Flash-MSA: Accelerating Million-Token Training with Sparse Attention Kernels (nanduruganesh.github.io) | HN Screenshots

YHacker Newsscreenshots

Screenshot of nanduruganesh.github.io: Flash-MSA: Accelerating Million-Token Training with Sparse Attention Kernels

First screen, captured 3 months ago.

Looks similar

Screenshot of aminediro.com: I rebuilt FlashAttention in Triton to understand the performance archaeology

I rebuilt FlashAttention in Triton to understand the performance archaeology (aminediro.com)

78 points by amindiro 10 months ago | 17 comments

Screenshot of pyspur.dev: DeepSeek's multi-head latent attention and other KV cache tricks

DeepSeek's multi-head latent attention and other KV cache tricks (pyspur.dev)

254 points by t55 1 year ago | 45 comments

Screenshot of chizkidd.github.io: Understanding FlashAttention Pt 1: Personal Notes

Understanding FlashAttention Pt 1: Personal Notes (chizkidd.github.io)

11 points by ibobev 1 month ago | 0 comments

Screenshot of Hacker News post: Hybrid Attention

Hybrid Attention

24 points by JohannaAlmeida 6 months ago | 6 comments

Screenshot of iaroslavelistratov.github.io: B200 Attention Kernel from Scratch to Near-SOTA in 60 Diagrams

B200 Attention Kernel from Scratch to Near-SOTA in 60 Diagrams (iaroslavelistratov.github.io)

13 points by magoghm 1 month ago | 0 comments

Screenshot of damek.github.io: Basic Facts about GPUs

Basic Facts about GPUs (damek.github.io)

306 points by ibobev 1 year ago | 66 comments