5 papers
From Error Prediction to Causal Influence: Auditing SAE Features in Multiple-Choice QA
Aria Nourbakhsh, Adelaide Danilov, Oleksandr Marchenko Breneur +1
Sparse autoencoders (SAEs) can predict internal features associated with model errors, but association alone does not establish causal influence. We study this distinction in multi…
Sparse Autoencoders for Mechanistic Interpretability in NLP: A Critical Survey of Recoverability, Interpretability, and Utility
Salima Lamsiyah, Adelaide Danilov, Oleksandr Marchenko Breneur +1
Sparse autoencoders (SAEs) have emerged as a promising approach to mechanistic interpretability by decomposing language-model activations into sparse, potentially interpretable fea…
"Many Are My Names": The Anatomy of the Assistant and Its Personas via Sparse Autoencoders
Adelaide Danilov, Aria Nourbakhsh, Oleksandr Marchenko Breneur +1
How a language model internally represents who is speaking, the Assistant, an assigned roleplay persona, or a narrated story character, remains underexplored. We study speaker repr…
CSULoRA: Closest Safe Update Low-Rank Adaptation
Oleksandr Marchenko Breneur, Adelaide Danilov, Aria Nourbakhsh +1
Low-rank adaptation has become a standard method for parameter-efficient fine-tuning of large language models, but even small amounts of unsafe or adversarial fine-tuning data can…
NOTAI.AI: Explainable Detection of Machine-Generated Text via Curvature and Feature Attribution
Oleksandr Marchenko Breneur, Adelaide Danilov, Aria Nourbakhsh +1
We present NotAI.AI, an explainable AI-generated text detection system. Instead of returning only a binary label or confidence score, the system shows which signals influenced the…