Toggle Poster Visibility
Oral
Fri Apr 25 07:30 PM -- 07:42 PM (PDT) @ Garnet 216-218 None
Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models
[
Slides]
[
OpenReview]
Oral
Fri Apr 25 07:42 PM -- 07:54 PM (PDT) @ Garnet 216-218 None
Unlearning-based Neural Interpretations
[
OpenReview]
Oral
Fri Apr 25 07:54 PM -- 08:06 PM (PDT) @ Garnet 216-218 None
Knowledge Entropy Decay during Language Model Pretraining Hinders New Knowledge Acquisition
[
OpenReview]
Oral
Fri Apr 25 08:06 PM -- 08:18 PM (PDT) @ Garnet 216-218 None
Cross-Entropy Is All You Need To Invert the Data Generating Process
[
OpenReview]
Oral
Fri Apr 25 08:18 PM -- 08:30 PM (PDT) @ Garnet 216-218 None
Can a MISL Fly? Analysis and Ingredients for Mutual Information Skill Learning
[
Slides]
[
OpenReview]
Successful Page Load