Skip to content

Commit

Permalink
Update
Browse files Browse the repository at this point in the history
[ghstack-poisoned]
  • Loading branch information
wconstab committed May 23, 2024
2 parents 9884f42 + 17b7bec commit 44ea24c
Show file tree
Hide file tree
Showing 2 changed files with 2 additions and 2 deletions.
2 changes: 1 addition & 1 deletion multinode_trainer.slurm
Original file line number Diff line number Diff line change
Expand Up @@ -54,7 +54,7 @@ export NCCL_BUFFSIZE=2097152
#export TORCH_DIST_INIT_BARRIER=1
export FI_EFA_SET_CUDA_SYNC_MEMOPS=0
#export USE_LIBUV=1
CONFIG_FILE=${CONFIG_FILE:-"./train_configs/llama_13b.toml"}
CONFIG_FILE=${CONFIG_FILE:-"./train_configs/llama2_13b.toml"}

dcgmi profile --pause
# adjust sbatch --ntasks and sbatch --nodes above and --nnodes below
Expand Down
2 changes: 1 addition & 1 deletion torchtitan/parallelisms/parallelize_llama.py
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@
CheckpointImpl,
)
from torch.distributed.pipelining import pipeline, SplitPoint
from torch.distributed.pipelining._PipelineStage import (
from torch.distributed.pipelining.PipelineStage import (
_PipelineStage,
ManualPipelineStage,
)
Expand Down

0 comments on commit 44ea24c

Please sign in to comment.