CE-Bench: Towards a Reliable Contrastive Evaluation Benchmark of Interpretability of Sparse Autoencoders
Paper
•
2509.00691
•
Published
•
2
Alex Gulko
GulkoA
AI & ML interests
Interpretable AI, NLP
Recent Activity
upvoted
a
paper
about 2 months ago
CE-Bench: Towards a Reliable Contrastive Evaluation Benchmark of
Interpretability of Sparse Autoencoders
updated
a dataset
about 2 months ago
GulkoA/contrastive-stories-v4
updated
a dataset
about 2 months ago
GulkoA/contrastive-stories-v4
Organizations
authored
a
paper
2 months ago