Update app.py

app.py

@@ -1,4 +1,5 @@
-import gradio as gr
+import
+ gradio as gr
 from peft import PeftModel, PeftConfig
 from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
 import torch
@@ -25,10 +26,9 @@ model = PeftModel.from_pretrained(model, model_id)
 
 def greet(text):
     with torch.no_grad():  # Disable gradient calculation for inference
-        batch = tokenizer(f'"{text}" ->:', return_tensors='pt')  # Move tensors to device
+        batch = tokenizer(f'### Human: {text}', return_tensors='pt')  # Move tensors to device
         with torch.cuda.amp.autocast():  # Enable mixed-precision if available
-            output_tokens = model.generate(**batch
-, max_new_tokens=15)
+            output_tokens = model.generate(**batch, max_new_tokens=25)
     return tokenizer.decode(output_tokens[0], skip_special_tokens=True)
 
 iface = gr.Interface(fn=greet, inputs="text", outputs="text", title="PEFT Model for Big Brain")