Eric J Michaud
4 papers · 2023–2025 · 1 venue · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Venues
Top co-authors
Research topics
Keywords
representation learning
(2)
sparse autoencoder
(2)
feature extraction
(1)
loss landscape
(1)
neural network activation
(1)
language model
(1)
mixture of expert
(1)
modular arithmetic
(1)
feature interpretability
(1)
neural network
(1)
feature ablation
(1)
algorithmic dataset
(1)
dictionary learning
(1)
circular feature
(1)
causal inference
(1)
Papers
Efficient Dictionary Learning with Switch Sparse Autoencoders
ICLR 2025
Not All Language Model Features Are One-Dimensionally Linear
ICLR 2025
Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models
ICLR 2025
Omnigrok: Grokking Beyond Algorithmic Data
ICLR 2023