You are browsing as a guest.
Sign up
(or log in) to start making projects!
Open comments for this post
- showed model performance before the interpretability analysis.
- separated task performance from probe results.
- combined all interpretability analyses into one pipeline.
- made CTCF the main case study.
- made the overall framing clearer.
- explained every analysis threshold.
- validated thresholds with sensitivity tests and control experiments.
- showed that the main results stay consistent across different thresholds.
Comments 0
No comments yet. Be the first!
Sign in to join the conversation.