retraining-free-pruning
retraining-free-pruning copied to clipboard
Hi, can this "the three-stage decomposition of the pruning process" be applied to GPT-X or any other NLG task? And how this could be done?