Remove block sparse attention to accommodate upstream change#285

lsy323 merged 2 commits into

mainvllm-project/tpu-inference:mainfrom

lsiyuan/fix-torchax-runnervllm-project/tpu-inference:lsiyuan/fix-torchax-runnerCopy head branch name to clipboard