@misc{oswald2022transformers,
  title={Transformers learn in-context by gradient descent},
  author={Johannes von Oswald and Eyvind Niklasson and Ettore Randazzo and João Sacramento and Alexander Mordvintsev and Andrey Zhmoginov and Max Vladymyrov},
  year={2022},
  eprint={2212.07677},
  archivePrefix={arXiv},
  primaryClass={cs.LG},
  url={https://arxiv.org/abs/2212.07677}
}