in more detail at Google's first Tensor Processing Unit
70
A ChatGPT clone, in 3000 bytes of C, backed by GPT-2 (2023)
This program is a dependency-free implementation of GPT-2, including byte-pair encoding and transformer inference, in ~3000 bytes of C. I then use this to create something like Chat GPT.
71
Claude's API now supports CORS requests, enabling client-side applications
Anthropic have enabled CORS support for their JSON APIs, which means it’s now possible to call the Claude LLMs directly from a user’s browser. This massively significant new feature is …
72
Fast LLM Inference From Scratch (using CUDA)
This post is about building an LLM inference engine using C++ and CUDA from scratch without libraries.
73
The OpenAI board was right
Even on little things, Sam is not consistently candid.
74
How Does GPT-4o Encode Images?
Here’s a fact: GPT-4o charges 170 tokens to process each 512x512 tile used in high-res mode. At ~0.75 tokens/word, this suggests a picture is worth about 227 words—only a factor of four off from the traditional saying. (There’s also an 85 tokens charge for a low-res ‘master thumbnail’ of each picture and higher…
75
My Python code is a neural network
This post translates a Python program to a recurrent neural network. It visualizes the network and explains each step of the translation in detail.
76
77
OpenAI’s board, paraphrased: ‘All we need is unimaginable sums of money’
OpenAI is to this decade’s generative-AI revolution what Netscape was to the 1990s’ internet revolution.
78
YC is wrong about LLMs for chip design
LLMs don’t make chips better. Good engineers do.
79
Probably pay attention to tokenizers
Last week I was helping a friend of mine to get one of his new apps off the ground. I can’t speak much about it at the moment, other than like most apps nowadays it has some AI sprinkled over it. Ok, maybe a bit maybe more just a bit – depends on the way you look at it, I suppose. There is a Retrieval-augmented…
80
81
82
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.
83
Diffusion Models
I spent 2022 learning to draw and was blindsided by the rise of AI art There is lots to say about this, whether it has been or will be a good thing for artists and society in the long run. I hope to write about it in another post. models like Stable Diffusion. Suddenly, the computer was a better artist than I could…
84
Audio Decomposition – open-source seperation of music to constituent instruments
Matthew Bird Portfolio
85
A Visual Guide to LLM Quantization
Exploring memory-efficient techniques for LLMs
86
Video scraping: extracting JSON from a 35s screen capture for 1/10th of a cent
The other day I found myself needing to add up some numeric values that were scattered across twelve different emails. I didn’t particularly feel like copying and pasting all of …
87
Interview with gwern
Legacy & Anonymity in the Age of AGI
88
The Illustrated AlphaFold
A visual walkthrough of the AlphaFold3 architecture, with more details and diagrams than you were probably looking for.
89
The deep learning boom caught almost everyone by surprise
"You’ve taken this idea way too far," a mentor told Prof. Fei-Fei Li.
90
