Eric J. Michaud
- Papers
- 5
Cite
Notes
Only stored in your browser.
5papers
Authored papers
5Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models
arXiv 2024
The Geometry of Concepts: Sparse Autoencoder Feature Structure
arXiv 2024
Not All Language Model Features Are Linear
arXiv 2024
Opening the AI black box: program synthesis via mechanistic interpretability
arXiv 2024
Towards Understanding Grokking: An Effective Theory of Representation Learning
arXiv 2022
Affiliations
No known affiliations.
Frequent co-authors
10from 5 papers