Ref-Pretrain-Qwen-104M
Ref-Pretrain-Qwen-104M is a 104M model with Qwen achitecture conventionally pre-trained from scratch on the Pile for 5B tokens.
We also open-source the tokenized pre-training corpus for reproducibility.
It is used as the reference model in the MiniPLM knwoledge distillation framework to construct the refined pre-training corpus. The data is then used to train MiniPLM models.
Evaluation
MiniPLM models achieves better performance given the same computation and scales well across model sizes:
Citation
@article{miniplm,
title={MiniPLM: Knowledge Distillation for Pre-Training Language Models},
author={Yuxian Gu and Hao Zhou and Fandong Meng and Jie Zhou and Minlie Huang},
journal={arXiv preprint arXiv:2410.17215},
year={2024}
}