Skip to main content

Imagenet Training report

Created on November 30|Last edited on November 30

Training Metrics for the two models

Notations:

  • happy-wildflower-150: 4 layers Attention Only
  • hearty-paper-146: 4 layers Attention+MLP

020k40k60k80kStep50100150200250300
20k40k60k80kStep1400160018002000220024002600
Run set
8