Update modeling_norbert.py

modeling_norbert.py

@@ -260,6 +260,7 @@ class NorbertModel(NorbertPreTrainedModel):
     def __init__(self, config, add_mlm_layer=False, gradient_checkpointing=False, **kwargs):
         super().__init__(config, **kwargs)
         self.config = config
+        self.hidden_size = config.hidden_size
 
         self.embedding = Embedding(config)
         self.transformer = Encoder(config, activation_checkpointing=gradient_checkpointing)