Update App.py

App.py

@@ -14,10 +14,12 @@ model = AutoModelForCausalLM.from_pretrained(
 
 # Simple function: one input → one output
 def ask_model(prompt):
-    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    # Add a guiding instruction
+    input_text = f"Answer briefly and clearly: {prompt}"
+    inputs = tokenizer(input_text, return_tensors="pt").to(model.device)
     outputs = model.generate(
         **inputs,
-        max_new_tokens=200,
+        max_new_tokens=150,   # reduce length so responses don’t get too long
         pad_token_id=tokenizer.eos_token_id
     )
     response = tokenizer.decode(outputs[0], skip_special_tokens=True)