What it does
Language models use a mechanism called attention to decide which words matter to each other. BertViz draws this as interactive pictures so you can see what is happening inside the model.
It offers several views: one layer at a time, a bird's-eye view of the whole model, and a close look at individual parts of the calculation. It works with most models from the Hugging Face library.
It runs inside Jupyter or Google Colab notebooks, and there is a ready-made Colab tutorial with all the views loaded.
Who it's for
Students, teachers and researchers who want to understand or explain how transformer models work.
Can I use it without coding?
Developers only. You run it from Python code in a notebook, though the Colab tutorial needs very little setup. New to this? Start here.
Can I run it?
- Cost
- FreeNothing to pay or sign up for.
- Runs on
- Web browser
- Computer
- An ordinary computer is fine
- Privacy
- Runs on your own device: your data stays with you
- API key
- Not needed
Try it
- Open the project's interactive Colab tutorial, which runs in your web browser with the views already loaded.
Using Claude Code, Cursor or another AI coding assistant? Paste the prompt and it will walk you through the setup.New to this?
Before you install
- Only download from the official GitHub page. Copies with similar names exist for many popular projects.
- License (Apache-2.0): Permissive: free to use, change and share, including for business.
- More on staying safe