Pinned Loading

1. mandala mandala Public

   A simple & elegant experiment tracking framework that integrates persistence logic & best practices directly into Python

   Jupyter Notebook 535 15

2. MadryLab/cifar10_challenge MadryLab/cifar10_challenge Public

   A challenge to explore adversarial robustness of neural networks on CIFAR10.

   Python 503 133

3. activation-patching-illusion activation-patching-illusion Public

   Code for "Is This the Subspace You Are Looking for? An Interpretability Illusion for Subspace Activation Patching"

   Jupyter Notebook 8

4. sae sae Public

   SAE evals https://arxiv.org/abs/2405.08366

   Jupyter Notebook 3