Skip to content
Open
Changes from 2 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions megatron/model/language_model.py
Original file line number Diff line number Diff line change
Expand Up @@ -543,6 +543,10 @@ def forward(self, enc_input_ids, enc_position_ids, enc_attn_mask,
else:
if args.curriculum_learning_legacy or args.data_efficiency_curriculum_learning:
rotary_pos_emb = self.rotary_pos_emb(args.curriculum_seqlen)
elif args.ds_sequence_parallel_size > 1:
parallel_seq_len = self.seq_length / args.ds_sequence_parallel_size
ds_sp_offset = mpu.get_sequence_parallel_rank() * parallel_seq_len
rotary_pos_emb = self.rotary_pos_emb(parallel_seq_len, ds_sp_offset)
else:
rotary_pos_emb = self.rotary_pos_emb(self.seq_length)

Expand Down