Unverified Commit 895d3946 authored by Lysandre Debut's avatar Lysandre Debut Committed by GitHub
Browse files

TF Flaubert w/ pre-norm (#6841)

parent 4561f05c
...@@ -296,7 +296,7 @@ class TFFlaubertMainLayer(TFXLMMainLayer): ...@@ -296,7 +296,7 @@ class TFFlaubertMainLayer(TFXLMMainLayer):
else: else:
tensor_normalized = self.layer_norm1[i](tensor) tensor_normalized = self.layer_norm1[i](tensor)
attn_outputs = self.attentions[i]( attn_outputs = self.attentions[i](
tensor_normalized, attn_mask, None, cache, head_mask[i], training=training tensor_normalized, attn_mask, None, cache, head_mask[i], output_attentions, training=training
) )
attn = attn_outputs[0] attn = attn_outputs[0]
if output_attentions: if output_attentions:
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment