SSRN paper: https://ssrn.com/abstract=5240330 (v1.1, last updated May 19, 2025)
Authors: Zeyuan Allen-Zhu

Slide show (best viewed on a computer)

@article{Allen2025-canon,
  author = {{Allen-Zhu}, Zeyuan},
  title = {{Physics of Language Models: Part 4.1, Architecture Design and the Magic of Canon Layers}},
  year = {2025},

  month = may,

  journal = {SSRN Electronic Journal},

  note = {\url{https://ssrn.com/abstract=5240330}} 

}