diff --git a/pytorch_transformers/modeling_bert.py b/pytorch_transformers/modeling_bert.py
index a8239038a7..c931ac6ab2 100644
--- a/pytorch_transformers/modeling_bert.py
+++ b/pytorch_transformers/modeling_bert.py
@@ -642,7 +642,7 @@ BERT_INPUTS_DOCSTRING = r"""
 @add_start_docstrings("The bare Bert Model transformer outputing raw hidden-states without any specific head on top.",
                       BERT_START_DOCSTRING, BERT_INPUTS_DOCSTRING)
 class BertModel(BertPreTrainedModel):
-    r"""
+    __doc__ = r"""
     Outputs: `Tuple` comprising various elements depending on the configuration (config) and inputs:
         **last_hidden_state**: ``torch.FloatTensor`` of shape ``(batch_size, sequence_length, hidden_size)``
             Sequence of hidden-states at the last layer of the model.
@@ -738,7 +738,7 @@ class BertModel(BertPreTrainedModel):
     a `masked language modeling` head and a `next sentence prediction (classification)` head. """,
     BERT_START_DOCSTRING, BERT_INPUTS_DOCSTRING)
 class BertForPreTraining(BertPreTrainedModel):
-    r"""
+    __doc__ = r"""
         **masked_lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
             Labels for computing the masked language modeling loss.
             Indices should be in ``[-1, 0, ..., config.vocab_size]`` (see ``input_ids`` docstring)
@@ -814,7 +814,7 @@ class BertForPreTraining(BertPreTrainedModel):
 @add_start_docstrings("""Bert Model transformer BERT model with a `language modeling` head on top. """,
     BERT_START_DOCSTRING, BERT_INPUTS_DOCSTRING)
 class BertForMaskedLM(BertPreTrainedModel):
-    r"""
+    __doc__ = r"""
         **masked_lm_labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
             Labels for computing the masked language modeling loss.
             Indices should be in ``[-1, 0, ..., config.vocab_size]`` (see ``input_ids`` docstring)
@@ -879,7 +879,7 @@ class BertForMaskedLM(BertPreTrainedModel):
 @add_start_docstrings("""Bert Model transformer BERT model with a `next sentence prediction (classification)` head on top. """,
     BERT_START_DOCSTRING, BERT_INPUTS_DOCSTRING)
 class BertForNextSentencePrediction(BertPreTrainedModel):
-    r"""
+    __doc__ = r"""
         **next_sentence_label**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size,)``:
             Labels for computing the next sequence prediction (classification) loss. Input should be a sequence pair (see ``input_ids`` docstring)
             Indices should be in ``[0, 1]``.
@@ -937,7 +937,7 @@ class BertForNextSentencePrediction(BertPreTrainedModel):
     the pooled output) e.g. for GLUE tasks. """,
     BERT_START_DOCSTRING, BERT_INPUTS_DOCSTRING)
 class BertForSequenceClassification(BertPreTrainedModel):
-    r"""
+    __doc__ = r"""
         **labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size,)``:
             Labels for computing the sequence classification/regression loss.
             Indices should be in ``[0, ..., config.num_labels]``.
@@ -1005,7 +1005,7 @@ class BertForSequenceClassification(BertPreTrainedModel):
     the pooled output and a softmax) e.g. for RocStories/SWAG tasks. """,
     BERT_START_DOCSTRING)
 class BertForMultipleChoice(BertPreTrainedModel):
-    r"""
+    __doc__ = r"""
     Inputs:
         **input_ids**: ``torch.LongTensor`` of shape ``(batch_size, num_choices, sequence_length)``:
             Indices of input sequence tokens in the vocabulary.
@@ -1110,7 +1110,7 @@ class BertForMultipleChoice(BertPreTrainedModel):
     the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks. """,
     BERT_START_DOCSTRING, BERT_INPUTS_DOCSTRING)
 class BertForTokenClassification(BertPreTrainedModel):
-    r"""
+    __doc__ = r"""
         **labels**: (`optional`) ``torch.LongTensor`` of shape ``(batch_size, sequence_length)``:
             Labels for computing the token classification loss.
             Indices should be in ``[0, ..., config.num_labels]``.