Fix by replacing view with reshape

alibaba · Sep 12, 2023 · 4da868f · 4da868f
1 parent 827c3e1
commit 4da868f
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/megatron/model/transformer.py b/megatron/model/transformer.py
@@ -312,7  312,7 @@ def forward(self, query_layer, key_layer,
                        value_layer.size(3))
 
         # change view [sk, b * np, hn]
-        value_layer = value_layer.view(value_layer.size(0),
         value_layer = value_layer.reshape(value_layer.size(0),
                                        output_size[0] * output_size[1], -1)
 
         # change view [b * np, sq, sk]