Eric J. Michaud

29 papers A* 6Journal 23

Year	Rank	Type	Title / Venue / Authors
2025	A*	`conf`	Efficient Dictionary Learning with Switch Sparse Autoencoders. ICLR Anish Mudide, Joshua Engels, Eric J. Michaud, Max Tegmark, Christian Schröder de Witt
2025	A*	`conf`	Not All Language Model Features Are One-Dimensionally Linear. ICLR Joshua Engels, Eric J. Michaud, Isaac Liao, Wes Gurnee, Max Tegmark
2025	J	`jnl`	On the creation of narrow AI: hierarchy and nonlocality of neural network skills. CoRR Eric J. Michaud, Asher Parker-Sartori, Max Tegmark
2025	J	`jnl`	Open Problems in Mechanistic Interpretability. CoRR Lee Sharkey, Bilal Chughtai, Joshua Batson, Jack Lindsey, Jeff Wu, Lucius Bushnaq, Nicholas Goldowsky-Dill, Stefan Heimersheim, Alejandro Ortega, Joseph Isaac Bloom, Stella Biderman, Adrià Garriga-Alonso, Arthur Conmy, Neel Nanda, Jessica Rumbelow, Martin Wattenberg, Nandi Schoots, Joseph Miller, Eric J. Michaud, Stephen Casper, Max Tegmark, William Saunders, David Bau, Eric Todd, Atticus Geiger, Mor Geva, Jesse Hoogland, Daniel Murfet, Tom McGrath
2025	J	`jnl`	Open Problems in Mechanistic Interpretability. Trans. Mach. Learn. Res. Lee Sharkey, Bilal Chughtai, Joshua Batson, Jack Lindsey, Jeffrey Wu, Lucius Bushnaq, Nicholas Goldowsky-Dill, Stefan Heimersheim, Alejandro Ortega, Joseph Isaac Bloom, Stella Biderman, Adrià Garriga-Alonso, Arthur Conmy, Neel Nanda, Jessica Rumbelow, Martin Wattenberg, Nandi Schoots, Joseph Miller, William Saunders, Eric J. Michaud, Stephen Casper, Max Tegmark, David Bau, Eric Todd, Atticus Geiger, Mor Geva, Jesse Hoogland, Daniel Murfet, Tom McGrath
2025	J	`jnl`	Physics of Skill Learning. CoRR Ziming Liu, Yizhou Liu, Eric J. Michaud, Jeff Gore, Max Tegmark
2025	A*	`conf`	Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models. ICLR Samuel Marks, Can Rager, Eric J. Michaud, Yonatan Belinkov, David Bau, Aaron Mueller
2025	J	`jnl`	The Geometry of Concepts: Sparse Autoencoder Feature Structure. Entropy Yuxiao Li, Eric J. Michaud, David D. Baek, Joshua Engels, Xiaoqing Sun, Max Tegmark
2025	J	`jnl`	Understanding sparse autoencoder scaling in the presence of feature manifolds. CoRR Eric J. Michaud, Liv Gorton, Tom McGrath
2024	J	`jnl`	Efficient Dictionary Learning with Switch Sparse Autoencoders. CoRR Anish Mudide, Joshua Engels, Eric J. Michaud, Max Tegmark, Christian Schröder de Witt
2024	J	`jnl`	Not All Language Model Features Are Linear. CoRR Joshua Engels, Isaac Liao, Eric J. Michaud, Wes Gurnee, Max Tegmark
2024	J	`jnl`	Opening the AI Black Box: Distilling Machine-Learned Algorithms into Code. Entropy Eric J. Michaud, Isaac Liao, Vedang Lad, Ziming Liu, Anish Mudide, Chloe Loughridge, Zifan Carl Guo, Tara Rezaei Kheirkhah, Mateja Vukelic, Max Tegmark
2024	J	`jnl`	Opening the AI black box: program synthesis via mechanistic interpretability. CoRR Eric J. Michaud, Isaac Liao, Vedang Lad, Ziming Liu, Anish Mudide, Chloe Loughridge, Zifan Carl Guo, Tara Rezaei Kheirkhah, Mateja Vukelic, Max Tegmark
2024	J	`jnl`	Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models. CoRR Samuel Marks, Can Rager, Eric J. Michaud, Yonatan Belinkov, David Bau, Aaron Mueller
2024	J	`jnl`	Survival of the Fittest Representation: A Case Study with Modular Addition. CoRR Xiaoman Delores Ding, Zifan Carl Guo, Eric J. Michaud, Ziming Liu, Max Tegmark
2024	J	`jnl`	The Geometry of Concepts: Sparse Autoencoder Feature Structure. CoRR Yuxiao Li, Eric J. Michaud, David D. Baek, Joshua Engels, Xiaoqing Sun, Max Tegmark
2023	A*	`conf`	Omnigrok: Grokking Beyond Algorithmic Data. ICLR Ziming Liu, Eric J. Michaud, Max Tegmark
2023	J	`jnl`	Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback. CoRR Stephen Casper, Xander Davies, Claudia Shi, Thomas Krendl Gilbert, Jérémy Scheurer, Javier Rando, Rachel Freedman, Tomasz Korbak, David Lindner, Pedro Freire, Tony Tong Wang, Samuel Marks, Charbel-Raphaël Ségerie, Micah Carroll, Andi Peng, Phillip J. K. Christoffersen, Mehul Damani, Stewart Slocum, Usman Anwar, Anand Siththaranjan, Max Nadeau, Eric J. Michaud, Jacob Pfau, Dmitrii Krasheninnikov, Xin Chen, Lauro Langosco, Peter Hase, Erdem Biyik, Anca D. Dragan, David Krueger, Dorsa Sadigh, Dylan Hadfield-Menell
2023	J	`jnl`	Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback. Trans. Mach. Learn. Res. Stephen Casper, Xander Davies, Claudia Shi, Thomas Krendl Gilbert, Jérémy Scheurer, Javier Rando, Rachel Freedman, Tomasz Korbak, David Lindner, Pedro Freire, Tony Tong Wang, Samuel Marks, Charbel-Raphaël Ségerie, Micah Carroll, Andi Peng, Phillip J. K. Christoffersen, Mehul Damani, Stewart Slocum, Usman Anwar, Anand Siththaranjan, Max Nadeau, Eric J. Michaud, Jacob Pfau, Dmitrii Krasheninnikov, Xin Chen, Lauro Langosco, Peter Hase, Erdem Biyik, Anca D. Dragan, David Krueger, Dorsa Sadigh, Dylan Hadfield-Menell
2023	J	`jnl`	Precision Machine Learning. Entropy Eric J. Michaud, Ziming Liu, Max Tegmark
2023	A*	`conf`	The Quantization Model of Neural Scaling. NeurIPS Eric J. Michaud, Ziming Liu, Uzay Girit, Max Tegmark
2023	J	`jnl`	The Quantization Model of Neural Scaling. CoRR Eric J. Michaud, Ziming Liu, Uzay Girit, Max Tegmark
2022	J	`jnl`	Omnigrok: Grokking Beyond Algorithmic Data. CoRR Ziming Liu, Eric J. Michaud, Max Tegmark
2022	J	`jnl`	Precision Machine Learning. CoRR Eric J. Michaud, Ziming Liu, Max Tegmark
2022	A*	`conf`	Towards Understanding Grokking: An Effective Theory of Representation Learning. NeurIPS Ziming Liu, Ouail Kitouni, Niklas Nolte, Eric J. Michaud, Max Tegmark, Mike Williams
2022	J	`jnl`	Towards Understanding Grokking: An Effective Theory of Representation Learning. CoRR Ziming Liu, Ouail Kitouni, Niklas Nolte, Eric J. Michaud, Max Tegmark, Mike Williams
2020	J	`jnl`	Examining the Causal Structures of Deep Neural Networks Using Information Theory. Entropy Scythia Marrow, Eric J. Michaud, Erik Hoel
2020	J	`jnl`	Examining the causal structures of deep neural networks using information theory. CoRR Simon Mattsson, Eric J. Michaud, Erik Hoel
2020	J	`jnl`	Understanding Learned Reward Functions. CoRR Eric J. Michaud, Adam Gleave, Stuart Russell