diff --git a/pytorch_transformers/modeling_bert.py b/pytorch_transformers/modeling_bert.py
index 0b1cfd6a..5d3c1606 100644
--- a/pytorch_transformers/modeling_bert.py
+++ b/pytorch_transformers/modeling_bert.py
@@ -852,7 +852,7 @@ class BertForMaskedLM(BertPreTrainedModel):
         sequence_output = outputs[0]
         prediction_scores = self.cls(sequence_output)
 
-        outputs = (prediction_scores,) + outputs[2:]  # Add hidden states and attention is they are here
+        outputs = (prediction_scores,) + outputs[2:]  # Add hidden states and attention if they are here
         if masked_lm_labels is not None:
             loss_fct = CrossEntropyLoss(ignore_index=-1)
             masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), masked_lm_labels.view(-1))