BertViz is a tool for visualizing attention in the Transformer model, supporting all models from the transformers library (BERT, GPT-2, XLNet, RoBERTa, XLM, CTRL, etc.).

Don't forget to tag @jessevig in your comment, otherwise they may not be notified.

Share this project
Similar projects
Visualizing Memorization in RNNs
Inspecting gradient magnitudes in context can be a powerful tool to see when recurrent units use short-term or long-term contextual understanding.
Visualizing Class Activation Maps for CNNs
Visualizing Class Activation Maps for Convolutional Neural Networks
Feature Visualization
How neural networks build up their understanding of images
Visualization toolkit for neural networks in PyTorch