Critical Batch Size Revisited: A Simple Empirical Approach to Large-Batch Language Model Training

Item #:

085713-3898

Details

DOI: https://doi.org/10.52202/085713-3898
Author(s): Will Merrill, Shane Arora, Dirk Groeneveld, Hanna Hajishirzi
Pages: 129578-129601 (24 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 38
Date/Location: Held 2-7 December 2025, San Diego, California, USA and 30 November - 5 December 2025, Mexico City, Mexico.
Series: Advances in Neural Information Processing Systems 38
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)

DOI: https://doi.org/10.52202/085713-3898
Author(s): Will Merrill, Shane Arora, Dirk Groeneveld, Hanna Hajishirzi
Pages: 129578-129601 (24 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 38
Date/Location: Held 2-7 December 2025, San Diego, California, USA and 30 November - 5 December 2025, Mexico City, Mexico.
Series: Advances in Neural Information Processing Systems 38
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)