GPU Kernel for a Transformer Layer
Hand-written Triton attention kernels, benchmarked against the baseline, with shape-aware dispatch choosing the fastest path per case with every optimization validated and explainable.
Hand-written Triton attention kernels, benchmarked against the baseline, with shape-aware dispatch choosing the fastest path per case with every optimization validated and explainable.
Log in or sign up for Devpost to join the conversation.