even stronger

lucidrains · Nov 23, 2024 · 28b693d · 28b693d
1 parent 3d61134
commit 28b693d
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 3 deletions.
diff --git a/setup.py b/setup.py
@@ -3,7 +3,7 @@
 setup(
   name = 'x-transformers',
   packages = find_packages(exclude=['examples']),
-  version = '1.42.15',
+  version = '1.42.16',
   license='MIT',
   description = 'X-Transformers - Pytorch',
   author = 'Phil Wang',

diff --git a/x_transformers/x_transformers.py b/x_transformers/x_transformers.py
@@ -1235,9 +1235,9 @@ def __init__(
         # maybe learned value residual mixer per token
 
         self.to_value_residual_mix = nn.Sequential(
-            nn.Linear(dim, 1),
+            nn.Linear(dim, heads),
             nn.Sigmoid(),
-            Rearrange('b n 1 -> b 1 n 1')
+            Rearrange('b n h -> b h n 1')
          ) if learned_value_residual_mix else always(0.5)
 
         # attention on attention