Exploiting Activation Sparsity with Dense to Dynamic-k Mixture-of-Experts Conversion

Item #:

079017-1370

Details

DOI: https://doi.org/10.52202/079017-1370
Author(s): Mikołaj Piórczyński, Simone Scardapane, Filip Szatkowski, Bartosz Wójcik
Pages: 43245-43273 (29 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 37
Date/Location: Held 10-15 December 2024, Vancouver, Canada.
Series: Advances in Neural Information Processing Systems 37
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)

DOI: https://doi.org/10.52202/079017-1370
Author(s): Mikołaj Piórczyński, Simone Scardapane, Filip Szatkowski, Bartosz Wójcik
Pages: 43245-43273 (29 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 37
Date/Location: Held 10-15 December 2024, Vancouver, Canada.
Series: Advances in Neural Information Processing Systems 37
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)