5 citations · 8 across the 19 of their papers we have counts for
3 papers · 1 filter
Sparse Autoencoders Encode Both Concepts and Functions: The Downstream Geometry of Feature Effects
Phu Gia Hoang, Anwoy Chatterjee, Tanmoy Chakraborty +2
The wide-scale use of sparse autoencoders (SAEs) as interpretability tools is limited by inconsistent links between SAE features and model behavior. Features with clear activation…
AICD Bench: A Challenging Benchmark for AI-Generated Code Detection
Daniil Orel, Dilshod Azizov, Indraneil Paul +3
Large language models (LLMs) are increasingly capable of generating functional source code, raising concerns about authorship, accountability, and security. While detecting AI-gene…
Early Period of Training Impacts Adaptation for Out-of-Distribution Generalization: An Empirical Study
Chen Cecilia Liu, Iryna Gurevych
Prior research shows that differences in the early period of neural network training significantly impact the performance of in-distribution (ID) data of tasks. Yet, the implicatio…