Sparse Autoencoders Encode Both Concepts and Functions: The Downstream Geometry of Feature Effects
Paper • 2607.24645 • Published • 1
Anwoy Chatterjee
anwoy
AI & ML interests
Natural Language Processing, Deep Learning, Machine Learning
Recent Activity
authored a paper 3 days ago
On the Effect of Instruction Tuning Loss on Generalization authored a paper 3 days ago
Sparse Autoencoders Encode Both Concepts and Functions: The Downstream Geometry of Feature EffectsOrganizations
None yet
upvoted 2
papers 3 days ago