Key points are not available for this paper at this time.
Efficiently training LLMs with long sequences is important yet challenged by the massive computation and memory requirements. Sequence parallelism has been proposed to tackle these problems, but existing methods suffer from scalability or efficiency issues. We propose LoongTrain, a novel system to efficiently train LLMs with long sequences at scale. The core of LoongTrain is the 2D-Attention mechanism, which combines both head-parallel and context-parallel techniques to break the scalability constraints while maintaining efficiency. We introduce Double-Ring-Attention and analyze the performance of device placement strategies to further speed up training. We implement LoongTrain with the hybrid ZeRO and Selective Checkpoint++ techniques. Experiment results show that LoongTrain outperforms state-of-the-art baselines, i.e., DeepSpeed-Ulysses and Megatron Context Parallelism, in both end-to-end training speed and scalability, and improves Model FLOPs Utilization (MFU) by up to 2.88x.
Building similarity graph...
Analyzing shared references across papers
Loading...
Diandian Gu
Peng Sun
Qinghao Hu
Building similarity graph...
Analyzing shared references across papers
Loading...
Gu et al. (Wed,) studied this question.
www.synapsesocial.com/papers/68e634cdb6db6435875c63b9 — DOI: https://doi.org/10.48550/arxiv.2406.18485
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: