MaxView

← Back to run

Log Summary

XPK Start: Thu Apr 16 22:35:24 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-16 22:35:47.935463: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303)
I0416 22:35:48.113301 132730069661504 max_utils.py:273] Attempting to initialize the jax distributed system...
I0416 22:35:57.153765 132730069661504 distributed.py:149] Starting JAX distributed service on [::]:8482
I0416 22:35:57.156602 132730069661504 distributed.py:172] Connecting to JAX distributed service on mt-07-eval-sk01k-slice-job-0-0.mt-07-eval-sk01k:8482
I0416 22:35:58.299477 132730069661504 max_utils.py:284] Jax distributed system initialized!
I0416 22:36:03.428548 132730069661504 max_utils.py:800] System Information: Jax Version: 0.9.2
I0416 22:36:03.428648 132730069661504 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0416 22:36:03.428686 132730069661504 max_utils.py:802] System Information: Jax Backend: PJRT C API
TFRT TPU v6 lite
Built on Mar 4 2026 11:32:08 (1772652728) cl/878335365
I0416 22:36:03.428719 132730069661504 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0416 22:36:04.096147 132730069661504 maxtext_utils.py:1687] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0416 22:36:04.208025 132730069661504 checkpointing.py:688] Setting up checkpoint logger...
I0416 22:36:04.208140 132730069661504 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0416 22:36:04.208184 132730069661504 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0416 22:36:04.208374 132730069661504 base_pytree_checkpoint_handler.py:441] Created BasePyTreeCheckpointHandler: use_ocdbt=True, use_zarr3=True, pytree_metadata_options=PyTreeMetadataOptions(support_rich_types=False), array_metadata_store=<orbax.checkpoint._src.metadata.array_metadata_store.Store object at 0x78b707b18410>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0416 22:36:07.550352 132730069661504 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0416 22:36:07.550585 132730069661504 checkpoint_manager.py:708] [process=0][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78a35862c050>}, handler_registry=None
I0416 22:36:07.550822 132730069661504 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78a35862c050>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`.
I0416 22:36:07.550885 132730069661504 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78a37010c650>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`.
I0416 22:36:07.550922 132730069661504 composite_checkpoint_handler.py:505] Initialized registry DefaultCheckpointHandlerRegistry({('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78a35862c050>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78a35862c050>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78a37010c650>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78a37010c650>}).
I0416 22:36:07.551244 132730069661504 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0416 22:36:07.551314 132730069661504 async_checkpointer.py:192] [process=0][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x78a35868a020> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0416 22:36:08.221239 132730069661504 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260416_221011/nnx_xpk_feat_nnx_post_train_fixes_20260416_221011_07_eval/checkpoints
I0416 22:36:08.659732 132730069661504 checkpoint_manager.py:929] [process=0][thread=MainThread] CheckpointManager created,  primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260416_221011/nnx_xpk_feat_nnx_post_train_fixes_20260416_221011_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x78a35862ef90>
I0416 22:36:08.659948 132730069661504 checkpointing.py:302] Checkpoint manager created!
I0416 22:36:09.608909 132730069661504 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0416 22:36:09.609025 132730069661504 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0416 22:36:10.994278 132730069661504 maxtext_utils.py:1805]  decoder/decoder_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0416 22:36:10.994393 132730069661504 maxtext_utils.py:1805]  decoder/layers/mlp/wi_0/kernel/value
    Shape:     float32[2048,16,7168]
    Physical:  ('fsdp', None, None)
I0416 22:36:10.994447 132730069661504 maxtext_utils.py:1805]  decoder/layers/mlp/wi_1/kernel/value
    Shape:     float32[2048,16,7168]
    Physical:  ('fsdp', None, None)
I0416 22:36:10.994490 132730069661504 maxtext_utils.py:1805]  decoder/layers/mlp/wo/kernel/value
    Shape:     float32[7168,16,2048]
    Physical:  (None, None, 'fsdp')
I0416 22:36:10.994526 132730069661504 maxtext_utils.py:1805]  decoder/layers/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048,16]
    Physical:  (None, None)
I0416 22:36:10.994560 132730069661504 maxtext_utils.py:1805]  decoder/layers/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048,16]
    Physical:  (None, None)
I0416 22:36:10.994595 132730069661504 maxtext_utils.py:1805]  decoder/layers/self_attention/key/kernel/value
    Shape:     float32[2048,16,16,128]
    Physical:  ('fsdp', None, None, None)
I0416 22:36:10.994630 132730069661504 maxtext_utils.py:1805]  decoder/layers/self_attention/out/kernel/value
    Shape:     float32[16,16,128,2048]
    Physical:  (None, None, None, 'fsdp')
I0416 22:36:10.994663 132730069661504 maxtext_utils.py:1805]  decoder/layers/self_attention/query/kernel/value
    Shape:     float32[2048,16,16,128]
    Physical:  ('fsdp', None, None, None)
I0416 22:36:10.994695 132730069661504 maxtext_utils.py:1805]  decoder/layers/self_attention/value/kernel/value
    Shape:     float32[2048,16,16,128]
    Physical:  ('fsdp', None, None, None)
I0416 22:36:10.994726 132730069661504 maxtext_utils.py:1805]  decoder/logits_dense/kernel/value
    Shape:     float32[2048,32000]
    Physical:  ('fsdp', None)
I0416 22:36:10.994757 132730069661504 maxtext_utils.py:1805]  token_embedder/embedding/value
    Shape:     float32[32000,2048]

    Physical:  (None, 'fsdp')
I0416 22:36:11.392358 132730069661504 nnx_decoders.py:465] nnx_decoders/carry Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0416 22:36:11.392450 132730069661504 nnx_decoders.py:465] nnx_decoders/carry Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0416 22:36:11.397727 132730069661504 nnx_decoders.py:465] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0416 22:36:11.397783 132730069661504 nnx_decoders.py:465] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0416 22:36:11.414057 132730069661504 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0416 22:36:11.414127 132730069661504 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0416 22:36:11.429718 132730069661504 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0416 22:36:11.429774 132730069661504 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0416 22:36:11.453647 132730069661504 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0416 22:36:11.453719 132730069661504 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0416 22:36:11.469421 132730069661504 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0416 22:36:11.469481 132730069661504 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0416 22:36:11.485111 132730069661504 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0416 22:36:11.485180 132730069661504 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0416 22:36:11.514607 132730069661504 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0416 22:36:11.514678 132730069661504 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0416 22:36:11.534430 132730069661504 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0416 22:36:11.534492 132730069661504 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0416 22:36:22.527027 132730069661504 max_utils.py:791] Total memory size: 1.5 GB, Output size: 0.4 GB, Temp size: 1.1 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0416 22:36:22.536111 132730069661504 metric_logger.py:289] number parameters: 1.104 billion
I0416 22:36:33.648274 132730069661504 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0416 22:36:33.800854 132730069661504 checkpointing.py:798] Waited 0.15254521369934082 seconds for step 0 to finish before starting checkpointing.
I0416 22:36:33.803461 132730069661504 checkpoint_manager.py:2009] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0416 22:36:33.805088 132730069661504 checkpoint_manager.py:1512] [process=0] Saving checkpoint at step 0
I0416 22:36:33.806556 132730069661504 event_tracking.py:70] [process=0] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260416_221011/nnx_xpk_feat_nnx_post_train_fixes_20260416_221011_07_eval/checkpoints/0.
I0416 22:36:34.509541 132730069661504 signaling_client.py:364] Using JaxDistributedSignalingClient
I0416 22:36:34.528825 132730069661504 jax_array_handlers.py:360] Scheduling D2H of 69 prioritized jax.Array.
I0416 22:36:34.528912 132730069661504 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False
I0416 22:36:34.764171 132603751425792 atomicity.py:140] Creating tmp directory gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260416_221011/nnx_xpk_feat_nnx_post_train_fixes_20260416_221011_07_eval/checkpoints/0
I0416 22:36:34.801213 132730069661504 base_pytree_checkpoint_handler.py:154] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.273795s
I0416 22:36:34.801984 132730069661504 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/blocking_gbytes_per_sec: 5.421 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.284534215927124 s) (per-host)
I0416 22:36:34.802059 132730069661504 base_pytree_checkpoint_handler.py:768] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.284630s (batch_requests_ready=0.003047s, total_serialization_initiated=0.280895s, others=0.000687s)
I0416 22:36:34.802146 132730069661504 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.288563s (all_items=0.000018s, per_item={'items': '0.00001788'}, temp_paths=0.288545)
I0416 22:36:34.802861 132730069661504 event_tracking.py:125] [process=0] [async] Finished blocking save in 1.00 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260416_221011/nnx_xpk_feat_nnx_post_train_fixes_20260416_221011_07_eval/checkpoints/0.
I0416 22:36:34.803132 132603759818496 async_checkpointer.py:76] [process=0][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-16 22:56:34.803104
I0416 22:36:34.811417 132730069661504 checkpoint_manager.py:1560] [process=0][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0416 22:36:34.811678 132603180984064 async_checkpointer.py:280] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0416 22:36:34.811799 132730069661504 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260416_221011/nnx_xpk_feat_nnx_post_train_fixes_20260416_221011_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776378993.8034432, 'wait_for_prev_duration_secs': 6.222724914550781e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776378993.8051248, 'checkpointer_blocking_duration_secs': 0.9981095790863037, 'get_old_steps_start_time': 1776378994.803258, 'get_old_steps_duration_secs': 3.0517578125e-05, 'checkpoint_manager_blocking_start_time': 1776378993.8012476, 'checkpoint_manager_blocking_duration_secs': 1.0105137825012207}
I0416 22:36:34.811924 132730069661504 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0416 22:36:34.811992 132730069661504 max_utils.py:750] 
Memstats: After params initialized:
I0416 22:36:34.812045 132730069661504 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_0(process=0,(0,0,0,0))
I0416 22:36:34.812083 132730069661504 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_1(process=0,(1,0,0,0))
I0416 22:36:34.812110 132730069661504 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_4(process=0,(0,1,0,0))
I0416 22:36:34.812135 132730069661504 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_5(process=0,(1,1,0,0))
I0416 22:36:35.155272 132730069661504 metric_logger.py:185] completed step: 0, seconds: 10.815, TFLOP/s/device: 1.256, Tokens/s/device: 189.366, total_weights: 65536, loss: 10.874
I0416 22:36:35.158097 132730069661504 metric_logger.py:269] To see full metrics 'tensorboard --logdir=gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260416_221011/nnx_xpk_feat_nnx_post_train_fixes_20260416_221011_07_eval/tensorboard/'
I0416 22:36:35.592418 132730069661504 metric_logger.py:185] completed step: 1, seconds: 1.504, TFLOP/s/device: 9.036, Tokens/s/device: 1362.073, total_weights: 65536, loss: 10.874
I0416 22:36:35.734346 132730069661504 metric_logger.py:185] completed step: 2, seconds: 0.438, TFLOP/s/device: 31.014, Tokens/s/device: 4674.764, total_weights: 65536, loss: 10.262
I0416 22:36:35.881778 132730069661504 metric_logger.py:185] completed step: 3, seconds: 0.012, TFLOP/s/device: 1131.035, Tokens/s/device: 170481.978, total_weights: 65536, loss: 9.732
I0416 22:36:36.694136 132603734640384 atomicity.py:140] Creating tmp directory gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260416_221011/nnx_xpk_feat_nnx_post_train_fixes_20260416_221011_07_eval/checkpoints/0/items
I0416 22:36:36.826036 132601582970624 checkpoint.py:188] Wrote Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776378996403023502, 'commit_timestamp_nsecs': None, 'custom_metadata': {}}, json={"item_handlers": null, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1776378996403023502, "commit_timestamp_nsecs": null, "custom_metadata": {}} to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260416_221011/nnx_xpk_feat_nnx_post_train_fixes_20260416_221011_07_eval/checkpoints/0/_CHECKPOINT_METADATA
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/deps/src/maxtext/trainers/pre_train/train.py", line 852, in <module>
    app.run(main)
  File "/usr/local/lib/python3.12/site-packages/absl/app.py", line 367, in run
    _run_main(main, args)
  File "/usr/local/lib/python3.12/site-packages/absl/app.py", line 312, in _run_main
    sys.exit(main(argv))
             ^^^^^^^^^^
  File "/deps/src/maxtext/trainers/pre_train/train.py", line 848, in main
    train_func()
  File "/deps/src/maxtext/trainers/pre_train/train.py", line 838, in train_func
    run(config, recorder, diagnostic_config)
  File "/deps/src/maxtext/trainers/pre_train/train.py", line 817, in run
    train_loop(config, recorder)
  File "/deps/src/maxtext/trainers/pre_train/train.py", line 715, in train_loop
    assert eval_data_iterator
           ^^^^^^^^^^^^^^^^^^
AssertionError
I0416 22:36:37.909063    2756 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
E0416 22:36:37.913418 132603189376768 future.py:359] [process=0][thread=write_metadata_after_commits][operation_id=1] _SignalingThread.run() raised an exception: cannot schedule new futures after interpreter shutdown
Traceback (most recent call last):
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 356, in run
    super().run()
  File "/usr/local/lib/python3.12/threading.py", line 1012, in run
    self._target(*self._args, **self._kwargs)
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 318, in _target_setting_result
    self._result = target()
                   ^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 435, in <lambda>
    target=lambda: asyncio_utils.run_sync(coro),
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/asyncio_utils.py", line 50, in run_sync
    return asyncio.run(coro)
           ^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/runners.py", line 195, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/base_events.py", line 691, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/handlers/base_pytree_checkpoint_handler.py", line 1251, in _write_metadata_after_commits
    await asyncio.to_thread(commit_future.result)
  File "/usr/local/lib/python3.12/asyncio/threads.py", line 25, in to_thread
    return await loop.run_in_executor(None, func_call)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/base_events.py", line 867, in run_in_executor
    executor.submit(func, *args), loop=self)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/concurrent/futures/thread.py", line 173, in submit
    raise RuntimeError('cannot schedule new futures after '
RuntimeError: cannot schedule new futures after interpreter shutdown
E0416 22:36:37.924304 132603197769472 future.py:359] [process=0][thread=array_type_handler][operation_id=1] _SignalingThread.run() raised an exception: cannot schedule new futures after interpreter shutdown
Traceback (most recent call last):
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 356, in run
    super().run()
  File "/usr/local/lib/python3.12/threading.py", line 1012, in run
    self._target(*self._args, **self._kwargs)
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 318, in _target_setting_result
    self._result = target()
                   ^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 435, in <lambda>
    target=lambda: asyncio_utils.run_sync(coro),
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/asyncio_utils.py", line 50, in run_sync
    return asyncio.run(coro)
           ^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/runners.py", line 195, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/base_events.py", line 691, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/serialization/jax_array_handlers.py", line 376, in _serialize_without_dispatcher
    await async_serialize_replica_slices_batch(
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/serialization/jax_array_handlers.py", line 625, in _async_serialize_replica_slices
    await asyncio.gather(*write_coros)
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/metadata/array_metadata_store.py", line 194, in write
    await self._maybe_create_base_dir(file_path.parent)
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/metadata/array_metadata_store.py", line 154, in _maybe_create_base_dir
    return await async_path.mkdir(base_dir, parents=True, exist_ok=True)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/path/async_path.py", line 50, in mkdir
    return await asyncio.to_thread(_mkdir_sync, **thread_kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/threads.py", line 25, in to_thread
    return await loop.run_in_executor(None, func_call)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/base_events.py", line 867, in run_in_executor
    executor.submit(func, *args), loop=self)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/concurrent/futures/thread.py", line 173, in submit
    raise RuntimeError('cannot schedule new futures after '
RuntimeError: cannot schedule new futures after interpreter shutdown
E0416 22:36:37.950725 132603759818496 async_checkpointer.py:244] [process=0] Failed to run 3 Handler Commit operations or the Commit callback in background save thread, directory: gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260416_221011/nnx_xpk_feat_nnx_post_train_fixes_20260416_221011_07_eval/checkpoints/0
Traceback (most recent call last):
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/checkpointers/async_checkpointer.py", line 229, in _thread_func
    _background_wait_for_commit_futures(
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/checkpointers/async_checkpointer.py", line 86, in _background_wait_for_commit_futures
    future.ChainedFuture(commit_futures, cb=lambda: None).result(
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 198, in result
    f.result(timeout=get_remaining_time(start, timeout))
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 198, in result
    f.result(timeout=get_remaining_time(start, timeout))
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 506, in result
    return self._f.result(timeout=timeout)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 442, in result
    return self._t.result(timeout=timeout)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 391, in result
    self.join(timeout=timeout)
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 388, in join
    raise self._exception
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 356, in run
    super().run()
  File "/usr/local/lib/python3.12/threading.py", line 1012, in run
    self._target(*self._args, **self._kwargs)
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 318, in _target_setting_result
    self._result = target()
                   ^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/futures/future.py", line 435, in <lambda>
    target=lambda: asyncio_utils.run_sync(coro),
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/asyncio_utils.py", line 50, in run_sync
    return asyncio.run(coro)
           ^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/runners.py", line 195, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/base_events.py", line 691, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/orbax/checkpoint/_src/handlers/base_pytree_checkpoint_handler.py", line 1251, in _write_metadata_after_commits
    await asyncio.to_thread(commit_future.result)
  File "/usr/local/lib/python3.12/asyncio/threads.py", line 25, in to_thread
    return await loop.run_in_executor(None, func_call)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/base_events.py", line 867, in run_in_executor
    executor.submit(func, *args), loop=self)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/concurrent/futures/thread.py", line 173, in submit
    raise RuntimeError('cannot schedule new futures after '
RuntimeError: cannot schedule new futures after interpreter shutdown
I0416 22:36:37.971694 132603180984064 async_checkpointer.py:288] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0416 22:36:37.971793 132603180984064 checkpoint_manager.py:2137] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0416 22:36:37.975696 132603180984064 checkpoint_manager.py:2146] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Thu Apr 16 22:36:49 UTC 2026
EXIT_CODE=1