William Dorrell
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
Neurons and Cognition×1ML×1
Research Timeline
2026
How Optimality Structures Sparse Dictionaries: A Theory for Understanding SAE Representations
The paper theoretically analyzes the properties that optimal sparse autoencoder (SAE) dictionaries must satisfy, deriving constraints that explain observed SAE behaviors like hierarchical splitting and residual structure.
Highlighted terms show continued research focus across papers