fix

Author: Southpika

paddlenlp/transformers/gemma/modeling.py

@@ -459,7 +459,6 @@ def forward(self, x):
 class GemmaAttention(nn.Layer):
     """Multi-headed attention from 'Attention Is All You Need' paper"""
 
-    # Ignore copy
     def __init__(self, config: GemmaConfig, layerwise_recompute: bool = False):
         super().__init__()