Skip to content

GitLab

Unverified Commit 204c3c6d authored by Tao He's avatar Tao He Committed by GitHub
Browse files

Fixes an error in comment (#785) 


Signed-off-by: default avatarTao He <sighingnow@gmail.com>
parent 197f2083
Hide whitespace changes
Inline Side-by-side
Showing with 2 additions and 2 deletions
README.md
View file @ 204c3c6d
...@@ -212,7 +212,7 @@ def flash_attn_with_kvcache( ...@@ -212,7 +212,7 @@ def flash_attn_with_kvcache(
rotary_sin [optional]: (seqlen_ro, rotary_dim / 2). Similar to rotary_cos. rotary_sin [optional]: (seqlen_ro, rotary_dim / 2). Similar to rotary_cos.
cache_seqlens: int, or (batch_size,), dtype torch.int32. The sequence lengths of the cache_seqlens: int, or (batch_size,), dtype torch.int32. The sequence lengths of the
KV cache. KV cache.
block_table [optional]: (num_blocks, max_num_blocks_per_seq), dtype torch.int32. block_table [optional]: (batch_size, max_num_blocks_per_seq), dtype torch.int32.
cache_batch_idx: (batch_size,), dtype torch.int32. The indices used to index into the KV cache. cache_batch_idx: (batch_size,), dtype torch.int32. The indices used to index into the KV cache.
If None, we assume that the batch indices are [0, 1, 2, ..., batch_size - 1]. If None, we assume that the batch indices are [0, 1, 2, ..., batch_size - 1].
If the indices are not distinct, and k and v are provided, the values updated in the cache If the indices are not distinct, and k and v are provided, the values updated in the cache
......
flash_attn/flash_attn_interface.py
View file @ 204c3c6d
...@@ -1149,7 +1149,7 @@ def flash_attn_with_kvcache( ...@@ -1149,7 +1149,7 @@ def flash_attn_with_kvcache(
rotary_sin [optional]: (seqlen_ro, rotary_dim / 2). Similar to rotary_cos. rotary_sin [optional]: (seqlen_ro, rotary_dim / 2). Similar to rotary_cos.
cache_seqlens: int, or (batch_size,), dtype torch.int32. The sequence lengths of the cache_seqlens: int, or (batch_size,), dtype torch.int32. The sequence lengths of the
KV cache. KV cache.
block_table [optional]: (num_blocks, max_num_blocks_per_seq), dtype torch.int32. block_table [optional]: (batch_size, max_num_blocks_per_seq), dtype torch.int32.
cache_batch_idx: (batch_size,), dtype torch.int32. The indices used to index into the KV cache. cache_batch_idx: (batch_size,), dtype torch.int32. The indices used to index into the KV cache.
If None, we assume that the batch indices are [0, 1, 2, ..., batch_size - 1]. If None, we assume that the batch indices are [0, 1, 2, ..., batch_size - 1].
If the indices are not distinct, and k and v are provided, the values updated in the cache If the indices are not distinct, and k and v are provided, the values updated in the cache
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment