Data-Selection/BSL-1B

BSL-1B

paper | code

BSL-1B is a 1B model with Mistral achitecture pre-trained from scratch on the CC split of Redpajama.

It is used as the baseline for PDS-1B.

Evaluation

PDS-selected data improves the performance of language models pre-trained from scratch and saves pre-training comptation. The improvement scales up to large model sizes.

Citation

@article{gu2024data,
  title={Data Selection via Optimal Control for Language Models},
  author={Gu, Yuxian and Dong, Li and Wang, Hongning and Hao, Yaru and Dong, Qingxiu and Wei, Furu and Huang, Minlie},
  journal={arXiv preprint arXiv:2410.07064},
  year={2024}
}