google-deepmind/penzai

Flax is a neural network library for JAX that is designed for flexibility.

JAX-based neural network library

Multi-framework implementation of Deep Kernel Shaping and Tailored Activation Transformations, which are methods that modify neural network models (and their initializations) to make them easier to train.

open source interpretability platform 🧠

Capture every activation and gradient of any PyTorch model — forward and backward — with automatic graph visualization, rich metadata, and live interventions. Works on any architecture, including dynamic and recurrent ones.

Unified access to Large Language Model modules using NNsight

The nnsight package enables interpreting and manipulating the internals of deep learned models.

Interactive 3D visualization platform for exploring transformer architectures, tensors, and real-time LLM inference.