fix-vocab-size (#17)

- fix vocab size (15c1aa443e6aa5877a0aef051055cf54ea01c85c)

tokenization_kimi.py

@@ -104,7 +104,7 @@ class TikTokenTokenizer(PreTrainedTokenizer):
         self.special_tokens = {
             special_tokens_mapping.get(i, f"<|reserved_token_{i}|>"): i
             for i in range(
-                num_base_tokens, num_base_tokens + self.num_reserved_special_tokens + 2
+                num_base_tokens, num_base_tokens + self.num_reserved_special_tokens
             )
         }