We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent ac80ee2 commit 3322ef8Copy full SHA for 3322ef8
LongNet/attention.py
@@ -13,7 +13,7 @@
13
dtype=torch.float16
14
15
16
-
+#
17
18
#second iteration the weighted sum of the different dilated + offsets for the different heads
19
class DilatedAttention(nn.Module):
0 commit comments