Pinned Loading

1. trak trak Public

   A fast, effective data attribution method for neural networks in PyTorch

   Python 190 27

2. robustness robustness Public

   A library for experimenting with, training and evaluating neural networks, with a focus on adversarial robustness.

   Jupyter Notebook 926 182

3. datamodels-data datamodels-data Public

   Data for "Datamodels: Predicting Predictions with Training Data"

   Python 95 3

4. modeldiff modeldiff Public

   ModelDiff: A Framework for Comparing Learning Algorithms

   Jupyter Notebook 55 4

5. modelcomponents modelcomponents Public

   Decomposing and Editing Predictions by Modeling Model Computation

   Jupyter Notebook 136 8

6. context-cite context-cite Public

   Attribute (or cite) statements generated by LLMs back to in-context information.

   Jupyter Notebook 193 16