Update app.py

app.py

@@ -20,10 +20,10 @@ quantization_config = BitsAndBytesConfig(
 model = AutoModelForCausalLM.from_pretrained(
     "shisa-ai/shisa-v1-qwen2-7b", quantization_config=quantization_config)
 tok = AutoTokenizer.from_pretrained("shisa-ai/shisa-v1-qwen2-7b")
-terminators = [
-    tok.eos_token_id,
-    tok.convert_tokens_to_ids("<|eot_id|>")
-]
+#terminators = [
+#    tok.eos_token_id,
+#    tok.convert_tokens_to_ids("<|eot_id|>")
+#]
 
 if torch.cuda.is_available():
     device = torch.device("cuda")
@@ -55,7 +55,7 @@ def chat(message, history, temperature,do_sample, max_tokens):
         max_new_tokens=max_tokens,
         do_sample=True,
         temperature=temperature,
-        eos_token_id=terminators,
+        eos_token_id=tok.eos_token_id,  # terminatorsをeos_token_idに変更
     )
     
     if temperature == 0: