Update gptx_tokenizer.py

gptx_tokenizer.py

@@ -245,7 +245,10 @@ class HFGPTXTokenizer(PreTrainedTokenizer):
         """
         output = self.tok.decode(input=token_ids, num_threads=num_threads)
         if skip_special_tokens:
-            output = [token for token in output if token not in self.additional_special_tokens]
+            warnings.warn(
+                "skip_special_tokens currently not implemented"
+            )
+                    
         if clean_up_tokenization_spaces:
             warnings.warn(
                 "when cleaning up tokenization spaces, this will not behave "
@@ -253,6 +256,7 @@ class HFGPTXTokenizer(PreTrainedTokenizer):
                 "`clean_up_tokenization_spaces=False` for decoding."
             )
             output = self.clean_up_tokenization(output)
+        
         return output
 
     def _convert_id_to_token(self, index: int) -> str: