[None][feat] Fix attention sink load in xqa#8836

PerkzZheng merged 1 commit intoNVIDIA:mainNVIDIA/TensorRT-LLM:mainfrom

qsang-nv:xqa_attentionsink_load_fixqsang-nv/TensorRT-LLM:xqa_attentionsink_load_fixCopy head branch name to clipboard