DeepSpeed
Account for expert parameters when calculating the total number of pa…
#3720
Merged

Commits
  • Account for expert parameters when calculating the total number of parameters in the model
    Alex Dubrovsky committed 3 years ago
  • Merge branch 'master' into flopsprofilertotalparams
    alito committed 3 years ago
  • Merge branch 'master' into flopsprofilertotalparams
    alito committed 3 years ago
  • Merge branch 'master' into flopsprofilertotalparams
    alito committed 3 years ago
Loading