Position: Mechanistic Interpretability Should Prioritize Feature Consistency in SAEs
Paper
•
2505.20254
•
Published
•
5
Zeyu Tang
zeyutang
AI & ML interests
Trustworthy AI
Recent Activity
authored
a paper
6 days ago
Position: Mechanistic Interpretability Should Prioritize Feature
Consistency in SAEs
liked
a Space
9 months ago
Shaoan/ConceptGAN
authored
a paper
about 1 year ago
Model Transferability With Responsive Decision Subjects
Organizations
None yet
zeyutang's activity
authored
a
paper
6 days ago
liked
a
Space
9 months ago