slicing

app.py

@@ -15,8 +15,9 @@ else:
 def generate_response(prompt):
     inputs = tokenizer(prompt, return_tensors="pt")
     outputs = model.generate(**inputs, max_new_tokens=5, temperature=1.0)
-    full_output = tokenizer.decode(outputs[0], skip_special_tokens=False)
-    new_tokens = full_output[len(inputs):]
+    input_length = inputs['input_ids'].shape[1]
+    new_token_ids = outputs[0][input_length:]
+    new_tokens = tokenizer.decode(new_token_ids, skip_special_tokens=False)
     return new_tokens
 
 iface = gr.Interface(