GitHub

Can Large Language Models Predict Parallel Code Performance?

We took this version of HeCBench and are modifying it to build the CUDA and OMP codes to gather their roofline performance data. So far we have a large portion of the CUDA and OMP codes building ...
IEEE

Multiplication of Sparse Matrices and their Transpose Using Compressed Sparse Diagonals

Abstract: Matrix-matrix multiplication is one of the most important kernel in linear algebra operations with a multitude of applications in scientific and engineering computing. Sparse matrix ...
IEEE

MH-SpGEMM: Efficient Sparse General Matrix-Matrix Multiplication on Modern GPUs via Masking and Hashing Cooperative Optimization

Abstract: Sparse General Matrix-Matrix Multiplication (SpGEMM) is a core operation in high-performance computing applications such as algebraic multigrid solvers, machine learning, and graph ...
GitHub

Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models

Sparse Autoencoders (SAEs) have recently gained attention as a means to improve the interpretability and steerability of Large Language Models (LLMs), both of which are essential for AI safety. In ...