Simplifying Outcomes of Language Model Component Analyses with ELIA

TL;DR

ELIA is an interactive web tool that simplifies language model analysis results, making them accessible to non-experts through natural language explanations and visualizations, validated by user studies.

Contribution

This work introduces ELIA, a novel system combining multiple analysis techniques with AI-generated explanations to improve interpretability accessibility.

Findings

AI explanations bridge knowledge gaps for non-experts

Interactive visualizations preferred over static ones

System reduces barriers regardless of user experience level

Abstract

While mechanistic interpretability has developed powerful tools to analyze the internal workings of Large Language Models (LLMs), their complexity has created an accessibility gap, limiting their use to specialists. We address this challenge by designing, building, and evaluating ELIA (Explainable Language Interpretability Analysis), an interactive web application that simplifies the outcomes of various language model component analyses for a broader audience. The system integrates three key techniques -- Attribution Analysis, Function Vector Analysis, and Circuit Tracing -- and introduces a novel methodology: using a vision-language model to automatically generate natural language explanations (NLEs) for the complex visualizations produced by these methods. The effectiveness of this approach was empirically validated through a mixed-methods user study, which revealed a clear preference for interactive, explorable interfaces over simpler, static visualizations. A key finding was that the AI-powered explanations helped bridge the knowledge gap for non-experts; a statistical analysis showed no significant correlation between a user's prior LLM experience and their comprehension scores, suggesting that the system reduced barriers to comprehension across experience levels. We conclude that an AI system can indeed simplify complex model analyses, but its true power is unlocked when paired with thoughtful, user-centered design that prioritizes interactivity, specificity, and narrative guidance.