500AI
← People

Santhosh Janakiraman

Source-listed: Graduate student · Rutgers, The State University of New Jersey

Website ↗GitHub ↗Institution ↗
  • Language models
  • Interpretability

Selected work

5
  1. LLM Refusal GeometryContributor↗
  2. VLM Agent Hallucination AuditContributor↗
  3. Grokking Under Differential PrivacyContributor↗
  4. Refusal Direction Experiments (LLM-Jailbreaks-Mechanistic-Interpretability)Maintainer↗
  5. CNN-Training-for-Space-BotContributor↗
Suggest an edit
About & privacy