Sparse Autoencoders Find Highly Interpretable Features in Language Models
Paper • 2309.08600 • Published • 15
mechanistic interpretability with sparse autoencoders
A collection of papers that I found useful for learning about using Sparse Autoencoders for finding interpretable features in language models
-
-
Scaling and evaluating sparse autoencoders
Paper • 2406.04093 • Published • 4 -
Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models
Paper • 2403.19647 • Published • 4 -
Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2
Paper • 2408.05147 • Published • 41 -
Jumping Ahead: Improving Reconstruction Fidelity with JumpReLU Sparse Autoencoders
Paper • 2407.14435 • Published • 7 -
Interpreting Attention Layer Outputs with Sparse Autoencoders
Paper • 2406.17759 • Published -
Disentangling Dense Embeddings with Sparse Autoencoders
Paper • 2408.00657 • Published • 1 -
Sparse Autoencoders Enable Scalable and Reliable Circuit Identification in Language Models
Paper • 2405.12522 • Published • 2 -
Towards Principled Evaluations of Sparse Autoencoders for Interpretability and Control
Paper • 2405.08366 • Published • 2