Update app.py

app.py

@@ -34,9 +34,9 @@ def generate_response(prompt, max_length=1024):
     inputs = tokenizer(
 [
     alpaca_prompt.format(
-                prompt, # instruction  
-                "", # input
-                "" #outputs
+        prompt, # instruction
+        "", # input
+        "", # output - leave this blank for generation!
     )
 ], return_tensors = "pt").to("cuda")
     outputs = model.generate(**inputs, max_length=max_length, pad_token_id=tokenizer.eos_token_id)