Pinned Loading

1. modded-nanogpt modded-nanogpt Public

   NanoGPT (124M) in 3 minutes

   Python 2.6k 315

2. cifar10-airbench cifar10-airbench Public

   CIFAR-10 speedruns: 94% in 2.6 seconds and 96% in 27 seconds

   Python 239 20

3. Muon Muon Public

   Muon: An optimizer for hidden layers in neural networks

   Python 679 34

4. REPAIR REPAIR Public

   Code release for REPAIR: REnormalizing Permuted Activations for Interpolation Repair

   Jupyter Notebook 47 7

5. Evaluate-CrossMax-Ensemble Evaluate-CrossMax-Ensemble Public

   An evaluation of the robust accuracy of the CrossMax Ensemble technique (Fort et al., 2024)

   Python 1

6. ResNet-PyTorch-CIFAR10 ResNet-PyTorch-CIFAR10 Public archive

   PyTorch implementation of residual networks trained on CIFAR-10 dataset (2017)

   Python 30 8