Pinned Loading

1. visual-token-pruning visual-token-pruning Public

   evaluating signals for pruning visual tokens in VLMs

   Python

2. ScalingIntelligence/KernelBench ScalingIntelligence/KernelBench Public

   KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)

   Jupyter Notebook 1k 169

3. vlm-norms vlm-norms Public

   investigating norm discrepancies in VLMs

   Python 1

4. vrlm-trajectories vrlm-trajectories Public

   dataset for creating RLM trajectories on video understanding tasks

   Python

5. forge forge Public

   single gpu machine job queuing cli for background work (i am so sick of clearml)

   Python