Any plan to add the stop token loss?

[pcgreat]

I notice that you added stop token in the model, but didn't calculate the stop_token in the loss. I tried to add stop_token myself, but then the model no longer converges. So I wonder if you have any plan to implement that?

[soobinseo]

Adding stop token is experimenting. Likewise, I do not seem to converge in the way when adding stop loss in my implementation, and I would appreciate it if you could give me a good test result.