Lecture 30: Quantized Training

Thien Tran discussed quantized training including low bit optimizers and full int8 precision training in torchao Slides and notebook available here https://github.com/gpu-mode/lectures/tree/main/lecture_030