@misc{yadlowsky2023pretraining,
  title={Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models},
  author={Steve Yadlowsky and Lyric Doshi and Nilesh Tripuraneni},
  year={2023},
  eprint={2311.00871},
  archivePrefix={arXiv},
  primaryClass={cs.LG},
  url={https://arxiv.org/abs/2311.00871}
}