Dodrio: Exploring Transformer Models with Interactive Visualization

TL;DR

Dodrio is an interactive visualization tool designed to help researchers analyze and understand the attention mechanisms in transformer models, revealing how they encode linguistic information to improve NLP performance.

Contribution

This paper introduces Dodrio, a novel open-source visualization tool that integrates overview and detailed views for analyzing attention in transformer models.

Findings

Dodrio enables detailed analysis of attention heads and linguistic features.

It helps identify how models encode syntactic and semantic information.

Case studies demonstrate Dodrio's effectiveness in revealing model behaviors.

Abstract

Why do large pre-trained transformer-based models perform so well across a wide variety of NLP tasks? Recent research suggests the key may lie in multi-headed attention mechanism's ability to learn and represent linguistic information. Understanding how these models represent both syntactic and semantic knowledge is vital to investigate why they succeed and fail, what they have learned, and how they can improve. We present Dodrio, an open-source interactive visualization tool to help NLP researchers and practitioners analyze attention mechanisms in transformer-based models with linguistic knowledge. Dodrio tightly integrates an overview that summarizes the roles of different attention heads, and detailed views that help users compare attention weights with the syntactic structure and semantic information in the input text. To facilitate the visual comparison of attention weights and linguistic knowledge, Dodrio applies different graph visualization techniques to represent attention weights scalable to longer input text. Case studies highlight how Dodrio provides insights into understanding the attention mechanism in transformer-based models. Dodrio is available at https://poloclub.github.io/dodrio/.