Search Results for author:
Found 1 papers, 0 papers with code
Optimizing Large Model Training through Overlapped Activation Recomputation
Our comprehensive evaluation using GPT models with 1. 3B-20B parameters shows that both OPT and HEU outperform the state-of-the-art recomputation approaches (e. g., Megatron-LM and Checkmake) by 1. 02-1. 53x.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
