Visualizing Attention in Transformer-Based Language Representation Models

4 Apr 2019 · Jesse Vig ·

We present an open-source tool for visualizing multi-head self-attention in Transformer-based language representation models. The tool extends earlier work by visualizing attention at three levels of granularity: the attention-head level, the model level, and the neuron level. We describe how each of these views can help to interpret the model, and we demonstrate the tool on the BERT model and the OpenAI GPT-2 model. We also present three use cases for analyzing GPT-2: detecting model bias, identifying recurring patterns, and linking neurons to model behavior.

PDF Abstract

Code

Add Remove Mark official

No code implementations yet. Submit your code now

Tasks

Add Remove

Language Modelling

Datasets

Add Datasets introduced or used in this paper

Results from the Paper

Add Remove

Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods

Add Remove

Adam • Attention Dropout • BPE • Cosine Annealing • Dense Connections • Discriminative Fine-Tuning • Dropout • GELU • GPT-2 • Layer Normalization • Linear Layer • Linear Warmup With Cosine Annealing • Multi-Head Attention • Residual Connection • Scaled Dot-Product Attention • Softmax • Weight Decay

Edit Social Preview

Visualizing Attention in Transformer-Based Language Representation Models

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit Add Remove

Methods Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Add Remove

Methods

Add Remove