Support seq_len > 64K in rotary embedding cuda kernel #20204
handle seq_len > 64k
d83f619e
gh-yewang
changed the title handle seq_len > 64k Support seq_len > 64K in rotary embedding cuda kernel 2 years ago
Update rotary_embedding_impl.cu
09a8bdfe
tianleiwu
approved these changes
on 2024-04-06
gh-yewang
merged
cc3faba6
into main 2 years ago
gh-yewang
deleted the wangyems-patch-4 branch 2 years ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub