tacit/nanochat
1
0
Fork 0
mirror of https://github.com/karpathy/nanochat.git synced 2026-02-05 10:09:57 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Update nanochat/flash_attention.py

Browse Source 
Code review suggestion

Co-authored-by: Sofie Van Landeghem <svlandeg@users.noreply.github.com>
This commit is contained in:
Franci Penov 2026-01-31 13:07:45 -08:00 committed by GitHub
parent 2e45b7800a
commit 4e70a2b678
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
nanochat/flash_attention.py
View File

@ -25,7 +25,7 @@ def _load_flash_attention_3():
if not torch.cuda.is_available():
return None
try:
major, minor = torch.cuda.get_device_capability()
major, _ = torch.cuda.get_device_capability()
# FA3 kernels are compiled for Hopper (sm90) only
# Ada (sm89), Blackwell (sm100) need SDPA fallback until FA3 is recompiled
if major != 9: