Skip to content

Commit

Permalink
update note
Browse files Browse the repository at this point in the history
Signed-off-by: Chen Zhang <[email protected]>
  • Loading branch information
heheda12345 committed Jan 29, 2025
1 parent 92d3397 commit e1399a8
Showing 1 changed file with 2 additions and 2 deletions.
4 changes: 2 additions & 2 deletions vllm/attention/layer.py
Original file line number Diff line number Diff line change
Expand Up @@ -149,8 +149,8 @@ def forward(
attn_metadata: AttentionMetadata,
) -> torch.Tensor:
# NOTE: please avoid accessing `kv_cache` and `attn_metadata` arguments
# directly. Instead, please use the `self.kv_cache` and
# `get_forward_context().attn_metadata` to access them.
# directly, use `self.kv_cache` and
# `get_forward_context().attn_metadata` instead.
if self.calculate_kv_scales:
ctx_attn_metadata = get_forward_context().attn_metadata
if ctx_attn_metadata.enable_kv_scales_calculation:
Expand Down

0 comments on commit e1399a8

Please sign in to comment.