Features that Make a Difference: Leveraging Gradients for Improved Dictionary Learning

Sparse Autoencoders (SAEs) are a promising approach for extracting neural network representations by learning a sparse and overcomplete decomposition of the network's internal activations. However, SAEs are traditionally trained considering only activation values and not the effect those activa...

Ausführliche Beschreibung

Bibliographische Detailangaben

Hauptverfasser:	Olmo, Jeffrey, Wilson, Jared, Forsey, Max, Hepner, Bryce, Howe, Thomas Vin, Wingate, David
Format:	Artikel
Sprache:	eng
Schlagworte:	Computer Science - Artificial Intelligence Computer Science - Computation and Language Computer Science - Learning
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!