Update app.py

app.py

@@ -13,6 +13,9 @@ class LlamaInterface:
         self.device = "cuda" if torch.cuda.is_available() else "cpu"
         
         self.tokenizer = AutoTokenizer.from_pretrained(base_model_name)
+        # Padding token'ı ayarla
+        self.tokenizer.pad_token = self.tokenizer.eos_token
+        
         self.model = AutoModelForCausalLM.from_pretrained(
             base_model_name,
             device_map="auto",