Zobrazeno 1 - 1
of 1
pro vyhledávání: '"Benington, Michael"'
Scaling Studies for Efficient Parameter Search and Parallelism for Large Language Model Pre-training
Autor:
Benington, Michael, Phan, Leo, Paul, Chris Pierre, Shoemaker, Evan, Ranade, Priyanka, Collett, Torstein, Perez, Grant Hodgson, Krieger, Christopher
Publikováno v:
Supercomputing 2023 (SC23) Student Research Poster Track
AI accelerator processing capabilities and memory constraints largely dictate the scale in which machine learning workloads (e.g., training and inference) can be executed within a desirable time frame. Training a state of the art, transformer-based m
Externí odkaz:
http://arxiv.org/abs/2310.05350
