Oct, 2023
GrowLength: 通过逐步增长训练长度来加速 LLMs 预训练
GrowLength: Accelerating LLMs Pretraining by Progressively Growing Training Length
Hongye Jin, Xiaotian Han, Jingfeng Yang, Zhimeng Jiang, Chia-Yuan Chang...
TL;DR通过增加训练长度加速大型语言模型 (LLL) 的预训练过程,从而提高效率、减少计算成本,并改善性能。