Update app.py

app.py

@@ -21,7 +21,7 @@ def make_inference(synopsis):
     )
 
     with torch.cuda.amp.autocast():
-        output_tokens = model.generate(**batch, max_new_tokens=600)
+        output_tokens = model.generate(**batch, max_new_tokens=400, temperature = 0.9)
 
     full_output = tokenizer.decode(output_tokens[0], skip_special_tokens=True)
     short_story = full_output.split("### Short Story:\n")[-1].strip()