PAIR-code/lit

The Learning Interpretability Tool: Interactively analyze ML models to understand their behavior in an extensible and framework agnostic interface.

56
/ 100
Established

This tool helps machine learning engineers and researchers understand how their ML models behave. You input your model (supporting text, image, or tabular data) and relevant datasets, and it outputs interactive visualizations and analyses in a web browser. This allows you to explore model predictions, identify errors, and debug performance issues.

3,640 stars.

Use this if you need to visually and interactively analyze why your machine learning model makes specific predictions or where it performs poorly.

Not ideal if you're looking for an automated model optimization solution or a tool for general data exploration unrelated to ML model behavior.

ML model analysis Model debugging Machine learning interpretability AI ethics Model performance evaluation
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 20 / 25

How are scores calculated?

Stars

3,640

Forks

371

Language

TypeScript

License

Apache-2.0

Last pushed

Mar 11, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/PAIR-code/lit"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.