fix bug in prepare inputs for language generation for xlm for effective batch_size > 1

src/transformers/modeling_xlm.py

@@ -674,7 +674,8 @@ class XLMWithLMHeadModel(XLMPreTrainedModel):
         mask_token_id = self.config.mask_token_id
         lang_id = self.config.lang_id
 
-        mask_token = torch.full((1, 1), mask_token_id, dtype=torch.long, device=input_ids.device)
+        effective_batch_size = input_ids.shape[0]
+        mask_token = torch.full((effective_batch_size, 1), mask_token_id, dtype=torch.long, device=input_ids.device)
         input_ids = torch.cat([input_ids, mask_token], dim=1)
         if lang_id is not None:
             langs = torch.full_like(input_ids, lang_id)