Unverified Commit 9869453c authored by Happy's avatar Happy Committed by GitHub
Browse files

Update test_flash_attn.py (#17102)


Signed-off-by: default avatarShuaibinLi <lishuaibin@live.cn>
parent 3642c59a
......@@ -145,7 +145,7 @@ def test_flash_attn_with_paged_kv(
v_descale = None
if q_dtype is not None:
# QKV are drawn from N(0, 1): no need for a fp8 scaling factor
maybe_quantized_query = query.to(q_dtype)
maybe_quantized_query = q.to(q_dtype)
maybe_quantized_key_cache = key_cache.to(q_dtype)
maybe_quantized_value_cache = value_cache.to(q_dtype)
......
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