PAIR code

facets

Visualizations for machine learning datasets

lit

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

saliency

Framework-agnostic implementation for state-of-the-art saliency methods (XRAI, BlurIG, SmoothGrad, and more).

what-if-tool

Source code/webpage/demos for the What-If Tool

umap-js

JavaScript implementation of UMAP

wordcraft

✨✍️ Wordcraft is an AI-powered text editor with an emphasis on short story writing

datacardsplaybook

The Data Cards Playbook helps dataset producers and publishers adopt a people-centered approach to transparency in dataset documentation.

federated-learning

Federated learning experiment using TensorFlow.js

scatter-gl

Interactive 3D / 2D webgl-accelerated scatter plot point renderer

understanding-umap

Understanding the theory behind UMAP

interpretability

PAIR.withgoogle.com and friend's work on interpretability methods

llm-comparator

LLM Comparator is an interactive data visualization tool for evaluating and analyzing LLM responses side-by-side, developed by the PAIR team.

ai-explorables

https://pair.withgoogle.com/explorables/

cococo

𝄡 Collaborative Convolutional Counterpoint

font-explorer

Font latent space explorer using tensorflow.js

megaplot

pair-code.github.io

depth-maps-art-and-illusions

recommendation-rudders

covid19_symptom_dataset

farsight

In situ interactive widgets for responsible AI 🌱

thehardway

Supplementary code repository to accompany Tic-Tac-Toe the Hard Way podcast

tiny-transformers

waterfall-of-meaning

deeplearnjs-legacy-loader

Deprecated: Legacy TensorFlow model loader for deeplearn.js

auto-histograms

llm-mediation-experiments

ml-vis-experiments

model-alignment

Model Alignment is a python library from the PAIR team that enable users to create model prompts through user feedback instead of manual prompt writing and editing. The technique makes use of constitutional principles to align prompts to users' desired values.

ts-llmt