Skip to main content

350M Param Model Training

Created on May 20|Last edited on May 20

Section 1


20406080Time (hours)2.533.5
50G100G150G200G250GStep2.533.5
100k200k300k400k500kStep3
100k200k300k400k500kStep1020304050
Run set
8



Run set
8



Run set
8



Run set
8



Run set
8