[mBART] fix erroneous italics in docstring (#20835)

* [mBART] fix erroneous italics in docstring

* fix-copies

src/transformers/models/bigbird_pegasus/modeling_bigbird_pegasus.py

@@ -1497,17 +1497,17 @@ class BigBirdPegasusDecoderLayer(nn.Module):
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states (`torch.FloatTensor`): input to the layer of shape *(seq_len, batch, embed_dim)*
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
             attention_mask (`torch.FloatTensor`): attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             encoder_hidden_states (`torch.FloatTensor`):
-                cross attention input to the layer of shape *(seq_len, batch, embed_dim)*
+                cross attention input to the layer of shape `(seq_len, batch, embed_dim)`
             encoder_attention_mask (`torch.FloatTensor`): encoder attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
-                *(encoder_attention_heads,)*.
+                `(encoder_attention_heads,)`.
             cross_attn_layer_head_mask (`torch.FloatTensor`): mask for cross-attention heads in a given layer of
-                size *(decoder_attention_heads,)*.
+                size `(decoder_attention_heads,)`.
             past_key_value (`Tuple(torch.FloatTensor)`): cached past key and value projection states
             output_attentions (`bool`, *optional*):
                 Whether or not to return the attentions tensors of all attention layers. See `attentions` under

src/transformers/models/blenderbot/modeling_blenderbot.py

@@ -307,11 +307,11 @@ class BlenderbotEncoderLayer(nn.Module):
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states (`torch.FloatTensor`): input to the layer of shape *(seq_len, batch, embed_dim)*
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
             attention_mask (`torch.FloatTensor`): attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
-                *(encoder_attention_heads,)*.
+                `(encoder_attention_heads,)`.
             output_attentions (`bool`, *optional*):
                 Whether or not to return the attentions tensors of all attention layers. See `attentions` under
                 returned tensors for more detail.
@@ -391,17 +391,17 @@ class BlenderbotDecoderLayer(nn.Module):
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states (`torch.FloatTensor`): input to the layer of shape *(seq_len, batch, embed_dim)*
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
             attention_mask (`torch.FloatTensor`): attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             encoder_hidden_states (`torch.FloatTensor`):
-                cross attention input to the layer of shape *(seq_len, batch, embed_dim)*
+                cross attention input to the layer of shape `(seq_len, batch, embed_dim)`
             encoder_attention_mask (`torch.FloatTensor`): encoder attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
-                *(encoder_attention_heads,)*.
+                `(encoder_attention_heads,)`.
             cross_attn_layer_head_mask (`torch.FloatTensor`): mask for cross-attention heads in a given layer of
-                size *(decoder_attention_heads,)*.
+                size `(decoder_attention_heads,)`.
             past_key_value (`Tuple(torch.FloatTensor)`): cached past key and value projection states
             output_attentions (`bool`, *optional*):
                 Whether or not to return the attentions tensors of all attention layers. See `attentions` under

src/transformers/models/m2m_100/modeling_m2m_100.py

@@ -375,11 +375,11 @@ class M2M100EncoderLayer(nn.Module):
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states (`torch.FloatTensor`): input to the layer of shape *(seq_len, batch, embed_dim)*
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
             attention_mask (`torch.FloatTensor`): attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
-                *(encoder_attention_heads,)*.
+                `(encoder_attention_heads,)`.
             output_attentions (`bool`, *optional*):
                 Whether or not to return the attentions tensors of all attention layers. See `attentions` under
                 returned tensors for more detail.
@@ -459,17 +459,17 @@ class M2M100DecoderLayer(nn.Module):
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states (`torch.FloatTensor`): input to the layer of shape *(seq_len, batch, embed_dim)*
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
             attention_mask (`torch.FloatTensor`): attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             encoder_hidden_states (`torch.FloatTensor`):
-                cross attention input to the layer of shape *(seq_len, batch, embed_dim)*
+                cross attention input to the layer of shape `(seq_len, batch, embed_dim)`
             encoder_attention_mask (`torch.FloatTensor`): encoder attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
-                *(encoder_attention_heads,)*.
+                `(encoder_attention_heads,)`.
             cross_attn_layer_head_mask (`torch.FloatTensor`): mask for cross-attention heads in a given layer of
-                size *(decoder_attention_heads,)*.
+                size `(decoder_attention_heads,)`.
             past_key_value (`Tuple(torch.FloatTensor)`): cached past key and value projection states
             output_attentions (`bool`, *optional*):
                 Whether or not to return the attentions tensors of all attention layers. See `attentions` under

src/transformers/models/mbart/modeling_mbart.py

@@ -326,11 +326,11 @@ class MBartEncoderLayer(nn.Module):
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states (`torch.FloatTensor`): input to the layer of shape *(seq_len, batch, embed_dim)*
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
             attention_mask (`torch.FloatTensor`): attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
-                *(encoder_attention_heads,)*.
+                `(encoder_attention_heads,)`.
             output_attentions (`bool`, *optional*):
                 Whether or not to return the attentions tensors of all attention layers. See `attentions` under
                 returned tensors for more detail.
@@ -409,17 +409,17 @@ class MBartDecoderLayer(nn.Module):
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states (`torch.FloatTensor`): input to the layer of shape *(seq_len, batch, embed_dim)*
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
             attention_mask (`torch.FloatTensor`): attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             encoder_hidden_states (`torch.FloatTensor`):
-                cross attention input to the layer of shape *(seq_len, batch, embed_dim)*
+                cross attention input to the layer of shape `(seq_len, batch, embed_dim)`
             encoder_attention_mask (`torch.FloatTensor`): encoder attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
-                *(encoder_attention_heads,)*.
+                `(encoder_attention_heads,)`.
             cross_attn_layer_head_mask (`torch.FloatTensor`): mask for cross-attention heads in a given layer of
-                size *(decoder_attention_heads,)*.
+                size `(decoder_attention_heads,)`.
             past_key_value (`Tuple(torch.FloatTensor)`): cached past key and value projection states
             output_attentions (`bool`, *optional*):
                 Whether or not to return the attentions tensors of all attention layers. See `attentions` under

src/transformers/models/pegasus/modeling_pegasus.py

@@ -322,11 +322,11 @@ class PegasusEncoderLayer(nn.Module):
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states (`torch.FloatTensor`): input to the layer of shape *(seq_len, batch, embed_dim)*
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
             attention_mask (`torch.FloatTensor`): attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
-                *(encoder_attention_heads,)*.
+                `(encoder_attention_heads,)`.
             output_attentions (`bool`, *optional*):
                 Whether or not to return the attentions tensors of all attention layers. See `attentions` under
                 returned tensors for more detail.
@@ -406,17 +406,17 @@ class PegasusDecoderLayer(nn.Module):
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states (`torch.FloatTensor`): input to the layer of shape *(seq_len, batch, embed_dim)*
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
             attention_mask (`torch.FloatTensor`): attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             encoder_hidden_states (`torch.FloatTensor`):
-                cross attention input to the layer of shape *(seq_len, batch, embed_dim)*
+                cross attention input to the layer of shape `(seq_len, batch, embed_dim)`
             encoder_attention_mask (`torch.FloatTensor`): encoder attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
-                *(encoder_attention_heads,)*.
+                `(encoder_attention_heads,)`.
             cross_attn_layer_head_mask (`torch.FloatTensor`): mask for cross-attention heads in a given layer of
-                size *(decoder_attention_heads,)*.
+                size `(decoder_attention_heads,)`.
             past_key_value (`Tuple(torch.FloatTensor)`): cached past key and value projection states
             output_attentions (`bool`, *optional*):
                 Whether or not to return the attentions tensors of all attention layers. See `attentions` under

src/transformers/models/xglm/modeling_xglm.py

@@ -431,17 +431,17 @@ class XGLMDecoderLayer(nn.Module):
     ) -> torch.Tensor:
         """
         Args:
-            hidden_states (`torch.FloatTensor`): input to the layer of shape *(seq_len, batch, embed_dim)*
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(seq_len, batch, embed_dim)`
             attention_mask (`torch.FloatTensor`): attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             encoder_hidden_states (`torch.FloatTensor`):
-                cross attention input to the layer of shape *(seq_len, batch, embed_dim)*
+                cross attention input to the layer of shape `(seq_len, batch, embed_dim)`
             encoder_attention_mask (`torch.FloatTensor`): encoder attention mask of size
-                *(batch, 1, tgt_len, src_len)* where padding elements are indicated by very large negative values.
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
-                *(encoder_attention_heads,)*.
+                `(encoder_attention_heads,)`.
             cross_attn_layer_head_mask (`torch.FloatTensor`): mask for cross-attention heads in a given layer of
-                size *(decoder_attention_heads,)*.
+                size `(decoder_attention_heads,)`.
             past_key_value (`Tuple(torch.FloatTensor)`): cached past key and value projection states
             output_attentions (`bool`, *optional*):
                 Whether or not to return the attentions tensors of all attention layers. See `attentions` under