fix GPT-2 tokenization to work also on python 3...

pytorch_pretrained_bert/tokenization_gpt2.py

@@ -220,7 +220,7 @@ class GPT2Tokenizer(object):
         """ Tokenize a string. """
         bpe_tokens = []
         for token in re.findall(self.pat, text):
-            token = ''.join(self.byte_encoder[ord(b)] for b in token.encode('utf-8'))
+            token = ''.join(self.byte_encoder[ord(b)] for b in token)
             bpe_tokens.extend(bpe_token for bpe_token in self.bpe(token).split(' '))
         return bpe_tokens