Don't add `bos` twice (#8)

- Don't add `bos` twice (d99561534d78bdf5ba991a406d31b6788550b14a)

model.py

@@ -37,7 +37,7 @@ def get_prompt(message: str, chat_history: list[tuple[str, str]],
 
 def get_input_token_length(message: str, chat_history: list[tuple[str, str]], system_prompt: str) -> int:
     prompt = get_prompt(message, chat_history, system_prompt)
-    input_ids = tokenizer([prompt], return_tensors='np')['input_ids']
+    input_ids = tokenizer([prompt], return_tensors='np', add_special_tokens=False)['input_ids']
     return input_ids.shape[-1]
 
 
@@ -49,7 +49,7 @@ def run(message: str,
         top_p: float = 0.95,
         top_k: int = 50) -> Iterator[str]:
     prompt = get_prompt(message, chat_history, system_prompt)
-    inputs = tokenizer([prompt], return_tensors='pt').to('cuda')
+    inputs = tokenizer([prompt], return_tensors='pt', add_special_tokens=False).to('cuda')
 
     streamer = TextIteratorStreamer(tokenizer,
                                     timeout=10.,