Skip to content

Add additional arguments to TransformerLayer subclasses #7377

Add additional arguments to TransformerLayer subclasses

Add additional arguments to TransformerLayer subclasses #7377

L2_Megatron_Change_Partitions_Increase_TP_Num_Partitions_-2_to_4-_and_PP_Num_Partitions_-1_to_2  /  main

succeeded Jan 9, 2025 in 1m 59s