Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings
This repository was archived by the owner on Jan 21, 2025. It is now read-only.
/meshPublic archive

Commit0a212ca

Browse files
author
Mesh TensorFlow Team
committed
Fix bug in shared_kv attention for autoregressive decoding.
PiperOrigin-RevId: 391048755
1 parent7c09bb9 commit0a212ca

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

‎mesh_tensorflow/transformer/transformer_layers.py‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -247,7 +247,7 @@ def call(self, context, x, losses=None):
247247
context.position,memory_length,dtype=context.activation_dtype)
248248
inv_one_hot=1.0-one_hot
249249
ifself.shared_kv:
250-
old_kv=context.get_states(1)
250+
old_kv,=context.get_states(1)
251251
kv=old_kv*inv_one_hot+kv*one_hot
252252
else:
253253
old_k,old_v=context.get_states(2)

0 commit comments

Comments
 (0)

[8]ページ先頭

©2009-2025 Movatter.jp