Martin Wattenberg
@wattenberg
Followers
17K
Following
2K
Media
84
Statuses
733
Human/AI interaction. Visualization as design, science, art. Professor at Harvard, and part-time at Google's People+AI Research initiative.
Joined July 2008
Here's a close-up of hue and brightness heads in a vision transformer. Work done with tremendous help from Aoyu Wu and @chenxcynthia!
0
2
14
Language models have some beautiful spiral plots reflecting positional patterns. And a vision model has heads that arrange images according to brightness and hues. But there’s a lot more to find! What else can you see?
1
4
22
Visualize transformer attention!. AttentionViz, created by Catherine Yeh and expanded by Yida Chen, helps you explore transformer self-attention by visualizing query and key vectors in a joint embedding. Paper: Website:
4
85
393
Toasters have blinking lights, cars have speedometers. Should chatbots have dashboards too?. A speculative essay: The System Model and the User Model: Exploring AI Dashboard Design.
1
4
25
RT @davidbau: I want to show the NSF there would be broad support+utility for a "National Deep Inference" service for >100b LLMs. If your….
0
52
0
RT @hseas: Students explore the aesthetics of computing in new computer science course at SEAS. "CS73: Code, Data, and Art" is co-taught by….
0
9
0
Thinking about grad school next year? Interested in visualization, machine learning interpretability, or human/AI interaction? Consider Harvard. @viegasf and I are continuing to build our lab!
1
18
111
RT @guardian: The labyrinthine patterns traced by birds on the wing – in pictures
theguardian.com
Spanish photographer Xavier Bou’s images of the flight paths of birds aim to make us look up at the wonder in the skies
0
15
0
I'm teaching with @OpenProcessing for the first time, and am completely impressed with how polished and friendly the system is. Every detail is on point. Last class a student spontaneously said, "OpenProcessing is just so great." I agree!.
1
1
23
The project is a systematic study of how a neural network represents more features than it has neurons, under a variety of conditions. (Oh, and I can't tweet without a typo, apparently. That's @AnthropicAI of course!).
1
1
17
Tiny neural networks have a surprisingly rich inner life, and may hold clues to how their larger cousins work. This image: evolution of feature vectors during learning. Full story: (in collaboration with the great interpretability team at @AnthopicAI)
11
89
695
Replies are showing a rough consensus of red/brown for the past, and blue/green for the future. But not everyone agrees, and that consensus hides wildly different metaphors. Fall foliage, receding galaxies, old photos, atmospheric fog. .
1
0
6
What are good chart colors to represent past, present, and future? Not looking to "gray out" a prediction, but just represent time.
15
1
26
All excellent images, but the "cyclops" picture is just startlingly good. Kudos for a brilliant prompt.
0
1
6
The new baby name data is out! Check out @NamerologyTalk to find out the latest trends. What alcoholic drink appeared as a name for the first time? Which famous robot is erasing a name? Which steamy series had the biggest effect?.
1
0
8
RT @NamerologyTalk: The year's biggest baby name style influencer: Raya and the Last Dragon. In a landmark for Asian characters, Raya was A….
0
2
0