[Longformer] Correct longformer docs (#12809)

* fix_torch_device_generate_test

* remove @

* correct longformer docs

Co-authored-by: Patrick von Platen <patrick@huggingface.co>

src/transformers/models/longformer/modeling_longformer.py

@@ -1609,7 +1609,6 @@ class LongformerModel(LongformerPreTrainedModel):
             >>> SAMPLE_TEXT = ' '.join(['Hello world! '] * 1000)  # long input document
             >>> input_ids = torch.tensor(tokenizer.encode(SAMPLE_TEXT)).unsqueeze(0)  # batch of size 1
 
-            >>> # Attention mask values -- 0: no attention, 1: local attention, 2: global attention
             >>> attention_mask = torch.ones(input_ids.shape, dtype=torch.long, device=input_ids.device) # initialize to local attention
             >>> global_attention_mask = torch.zeros(input_ids.shape, dtype=torch.long, device=input_ids.device) # initialize to global attention to be deactivated for all tokens
             >>> global_attention_mask[:, [1, 4, 21,]] = 1  # Set global attention to random tokens for the sake of this example