Martin Wattenberg

Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task

Oct 25, 2022
Kenneth Li, Aspen K. Hopkins, David Bau, Fernanda Viégas, Hanspeter Pfister, Martin Wattenberg

Figure 2 for Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task

Figure 3 for Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task

Figure 4 for Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task

* code: https://github.com/likenneth/othello_world

Access Paper or Ask Questions

Toy Models of Superposition

Sep 21, 2022
Nelson Elhage, Tristan Hume, Catherine Olsson, Nicholas Schiefer, Tom Henighan, Shauna Kravec, Zac Hatfield-Dodds, Robert Lasenby, Dawn Drain, Carol Chen, Roger Grosse, Sam McCandlish, Jared Kaplan, Dario Amodei, Martin Wattenberg, Christopher Olah

* Also available at https://transformer-circuits.pub/2022/toy_model/index.html

Access Paper or Ask Questions

Interpreting a Machine Learning Model for Detecting Gravitational Waves

Feb 15, 2022
Mohammadtaher Safarzadeh, Asad Khan, E. A. Huerta, Martin Wattenberg

Figure 2 for Interpreting a Machine Learning Model for Detecting Gravitational Waves

Figure 3 for Interpreting a Machine Learning Model for Detecting Gravitational Waves

Figure 4 for Interpreting a Machine Learning Model for Detecting Gravitational Waves

* 19 pages, to be submitted, comments are welcome. Movies based on this work can be accessed via: https://www.youtube.com/watch?v=SXFGMOtJwn0 https://www.youtube.com/watch?v=itVCj9gpmAs

Access Paper or Ask Questions

An Interpretability Illusion for BERT

Apr 14, 2021
Tolga Bolukbasi, Adam Pearce, Ann Yuan, Andy Coenen, Emily Reif, Fernanda Viégas, Martin Wattenberg

Figure 2 for An Interpretability Illusion for BERT

Figure 3 for An Interpretability Illusion for BERT

Figure 4 for An Interpretability Illusion for BERT

Access Paper or Ask Questions

The What-If Tool: Interactive Probing of Machine Learning Models

Jul 09, 2019
James Wexler, Mahima Pushkarna, Tolga Bolukbasi, Martin Wattenberg, Fernanda Viegas, Jimbo Wilson

Figure 2 for The What-If Tool: Interactive Probing of Machine Learning Models

Figure 3 for The What-If Tool: Interactive Probing of Machine Learning Models

Figure 4 for The What-If Tool: Interactive Probing of Machine Learning Models

* 10 pages, 6 figures, 2 tables. To be presented at IEEE VAST 2019

Access Paper or Ask Questions

Visualizing and Measuring the Geometry of BERT

Jun 06, 2019
Andy Coenen, Emily Reif, Ann Yuan, Been Kim, Adam Pearce, Fernanda Viégas, Martin Wattenberg

Figure 2 for Visualizing and Measuring the Geometry of BERT

Figure 3 for Visualizing and Measuring the Geometry of BERT

Figure 4 for Visualizing and Measuring the Geometry of BERT

* 8 pages, 5 figures

Access Paper or Ask Questions

Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure

Mar 21, 2019
Been Kim, Emily Reif, Martin Wattenberg, Samy Bengio

Figure 2 for Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure

Figure 3 for Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure

Figure 4 for Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure

Access Paper or Ask Questions

TensorFlow.js: Machine Learning for the Web and Beyond

Jan 16, 2019
Daniel Smilkov, Nikhil Thorat, Yannick Assogba, Ann Yuan, Nick Kreeger, Ping Yu, Kangyi Zhang, Shanqing Cai, Eric Nielsen, David Soergel, Stan Bileschi, Michael Terry, Charles Nicholson, Sandeep N. Gupta, Sarah Sirajuddin, D. Sculley, Rajat Monga, Greg Corrado, Fernanda B. Viegas, Martin Wattenberg

Figure 2 for TensorFlow.js: Machine Learning for the Web and Beyond

Figure 3 for TensorFlow.js: Machine Learning for the Web and Beyond

Figure 4 for TensorFlow.js: Machine Learning for the Web and Beyond

* 10 pages

Access Paper or Ask Questions

Adversarial Spheres

Sep 10, 2018
Justin Gilmer, Luke Metz, Fartash Faghri, Samuel S. Schoenholz, Maithra Raghu, Martin Wattenberg, Ian Goodfellow

Access Paper or Ask Questions

GAN Lab: Understanding Complex Deep Generative Models using Interactive Visual Experimentation

Sep 05, 2018
Minsuk Kahng, Nikhil Thorat, Duen Horng Chau, Fernanda Viégas, Martin Wattenberg

Figure 2 for GAN Lab: Understanding Complex Deep Generative Models using Interactive Visual Experimentation

Figure 3 for GAN Lab: Understanding Complex Deep Generative Models using Interactive Visual Experimentation

Figure 4 for GAN Lab: Understanding Complex Deep Generative Models using Interactive Visual Experimentation

* This paper will be published in the IEEE Transactions on Visualization and Computer Graphics, 25(1), January 2019, and presented at IEEE VAST 2018

Access Paper or Ask Questions

1

2

>>