Fix annotations (#24571)

* fix annotations

* fix copies

src/transformers/models/blenderbot/modeling_blenderbot.py

@@ -307,7 +307,7 @@ 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 `(batch, seq_len, 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.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size

src/transformers/models/m2m_100/modeling_m2m_100.py

@@ -375,7 +375,7 @@ 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 `(batch, seq_len, 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.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size

src/transformers/models/mbart/modeling_mbart.py

@@ -315,7 +315,7 @@ 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 `(batch, seq_len, 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.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size

src/transformers/models/pegasus/modeling_pegasus.py

@@ -322,7 +322,7 @@ 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 `(batch, seq_len, 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.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size

src/transformers/models/speech_to_text/modeling_speech_to_text.py

@@ -382,7 +382,7 @@ class Speech2TextEncoderLayer(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 `(batch, seq_len, 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.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size

src/transformers/models/whisper/modeling_whisper.py

@@ -491,7 +491,7 @@ class WhisperEncoderLayer(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 `(batch, seq_len, 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.
             layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size