Upload tokenizer

tokenizer.py

@@ -253,6 +253,8 @@ class ProteinTokenizer(PreTrainedTokenizerFast):
         # Add special tokens
         if add_special_tokens:
             encoded_inputs["input_ids"] = [[self.bos_token_id] + seq + [self.eos_token_id] for seq in encoded_inputs["input_ids"]]
+            encoded_inputs["attention_mask"] = [1, 1] + encoded_inputs["attention_mask"]
+            encoded_inputs["special_tokens_mask"] = [1] + encoded_inputs["special_tokens_mask"] + [1]
 
         # Truncate
         if truncation: