github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-10-09 20:14:29 +00:00
Code Issues Packages Projects Releases Wiki Activity

[misc] update pre-commit and run all files (#4752)

Browse Source 
* [misc] update pre-commit

* [misc] run pre-commit

* [misc] remove useless configuration files

* [misc] ignore cuda for clang-format
This commit is contained in:
Hongxin Liu
2023-09-19 14:20:26 +08:00
committed by GitHub
parent 3c6b831c26
commit 079bf3cb26
1268 changed files with 50037 additions and 38444 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
examples/language/llama2/attn.py
View File

@@ -9,12 +9,14 @@ SUPPORT_XFORMERS = False
SUPPORT_FLASH2 = False
try:
import xformers.ops as xops
SUPPORT_XFORMERS = True
except ImportError:
pass
try:
from flash_attn import flash_attn_func
SUPPORT_FLASH2 = True
except ImportError:
pass
@@ -62,10 +64,9 @@ def llama_flash_attention(
if SUPPORT_FLASH2:
attn_output = flash_attn_func(query_states, key_states, value_states, causal=True)
else:
attn_output = xops.memory_efficient_attention(query_states,
key_states,
value_states,
attn_bias=xops.LowerTriangularMask())
attn_output = xops.memory_efficient_attention(
query_states, key_states, value_states, attn_bias=xops.LowerTriangularMask()
)
attn_output = attn_output.reshape(bsz, q_len, self.hidden_size)