2024/07/10 by Sami Jaghouar, Jaghouar, Sami, Jack Min Ong +3 · 1 voice · 1 citation
#cs.LG #cs.DC
paper · pdf · doi:10.48550/arxiv.2407.07852
OpenDiLoCo is an open-source implementation and replication of the Distributed Low-Communication (DiLoCo) training method for large language models. We provide a reproducible implementation of the DiLoCo experiments, offering it within a scalable, decentralized training framework using the Hivemind library. We demonstrate its effectiveness by training a model across two continents and three countries, while maintaining 90-95% compute utilization. Additionally, we conduct ablations studies focusing on the algorithm's compute efficiency, scalability in the number of workers and show that its gradients can be all-reduced using FP16 without any performance degradation. Furthermore, we scale OpenDiLoCo to 3x the size of the original work, demonstrating its effectiveness for billion parameter models.