Why not? You have far more results to review.
If you run a model on slower hardware are you getting more experience? Surely its a factor of model output reviewed and not human time.
If you run a model on slower hardware are you getting more experience? Surely its a factor of model output reviewed and not human time.