t

app.py

@@ -52,7 +52,7 @@ def get_completion(msg):
 
     with torch.no_grad():
         outputs = peft_model.generate(
-            **inputs, max_new_tokens=384, pad_token_id = tokenizer.eos_token_id
+            **inputs, max_new_tokens=512, pad_token_id = tokenizer.eos_token_id
         )
 
     #peft_model.to(cpu_device)
@@ -75,5 +75,5 @@ def greet(input):
 	###return json.dumps(extract_json(gen_text, -1))
 	return response
 
-demo = gr.Interface(fn=greet, inputs=[gr.Textbox(label="Elevator pitcher", lines=3)], outputs=gr.Text())
+demo = gr.Interface(fn=greet, inputs=[gr.Textbox(label="Elevator pitcher", lines=1)], outputs=gr.Text())
 demo.launch()