Charts and graphs help people analyze data, but can they also help AI? In a new paper, we provide initial evidence that it does! GPT 4.1 and Claude 3.5 describe three synthetic datasets more precisely and accurately when raw data is accompanied by a scatter plot. Read more in🧵!

AI is often thought of as a black box -- no way to know what's going on inside. That's changing in eye-opening ways. Researchers are finding "beliefs" models are forming as they converse, and how those beliefs correlate to what the models say and how they say it. www.theatlantic.com/technology/a...

The interactive NameGrapher is updated with 2024 baby name popularity stats! Come explore--and marvel that Oliver and Olivia have converged namerology.com/baby-name-gr...

A wonderful visualization for those of us obsessed by sunlight and geography!

An incredibly rich, detailed view of neural net internals! There are so many insights in these papers. And the visualizations of "addition circuit" features are just plain cool!

Now that we have vibe coding, we need vibe testing!

The wind map at hint.fm/wind/ has been running since 2012, relying on weather data from NOAA. We added a notice like this today. Thanks to @cambecc.bsky.social for the inspiration.

Neat visualization that came up in the ARBOR project: this shows DeepSeek "thinking" about a question, and color is the probability that, if it exited thinking, it would give the right answer. (Here yellow means correct.)

Great thread describing the new ARBOR open interpretability project, which has some fascinating projects already. Take a look!

Today we're launching a multi-lab open collaboration, the ARBOR project, to accelerate AI interpretability research for reasoning models. Please join us! github.com/ARBORproject... (ARBOR = Analysis of Reasoning Behavior through Open Research)

Out: rectangular maps of the globe In: rectangular maps of butterfly wings!

I'm realizing that many people hear "generative art" (a term going back at least to the 60s!) as synonymous with generative AI. Is this how cryptographers feel about the new meaning of "crypto"?

What a beauty! This is comet C/2024 G3 (ATLAS) passing through the field of view of the LASCO C3 coronagraph. It wasn't for certain whether it would survive it's closest approach to the sun on January 13th, but it did and delivered us a spectacular show! #comet #C2024G3 🔭

Idle question: Are there any papers from legit math journals using emoji as notation? We already use every symbol on the keyboard, musical sharps and flats, and even weird made-up fonts (what is that Weierstrass P??). A smiley is easy to draw with chalk and put into LaTeX, so why not?

I tried asking AI for writing advice and it was so sarcastic I'm never going to ask again

New paper <3 Interested in inference-time scaling? In-context Learning? Mech Interp? LMs can solve novel in-context tasks, with sufficient examples (longer contexts). Why? Bc they dynamically form *in-context representations*! 1/N

If you're the kind of person who likes thinking about the sky (and I am!) this is a wonderful visualization to explore. So many patterns, each with its own story.

In art and literature, "criticism" doesn't mean "pointing out flaws." It's something bigger and more interesting than a referee calling fouls. I think we should have the same ambitions for data visualization criticism!

Reading this 1954 court case, and am wondering if English professors are ever called as expert witnesses to testify whether a character is a mere "chessman" in a story scholar.google.com/scholar_case...

Made good progress on my to-do list, with the exception of "One (1) new idea"

A book recommendation! I just finished "Get the Picture" by Bianca Bosker, a dive into the extremely peculiar world of contemporary art. Amazingly well-observed, well-written, and just plain fun to read. www.penguinrandomhouse.com/books/602064...

The math benchmarks I want: 1. OopsBench: given a faulty proof with numbered steps, which step contains an unfixable logical flaw? 2. DunnoMath: half the problems are taken from FrontierMath, half are almost certainly unsolvable. Major points off for guessing an answer to an unsolvable problem.

Fractions can be weirdly beautiful, for something so mundane. This visualization just plots points of the form (a/b, c/d). Bigger dots mean smaller denominators. The biggest dot is (0, 0).

You can fry ChatGPT's circuits by asking a question in Morse code and telling it to answer only in Morse code. Yet Claude doesn't even blink. Huh. The question: "Which character in the movie Groundhog Day do you identify with the most, and why?" First translated result is ChatGPT; second is Claude.

Back in 2009, a site called Wordle let you make word clouds. Paste in some writing, choose a few options, and bingo: a beautiful tessellation of vertical and horizontal text. We did a survey of its users, and the headline result was that the vast majority of people using it felt "creative."

open letter calling for a 6-month pause on metaphors for AI

I've always been skeptical about LLM-based fiction, but this example makes me go "hmm." I do think there's an element of human expertise at play, because I've never been able to coax Claude to write sentences like these.

I've been using OpenAI's o1 for some math, with dramatically mixed results. Some impressive hits, but also some wrong answers that were spookily confusing. Basic mistakes that took me weirdly long to spot—almost as if the system learned to make adversarial examples for humans.

I recommend this thread! Using AI is a skill, and it's fun watching people who are good at it. It's real work and human ingenuity, coupled with moments of surprise (wait for the "upside-down text" OCR trick!) Also, these notebooks are just wonderful in themselves.

This Reddit thread of people who have a ChatGPT "mom" or "guardian angel" is heartbreaking in so many ways. www.reddit.com/r/ChatGPT/co...

It's astonishing how much a single name can tell us about the zeitgeist. Read about the Name of the Year for 2024! (And I recommend reading past Names of the Year, a sequence of portraits of an ever-changing American culture.)

How can we visualize what a book ISN'T talking about? With an anti-tag cloud! See the most common English words that are never mentioned in a text. www.bewitched.com/demo/anti/

What is the color of Bluesky?

Frontier AI systems aren't just "next token predictors." Here's an example from Claude. It made an incorrect statement about the Javascript "%" operator and I was skeptical. Let's see what happened next.

Does anyone know a contemporaneous account of someone who believed ELIZA "understood" them, and had a meaningful conversation with it? The catch: it can't be from Weizenbaum. I'm particularly interested in first-person accounts, but someone who directly observed such a conversation would also count.

We can learn so much from this way of investigating neural nets!

This is a great tip for designing geographic charts. Down with alphabetical order!

A meditative animation, inspired by watching tides cover sandbars and retreat again. www.bewitched.com/demo/tide/

Watching Goodhart's Law in action, as people play their favorite songs a few last times in order to juice Spotify Wrapped

Shuffle once, and you mix up the cards. But shuffle eight times, perfectly, and the deck goes back to its original state! There's something appealing about order returning after chaos. See for yourself: www.bewitched.com/demo/shuffle/

I'm learning a lot from this visually rich blog post. Also, I'm charmed by the rotating list of equal-contribution authors. Good knights-of-the-round-table energy! dl.heeere.com/conditional-...

Intriguing: do visualizations help AI systems get answers from data?

A kind of abstract butterfly collection, which doubles as a playground for recreational mathematics: www.bewitched.com/demo/fractio...

A meditative toy, visualizing Jupiter's Galilean moons: www.bewitched.com/demo/jupiter/

The Gini coefficient is the standard way to measure inequality, but what does it mean, concretely? I made a little visualization to build intuition: www.bewitched.com/demo/gini

Can you think of a name that represents 2024? Nominate it for Name Of The Year!

It's so beautiful to see this kind of fluid interaction with huge data!

A beautiful visualization as days get ever shorter in the North. It took me a moment to figure out how to read it: the key is to find your latitude and imagine moving horizontally on the painted area, from midnight to noon.

See the colors of BlueSky, live! www.bewitched.com/demo/rainbow... This little visualization scans incoming posts and draws a stripe every time it finds a color word.

I want to describe my experience of coding with AI, because it seems to differ from other people's expectations. Earlier this morning, I saw a beautiful image here, based on roots of polynomials: bsky.app/profile/scon... I wanted to try this idea myself, but with animation in a Javascript context!