Feature-Level Insights into Artificial Text Detection with Sparse Autoencoders

Kristian Kuznetsov, Kushnareva, Polina Druzhinina, Anton Razzhigaev, Anastasia Voznyuk, Irina Piontkovskaya, Evgeny Burnaev, Serguei Barannikov

Feature-Level Insights into Artificial Text Detection with Sparse Autoencoders: 234 upvotes on Hugging Face Daily Papers, #1 of 50 papers on 2025-03-11. Day-by-day upvote history.

Artificial Text Detection (ATD) is becoming increasingly important with the rise of advanced Large Language Models (LLMs). Despite numerous efforts, no single algorithm performs consistently well across different types of unseen text or guarantees effective generalization to new LLMs. Interpretability plays a crucial role in achieving this goal. In this study, we enhance ATD interpretability by using Sparse Autoencoders (SAE) to extract features from Gemma-2-2b residual stream. We identify both interpretable and efficient features, analyzing their semantics and relevance through domain- and model-specific statistics, a steering approach, and manual or LLM-based interpretation. Our methods offer valuable insights into how texts from various models differ from human-written content. We show that modern LLMs have a distinct writing style, especially in information-dense domains, even though they can produce human-like outputs with personalized prompts.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.