Peyton
@crazywoola is there any progress? Now, the second bug is fixed using python. But the 3rd bug still exists.
In fact, these code is only used for inference, which is correspond to the equation found in our paper. As you can see, X is just part of the search...
Thanks for you caring. I have fixed the code in https://github.com/pprp/Pruner-Zero/commit/db70134ed7d4c521a57060fd698f353d04268429. Please let me know if there are any further bugs.
Sorry, there are too many conflicts in this PR.
Thank you for your interest. In most LLM pruning studies, a 25%/50 % compression ratio is the norm; for reference, Tables 2 & 3 in our paper report experiments on...