Sidharth Baskaran
email 路 github 路 linkedin 路 饾晱 路 scholar
I am interested in building systems and novel methods to understand and improve language model capabilities.
Research
HyperSteer: Activation Steering at Scale with Hypernetworks
Jiuding Sun*, Sidharth Baskaran*, Zhengxuan Wu, Michael Sklar, Christopher Potts, Atticus Geiger
HyperDAS: Towards Automating Mechanistic Interpretability with Hypernetworks
Jiuding Sun, Jing Huang, Sidharth Baskaran, Karel D'Oosterlinck, Christopher Potts, Michael Sklar*, Atticus Geiger*
Rebuilding ROME: Resolving Model Collapse during Sequential Model Editing
Akshat Gupta, Sidharth Baskaran, Gopala Anumanchipalli
* denotes equal contribution
Software
Ray Job Submission launcher plugin for the Hydra library.
Simple orchestration of containers and remote workflows for research.