xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-14 15:45:28 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix some typos : supported_head_sizes (#14627)

Browse Source
This commit is contained in:
Yang.Tao 2025-03-12 01:38:24 +08:00 committed by GitHub
parent 4cbf286794
commit 53056731fd
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
vllm/attention/backends/blocksparse_attn.py
View File

@ -335,11 +335,11 @@ class BlocksparseFlashAttentionImpl(AttentionImpl):
self.sparse_block_size = self.blocksparse_params.block_size self.sparse_block_size = self.blocksparse_params.block_size
self.head_sliding_step = self.blocksparse_params.head_sliding_step self.head_sliding_step = self.blocksparse_params.head_sliding_step
suppored_head_sizes = PagedAttention.get_supported_head_sizes() supported_head_sizes = PagedAttention.get_supported_head_sizes()
if head_size not in suppored_head_sizes: if head_size not in supported_head_sizes:
raise ValueError( raise ValueError(
f"Head size {head_size} is not supported by PagedAttention. " f"Head size {head_size} is not supported by PagedAttention. "
f"Supported head sizes are: {suppored_head_sizes}.") f"Supported head sizes are: {supported_head_sizes}.")
self.tp_size = get_tensor_model_parallel_world_size() self.tp_size = get_tensor_model_parallel_world_size()
self.tp_rank = get_tensor_model_parallel_rank() self.tp_rank = get_tensor_model_parallel_rank()