Packing Analysis: Packing Is More Appropriate for Large Models or Datasets in Supervised Fine-tuning.

Shuhe Wang Guoyin Wang 0002 Jiwei Li 0001 Eduard H. Hovy Chen Guo Packing Analysis: Packing Is More Appropriate for Large Models or Datasets in Supervised Fine-tuning. 2024 abs/2410.08081 CoRR https://doi.org/10.48550/arXiv.2410.08081 db/journals/corr/corr2410.html#abs-2410-08081 streams/journals/corr