Aditya Raj

small tar­geted train­ing as a mechanis­tic “am­plifier” for dis­cov­er­ing la­tent/​learned de­cep­tive circuitry

Chain-of-Thought as Con­tex­tual Sta­bi­liza­tion and As­so­ci­a­tive Retrieval

Why Safety Con­straints in LLMs Are Easily Break­able? Knowl­edge as a Net­work of Gated Circuits