Skip to content

Commit

Permalink
Merge
Browse files Browse the repository at this point in the history
  • Loading branch information
SujeethJinesh committed Jan 9, 2025
2 parents 1543d20 + 07789f1 commit e5114d4
Show file tree
Hide file tree
Showing 2 changed files with 2 additions and 1 deletion.
2 changes: 1 addition & 1 deletion MaxText/checkpointing.py
Original file line number Diff line number Diff line change
Expand Up @@ -126,7 +126,7 @@ def create_orbax_emergency_replicator_checkpoint_manager(
save_interval_steps=save_interval_steps,
)
manager = emergency_replicator_checkpoint_manager.ReplicatorCheckpointManager(
local_checkpoint_dir,
epath.Path(local_checkpoint_dir),
options,
global_mesh=global_mesh,
)
Expand Down
1 change: 1 addition & 0 deletions MaxText/train.py
Original file line number Diff line number Diff line change
Expand Up @@ -960,6 +960,7 @@ def train_loop(config, state=None):
record_goodput(recorder, config, recorder.record_job_end_time if recorder else None)
clear_buffered_metrics()
with mesh, nn_partitioning.axis_rules(config.logical_axis_rules):
# pytype: disable=attribute-error
compiled = p_train_step.lower(state, example_batch, nextrng).compile()
compiled_stats = compiled.memory_analysis()
max_logging.log(
Expand Down

0 comments on commit e5114d4

Please sign in to comment.