update BART docs (#17212)

src/transformers/models/bart/modeling_bart.py

@@ -624,9 +624,9 @@ BART_INPUTS_DOCSTRING = r"""
             Default behavior: generate a tensor that ignores pad tokens in `decoder_input_ids`. Causal mask will also
             be used by default.
 
-            If you want to change padding behavior, you should read [`modeling_bart._prepare_decoder_inputs`] and
-            modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more information
-            on the default strategy.
+            If you want to change padding behavior, you should read [`modeling_bart._prepare_decoder_attention_mask`]
+            and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
+            information on the default strategy.
         head_mask (`torch.Tensor` of shape `(encoder_layers, encoder_attention_heads)`, *optional*):
             Mask to nullify selected heads of the attention modules in the encoder. Mask values selected in `[0, 1]`:
 

src/transformers/models/bigbird_pegasus/modeling_bigbird_pegasus.py

@@ -1685,8 +1685,8 @@ BIGBIRD_PEGASUS_INPUTS_DOCSTRING = r"""
             be used by default.
 
             If you want to change padding behavior, you should read
-            [`modeling_bigbird_pegasus._prepare_decoder_inputs`] and modify to your needs. See diagram 1 in [the
-            paper](https://arxiv.org/abs/1910.13461) for more information on the default strategy.
+            [`modeling_bigbird_pegasus._prepare_decoder_attention_mask`] and modify to your needs. See diagram 1 in
+            [the paper](https://arxiv.org/abs/1910.13461) for more information on the default strategy.
 
         decoder_head_mask (`torch.Tensor` of shape `(num_layers, num_heads)`, *optional*):
             Mask to nullify selected heads of the attention modules in the decoder. Mask values selected in `[0, 1]`:

src/transformers/models/opt/modeling_opt.py

@@ -464,9 +464,9 @@ OPT_INPUTS_DOCSTRING = r"""
             If `past_key_values` is used, optionally only the last `decoder_input_ids` have to be input (see
             `past_key_values`).
 
-            If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_inputs`] and modify
-            to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more information on the
-            default strategy.
+            If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
+            and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
+            information on the default strategy.
         head_mask (`torch.Tensor` of shape `(encoder_layers, encoder_attention_heads)`, *optional*):
             Mask to nullify selected heads of the attention modules in the encoder. Mask values selected in `[0, 1]`:
 

src/transformers/models/speech_to_text/modeling_speech_to_text.py

@@ -625,9 +625,9 @@ SPEECH_TO_TEXT_INPUTS_DOCSTRING = r"""
             Default behavior: generate a tensor that ignores pad tokens in `decoder_input_ids`. Causal mask will also
             be used by default.
 
-            If you want to change padding behavior, you should read [`modeling_speech_to_text._prepare_decoder_inputs`]
-            and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
-            information on the default strategy.
+            If you want to change padding behavior, you should read
+            [`modeling_speech_to_text._prepare_decoder_attention_mask`] and modify to your needs. See diagram 1 in [the
+            paper](https://arxiv.org/abs/1910.13461) for more information on the default strategy.
         head_mask (`torch.Tensor` of shape `(encoder_layers, encoder_attention_heads)`, *optional*):
             Mask to nullify selected heads of the attention modules in the encoder. Mask values selected in `[0, 1]`:
 

templates/adding_a_new_model/cookiecutter-template-{{cookiecutter.modelname}}/modeling_{{cookiecutter.lowercase_modelname}}.py

@@ -2100,7 +2100,7 @@ class {{cookiecutter.camelcase_modelname}}PreTrainedModel(PreTrainedModel):
             Default behavior: generate a tensor that ignores pad tokens in `decoder_input_ids`. Causal mask will
             also be used by default.
 
-            If you want to change padding behavior, you should read [`modeling_{{cookiecutter.lowercase_modelname}}._prepare_decoder_inputs`] and
+            If you want to change padding behavior, you should read [`modeling_{{cookiecutter.lowercase_modelname}}._prepare_decoder_attention_mask`] and
             modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
             information on the default strategy.
         head_mask (`torch.Tensor` of shape `(encoder_layers, encoder_attention_heads)`, *optional*):