https://samankeon.com/https://samankeon.com/blog/https://samankeon.com/blog/2d-workloads/https://samankeon.com/blog/add-kernel-1/https://samankeon.com/blog/add-kernel-2/https://samankeon.com/blog/add-kernel-3/https://samankeon.com/blog/add-kernel-4/https://samankeon.com/blog/attention-isnt-all-you-need-you-also/https://samankeon.com/blog/breaking-the-quadratic-barrier/https://samankeon.com/blog/continuous-batching-the-secret-sauce/https://samankeon.com/blog/from-brute-force-to-finesse-the-story/https://samankeon.com/blog/from-greedy-to-nucleus-how-llms-choose/https://samankeon.com/blog/from-word2vec-to-bert-the-evolution/https://samankeon.com/blog/how-moe-powers-modern-llms/https://samankeon.com/blog/matmul/https://samankeon.com/blog/rag-giving-llms-an-external-brain/https://samankeon.com/blog/rope-the-phase-geometry-behind-long/https://samankeon.com/blog/softmax-kernel-1/https://samankeon.com/blog/softmax-kernel-2/https://samankeon.com/blog/softmax-kernel-3/https://samankeon.com/blog/tag/benchmarking/https://samankeon.com/blog/tag/cuda-kernel/https://samankeon.com/blog/tag/image-processing/https://samankeon.com/blog/tag/kernels/https://samankeon.com/blog/tag/llm/https://samankeon.com/blog/tag/positional%20embedding/https://samankeon.com/blog/tag/rope/https://samankeon.com/blog/tag/triton-kernel/https://samankeon.com/blog/the-command-line-is-learning-to-talk/https://samankeon.com/publications/