Update app.py

app.py

@@ -26,8 +26,7 @@ def generate_text(inp, history):
     input_ids = tokenizer(inp, return_tensors='pt').to("cuda")
     beam_output = model.generate(**input_ids, generation_config=generation_config)
     output = tokenizer.decode(beam_output[0], skip_special_token=True)
-    output = output.replace(inp,"")
-    print(output)
+    output = output.replace(inp,"").replace("<s>","").replace("</s>","")
     return output
 
 gr.ChatInterface(generate_text).launch()