MaxView

XPK Start: Fri Apr 17 01:34:51 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-17 01:35:15.039266: 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)
I0417 01:35:15.565700 133638603818816 max_utils.py:273] Attempting to initialize the jax distributed system...
I0417 01:35:24.606503 133638603818816 distributed.py:149] Starting JAX distributed service on [::]:8482
I0417 01:35:24.608691 133638603818816 distributed.py:172] Connecting to JAX distributed service on mt-13-scan-layers-false-zspy9-slice-job-0-0.mt-13-scan-layers-false-zspy9:8482
I0417 01:35:26.264374 133638603818816 max_utils.py:284] Jax distributed system initialized!
I0417 01:35:32.208662 133638603818816 max_utils.py:800] System Information: Jax Version: 0.9.2
I0417 01:35:32.208760 133638603818816 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0417 01:35:32.208798 133638603818816 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
I0417 01:35:32.208832 133638603818816 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0417 01:35:32.879540 133638603818816 maxtext_utils.py:1687] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0417 01:35:33.689629 133638603818816 checkpointing.py:688] Setting up checkpoint logger...
I0417 01:35:33.689750 133638603818816 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0417 01:35:33.689802 133638603818816 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0417 01:35:33.690040 133638603818816 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 0x798a9811b140>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0417 01:35:36.616004 133638603818816 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0417 01:35:36.616239 133638603818816 checkpoint_manager.py:708] [process=1][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7976f445d850>}, handler_registry=None
I0417 01:35:36.616505 133638603818816 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7976f445d850>` 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`.
I0417 01:35:36.616557 133638603818816 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7976f445d100>` 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`.
I0417 01:35:36.616595 133638603818816 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 0x7976f445d850>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7976f445d850>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7976f445d100>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7976f445d100>}).
I0417 01:35:36.616929 133638603818816 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0417 01:35:36.617003 133638603818816 async_checkpointer.py:192] [process=1][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7976f42ad800> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0417 01:35:38.537812 133638603818816 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_003418/nnx_xpk_feat_nnx_post_train_fixes_20260417_003418_13_scan_layers_false/checkpoints
I0417 01:35:38.540005 133638603818816 checkpoint_manager.py:929] [process=1][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_20260417_003418/nnx_xpk_feat_nnx_post_train_fixes_20260417_003418_13_scan_layers_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7976f447ef90>
I0417 01:35:38.540122 133638603818816 checkpointing.py:302] Checkpoint manager created!
I0417 01:35:40.663367 133638603818816 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0417 01:35:40.663510 133638603818816 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0417 01:35:48.575947 133638603818816 maxtext_utils.py:1805]  decoder/decoder_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.576069 133638603818816 maxtext_utils.py:1805]  decoder/layers/0/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.576120 133638603818816 maxtext_utils.py:1805]  decoder/layers/0/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.576162 133638603818816 maxtext_utils.py:1805]  decoder/layers/0/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.576200 133638603818816 maxtext_utils.py:1805]  decoder/layers/0/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.576233 133638603818816 maxtext_utils.py:1805]  decoder/layers/0/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.576267 133638603818816 maxtext_utils.py:1805]  decoder/layers/0/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.576300 133638603818816 maxtext_utils.py:1805]  decoder/layers/0/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.576333 133638603818816 maxtext_utils.py:1805]  decoder/layers/0/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.576364 133638603818816 maxtext_utils.py:1805]  decoder/layers/0/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.576397 133638603818816 maxtext_utils.py:1805]  decoder/layers/1/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.576429 133638603818816 maxtext_utils.py:1805]  decoder/layers/1/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.576471 133638603818816 maxtext_utils.py:1805]  decoder/layers/1/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.576503 133638603818816 maxtext_utils.py:1805]  decoder/layers/1/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.576533 133638603818816 maxtext_utils.py:1805]  decoder/layers/1/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.576564 133638603818816 maxtext_utils.py:1805]  decoder/layers/1/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.576596 133638603818816 maxtext_utils.py:1805]  decoder/layers/1/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.576627 133638603818816 maxtext_utils.py:1805]  decoder/layers/1/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.576657 133638603818816 maxtext_utils.py:1805]  decoder/layers/1/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.576687 133638603818816 maxtext_utils.py:1805]  decoder/layers/2/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.576717 133638603818816 maxtext_utils.py:1805]  decoder/layers/2/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.576746 133638603818816 maxtext_utils.py:1805]  decoder/layers/2/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.576775 133638603818816 maxtext_utils.py:1805]  decoder/layers/2/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.576803 133638603818816 maxtext_utils.py:1805]  decoder/layers/2/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.576833 133638603818816 maxtext_utils.py:1805]  decoder/layers/2/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.576885 133638603818816 maxtext_utils.py:1805]  decoder/layers/2/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.576918 133638603818816 maxtext_utils.py:1805]  decoder/layers/2/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.576948 133638603818816 maxtext_utils.py:1805]  decoder/layers/2/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.576978 133638603818816 maxtext_utils.py:1805]  decoder/layers/3/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.577012 133638603818816 maxtext_utils.py:1805]  decoder/layers/3/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.577047 133638603818816 maxtext_utils.py:1805]  decoder/layers/3/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.577092 133638603818816 maxtext_utils.py:1805]  decoder/layers/3/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.577135 133638603818816 maxtext_utils.py:1805]  decoder/layers/3/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.577168 133638603818816 maxtext_utils.py:1805]  decoder/layers/3/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.577199 133638603818816 maxtext_utils.py:1805]  decoder/layers/3/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.577228 133638603818816 maxtext_utils.py:1805]  decoder/layers/3/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.577261 133638603818816 maxtext_utils.py:1805]  decoder/layers/3/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.577291 133638603818816 maxtext_utils.py:1805]  decoder/layers/4/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.577319 133638603818816 maxtext_utils.py:1805]  decoder/layers/4/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.577347 133638603818816 maxtext_utils.py:1805]  decoder/layers/4/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.577374 133638603818816 maxtext_utils.py:1805]  decoder/layers/4/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.577402 133638603818816 maxtext_utils.py:1805]  decoder/layers/4/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.577440 133638603818816 maxtext_utils.py:1805]  decoder/layers/4/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.577489 133638603818816 maxtext_utils.py:1805]  decoder/layers/4/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.577521 133638603818816 maxtext_utils.py:1805]  decoder/layers/4/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.577550 133638603818816 maxtext_utils.py:1805]  decoder/layers/4/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.577580 133638603818816 maxtext_utils.py:1805]  decoder/layers/5/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.577609 133638603818816 maxtext_utils.py:1805]  decoder/layers/5/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.577637 133638603818816 maxtext_utils.py:1805]  decoder/layers/5/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.577664 133638603818816 maxtext_utils.py:1805]  decoder/layers/5/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.577691 133638603818816 maxtext_utils.py:1805]  decoder/layers/5/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.577719 133638603818816 maxtext_utils.py:1805]  decoder/layers/5/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.577748 133638603818816 maxtext_utils.py:1805]  decoder/layers/5/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.577776 133638603818816 maxtext_utils.py:1805]  decoder/layers/5/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.577804 133638603818816 maxtext_utils.py:1805]  decoder/layers/5/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.577832 133638603818816 maxtext_utils.py:1805]  decoder/layers/6/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.577860 133638603818816 maxtext_utils.py:1805]  decoder/layers/6/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.577890 133638603818816 maxtext_utils.py:1805]  decoder/layers/6/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.577917 133638603818816 maxtext_utils.py:1805]  decoder/layers/6/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.577944 133638603818816 maxtext_utils.py:1805]  decoder/layers/6/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.577973 133638603818816 maxtext_utils.py:1805]  decoder/layers/6/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578007 133638603818816 maxtext_utils.py:1805]  decoder/layers/6/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.578037 133638603818816 maxtext_utils.py:1805]  decoder/layers/6/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578065 133638603818816 maxtext_utils.py:1805]  decoder/layers/6/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578093 133638603818816 maxtext_utils.py:1805]  decoder/layers/7/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.578123 133638603818816 maxtext_utils.py:1805]  decoder/layers/7/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.578151 133638603818816 maxtext_utils.py:1805]  decoder/layers/7/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.578177 133638603818816 maxtext_utils.py:1805]  decoder/layers/7/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.578204 133638603818816 maxtext_utils.py:1805]  decoder/layers/7/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.578232 133638603818816 maxtext_utils.py:1805]  decoder/layers/7/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578261 133638603818816 maxtext_utils.py:1805]  decoder/layers/7/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.578289 133638603818816 maxtext_utils.py:1805]  decoder/layers/7/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578317 133638603818816 maxtext_utils.py:1805]  decoder/layers/7/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578345 133638603818816 maxtext_utils.py:1805]  decoder/layers/8/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.578373 133638603818816 maxtext_utils.py:1805]  decoder/layers/8/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.578401 133638603818816 maxtext_utils.py:1805]  decoder/layers/8/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.578428 133638603818816 maxtext_utils.py:1805]  decoder/layers/8/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.578466 133638603818816 maxtext_utils.py:1805]  decoder/layers/8/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.578498 133638603818816 maxtext_utils.py:1805]  decoder/layers/8/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578527 133638603818816 maxtext_utils.py:1805]  decoder/layers/8/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.578557 133638603818816 maxtext_utils.py:1805]  decoder/layers/8/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578585 133638603818816 maxtext_utils.py:1805]  decoder/layers/8/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578613 133638603818816 maxtext_utils.py:1805]  decoder/layers/9/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.578643 133638603818816 maxtext_utils.py:1805]  decoder/layers/9/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.578671 133638603818816 maxtext_utils.py:1805]  decoder/layers/9/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.578698 133638603818816 maxtext_utils.py:1805]  decoder/layers/9/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.578725 133638603818816 maxtext_utils.py:1805]  decoder/layers/9/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.578753 133638603818816 maxtext_utils.py:1805]  decoder/layers/9/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578781 133638603818816 maxtext_utils.py:1805]  decoder/layers/9/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.578809 133638603818816 maxtext_utils.py:1805]  decoder/layers/9/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578837 133638603818816 maxtext_utils.py:1805]  decoder/layers/9/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.578866 133638603818816 maxtext_utils.py:1805]  decoder/layers/10/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.578893 133638603818816 maxtext_utils.py:1805]  decoder/layers/10/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.578921 133638603818816 maxtext_utils.py:1805]  decoder/layers/10/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.578948 133638603818816 maxtext_utils.py:1805]  decoder/layers/10/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.578975 133638603818816 maxtext_utils.py:1805]  decoder/layers/10/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.579009 133638603818816 maxtext_utils.py:1805]  decoder/layers/10/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579039 133638603818816 maxtext_utils.py:1805]  decoder/layers/10/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.579067 133638603818816 maxtext_utils.py:1805]  decoder/layers/10/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579095 133638603818816 maxtext_utils.py:1805]  decoder/layers/10/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579123 133638603818816 maxtext_utils.py:1805]  decoder/layers/11/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.579151 133638603818816 maxtext_utils.py:1805]  decoder/layers/11/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.579179 133638603818816 maxtext_utils.py:1805]  decoder/layers/11/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.579206 133638603818816 maxtext_utils.py:1805]  decoder/layers/11/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.579232 133638603818816 maxtext_utils.py:1805]  decoder/layers/11/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.579260 133638603818816 maxtext_utils.py:1805]  decoder/layers/11/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579288 133638603818816 maxtext_utils.py:1805]  decoder/layers/11/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.579316 133638603818816 maxtext_utils.py:1805]  decoder/layers/11/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579344 133638603818816 maxtext_utils.py:1805]  decoder/layers/11/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579371 133638603818816 maxtext_utils.py:1805]  decoder/layers/12/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.579400 133638603818816 maxtext_utils.py:1805]  decoder/layers/12/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.579427 133638603818816 maxtext_utils.py:1805]  decoder/layers/12/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.579465 133638603818816 maxtext_utils.py:1805]  decoder/layers/12/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.579494 133638603818816 maxtext_utils.py:1805]  decoder/layers/12/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.579523 133638603818816 maxtext_utils.py:1805]  decoder/layers/12/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579552 133638603818816 maxtext_utils.py:1805]  decoder/layers/12/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.579580 133638603818816 maxtext_utils.py:1805]  decoder/layers/12/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579608 133638603818816 maxtext_utils.py:1805]  decoder/layers/12/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579641 133638603818816 maxtext_utils.py:1805]  decoder/layers/13/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.579669 133638603818816 maxtext_utils.py:1805]  decoder/layers/13/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.579698 133638603818816 maxtext_utils.py:1805]  decoder/layers/13/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.579724 133638603818816 maxtext_utils.py:1805]  decoder/layers/13/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.579751 133638603818816 maxtext_utils.py:1805]  decoder/layers/13/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.579779 133638603818816 maxtext_utils.py:1805]  decoder/layers/13/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579807 133638603818816 maxtext_utils.py:1805]  decoder/layers/13/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.579835 133638603818816 maxtext_utils.py:1805]  decoder/layers/13/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579863 133638603818816 maxtext_utils.py:1805]  decoder/layers/13/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.579891 133638603818816 maxtext_utils.py:1805]  decoder/layers/14/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.579920 133638603818816 maxtext_utils.py:1805]  decoder/layers/14/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.579949 133638603818816 maxtext_utils.py:1805]  decoder/layers/14/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.579975 133638603818816 maxtext_utils.py:1805]  decoder/layers/14/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.580007 133638603818816 maxtext_utils.py:1805]  decoder/layers/14/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.580036 133638603818816 maxtext_utils.py:1805]  decoder/layers/14/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.580065 133638603818816 maxtext_utils.py:1805]  decoder/layers/14/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.580093 133638603818816 maxtext_utils.py:1805]  decoder/layers/14/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.580121 133638603818816 maxtext_utils.py:1805]  decoder/layers/14/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.580149 133638603818816 maxtext_utils.py:1805]  decoder/layers/15/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.580177 133638603818816 maxtext_utils.py:1805]  decoder/layers/15/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 01:35:48.580205 133638603818816 maxtext_utils.py:1805]  decoder/layers/15/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 01:35:48.580231 133638603818816 maxtext_utils.py:1805]  decoder/layers/15/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.580258 133638603818816 maxtext_utils.py:1805]  decoder/layers/15/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 01:35:48.580286 133638603818816 maxtext_utils.py:1805]  decoder/layers/15/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.580315 133638603818816 maxtext_utils.py:1805]  decoder/layers/15/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 01:35:48.580343 133638603818816 maxtext_utils.py:1805]  decoder/layers/15/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.580371 133638603818816 maxtext_utils.py:1805]  decoder/layers/15/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 01:35:48.580399 133638603818816 maxtext_utils.py:1805]  decoder/logits_dense/kernel/value
    Shape:     float32[2048,32000]
    Physical:  ('fsdp', None)
I0417 01:35:48.580427 133638603818816 maxtext_utils.py:1805]  token_embedder/embedding/value
    Shape:     float32[32000,2048]
    Physical:  (None, 'fsdp')

I0417 01:35:49.072257 133638603818816 nnx_decoders.py:1023] nnx_decoders/y_in Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0417 01:35:49.072350 133638603818816 nnx_decoders.py:1023] nnx_decoders/y_in Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0417 01:35:49.077810 133638603818816 nnx_decoders.py:1023] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0417 01:35:49.077866 133638603818816 nnx_decoders.py:1023] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0417 01:35:49.093641 133638603818816 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0417 01:35:49.093699 133638603818816 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0417 01:35:49.109018 133638603818816 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0417 01:35:49.109077 133638603818816 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0417 01:35:49.131443 133638603818816 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0417 01:35:49.131521 133638603818816 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0417 01:35:49.147243 133638603818816 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0417 01:35:49.147301 133638603818816 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0417 01:35:49.162825 133638603818816 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0417 01:35:49.162882 133638603818816 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0417 01:35:49.188976 133638603818816 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0417 01:35:49.189045 133638603818816 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0417 01:35:49.208261 133638603818816 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0417 01:35:49.208324 133638603818816 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0417 01:36:48.417353 133638603818816 max_utils.py:791] Total memory size: 1.8 GB, Output size: 0.4 GB, Temp size: 1.4 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0417 01:36:48.428271 133638603818816 metric_logger.py:289] number parameters: 1.104 billion
I0417 01:37:48.600907 133638603818816 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0417 01:37:48.821080 133638603818816 checkpointing.py:798] Waited 0.22018957138061523 seconds for step 0 to finish before starting checkpointing.
I0417 01:37:48.824821 133638603818816 checkpoint_manager.py:2009] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0417 01:37:48.826835 133638603818816 checkpoint_manager.py:1512] [process=1] Saving checkpoint at step 0
I0417 01:37:48.828428 133638603818816 event_tracking.py:70] [process=1] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_003418/nnx_xpk_feat_nnx_post_train_fixes_20260417_003418_13_scan_layers_false/checkpoints/0.
I0417 01:37:49.648195 133638603818816 signaling_client.py:364] Using JaxDistributedSignalingClient
I0417 01:37:49.649536 133638603818816 jax_array_handlers.py:360] Scheduling D2H of 648 prioritized jax.Array.
I0417 01:37:49.649651 133638603818816 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
I0417 01:37:49.951951 133638603818816 base_pytree_checkpoint_handler.py:154] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.308252s
I0417 01:37:49.952115 133638603818816 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/blocking_gbytes_per_sec: 4.234 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.36431288719177246 s) (per-host)
I0417 01:37:49.952167 133638603818816 base_pytree_checkpoint_handler.py:768] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.364375s (batch_requests_ready=0.024146s, total_serialization_initiated=0.340160s, others=0.000069s)
I0417 01:37:49.952581 133638603818816 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.368886s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.368869)
I0417 01:37:49.953423 133638603818816 event_tracking.py:125] [process=1] [async] Finished blocking save in 1.13 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_003418/nnx_xpk_feat_nnx_post_train_fixes_20260417_003418_13_scan_layers_false/checkpoints/0.
I0417 01:37:49.953754 133514292885248 async_checkpointer.py:76] [process=1][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-17 01:57:49.953718
I0417 01:37:49.998034 133638603818816 checkpoint_manager.py:1560] [process=1][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0417 01:37:49.998385 133514276099840 async_checkpointer.py:280] [process=1][thread=save_finalize] Waiting for background save thread=async_save.
I0417 01:37:49.998560 133638603818816 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_003418/nnx_xpk_feat_nnx_post_train_fixes_20260417_003418_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776389868.82479, 'wait_for_prev_duration_secs': 0.0001220703125, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776389868.8268726, 'checkpointer_blocking_duration_secs': 1.1270332336425781, 'get_old_steps_start_time': 1776389869.953933, 'get_old_steps_duration_secs': 2.9802322387695312e-05, 'checkpoint_manager_blocking_start_time': 1776389868.8229527, 'checkpoint_manager_blocking_duration_secs': 1.1755692958831787}
I0417 01:37:49.998807 133638603818816 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0417 01:37:49.999018 133638603818816 max_utils.py:750] 
Memstats: After params initialized:
I0417 01:37:49.999073 133638603818816 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_2(process=1,(2,0,0,0))
I0417 01:37:49.999107 133638603818816 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_3(process=1,(3,0,0,0))
I0417 01:37:49.999135 133638603818816 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_6(process=1,(2,1,0,0))
I0417 01:37:49.999160 133638603818816 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_7(process=1,(3,1,0,0))
I0417 01:37:50.371045 133638603818816 metric_logger.py:185] completed step: 0, seconds: 60.166, TFLOP/s/device: 0.226, Tokens/s/device: 34.039, total_weights: 65536, loss: 10.874
I0417 01:37:50.497194 133638603818816 metric_logger.py:185] completed step: 1, seconds: 1.760, TFLOP/s/device: 7.722, Tokens/s/device: 1163.919, total_weights: 65536, loss: 10.874
I0417 01:37:51.040817 133638603818816 metric_logger.py:185] completed step: 2, seconds: 0.053, TFLOP/s/device: 254.488, Tokens/s/device: 38359.243, total_weights: 65536, loss: 10.265
I0417 01:37:51.176099 133638603818816 metric_logger.py:185] completed step: 3, seconds: 0.428, TFLOP/s/device: 31.774, Tokens/s/device: 4789.310, total_weights: 65536, loss: 9.736
I0417 01:37:51.451220 133638603818816 metric_logger.py:185] completed step: 4, seconds: 0.217, TFLOP/s/device: 62.541, Tokens/s/device: 9426.928, total_weights: 65536, loss: 9.279
I0417 01:37:51.473398 133638603818816 metric_logger.py:185] completed step: 5, seconds: 0.135, TFLOP/s/device: 100.754, Tokens/s/device: 15186.682, total_weights: 65536, loss: 8.894
I0417 01:37:53.006283    2723 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0417 01:37:56.074680 133516968097536 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 648 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_003418/nnx_xpk_feat_nnx_post_train_fixes_20260417_003418_13_scan_layers_false/checkpoints/0/items/array_metadatas/process_1
I0417 01:38:14.925067 133638603818816 metric_logger.py:185] completed step: 6, seconds: 0.276, TFLOP/s/device: 49.288, Tokens/s/device: 7429.200, total_weights: 65536, loss: 8.595
I0417 01:38:15.060928 133638603818816 metric_logger.py:185] completed step: 7, seconds: 23.328, TFLOP/s/device: 0.582, Tokens/s/device: 87.791, total_weights: 65536, loss: 8.387
I0417 01:38:15.196075 133638603818816 metric_logger.py:185] completed step: 8, seconds: 0.146, TFLOP/s/device: 93.318, Tokens/s/device: 14065.934, total_weights: 65536, loss: 8.258
I0417 01:38:15.338885 133638603818816 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0417 01:38:15.343231 133638603818816 checkpointing.py:798] Waited 0.004355430603027344 seconds for step 9 to finish before starting checkpointing.
I0417 01:38:15.346795 133638603818816 checkpoint_manager.py:2020] [process=1][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0417 01:38:22.400149 133514292885248 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/gbytes_per_sec: 48.140 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 32.8123140335083 s) (per-host)
I0417 01:38:22.400276 133514292885248 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 32.446414s.
I0417 01:38:33.384248 133514292885248 async_checkpointer.py:160] [process=1][thread=async_save] Background save thread done. Time taken: 43.430362s.
I0417 01:38:33.384563 133514276099840 async_checkpointer.py:288] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save.
I0417 01:38:33.384681 133514276099840 async_checkpointer.py:298] [process=1][thread=save_finalize] No errors found in background save thread=async_save.
I0417 01:38:33.384733 133514276099840 checkpoint_manager.py:2137] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0417 01:38:33.386153 133514276099840 checkpoint_manager.py:2146] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0417 01:38:33.386387 133638603818816 checkpoint_manager.py:2032] [process=1][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0417 01:38:33.386531 133638603818816 checkpoint_manager.py:1452] Waiting for previous save to complete took 18.039736 seconds. If this number is high, consider checkpointing less frequently.
I0417 01:38:33.389111 133638603818816 checkpoint_manager.py:1512] [process=1] Saving checkpoint at step 9
I0417 01:38:33.391210 133638603818816 event_tracking.py:70] [process=1] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_003418/nnx_xpk_feat_nnx_post_train_fixes_20260417_003418_13_scan_layers_false/checkpoints/9.
I0417 01:38:34.147694 133638603818816 jax_array_handlers.py:360] Scheduling D2H of 648 prioritized jax.Array.
I0417 01:38:34.147833 133638603818816 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
I0417 01:38:34.271376 133638603818816 base_pytree_checkpoint_handler.py:154] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.129325s
I0417 01:38:34.271549 133638603818816 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/blocking_gbytes_per_sec: 8.308 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.1856703758239746 s) (per-host)
I0417 01:38:34.271600 133638603818816 base_pytree_checkpoint_handler.py:768] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.185730s (batch_requests_ready=0.023779s, total_serialization_initiated=0.161884s, others=0.000067s)
I0417 01:38:34.272005 133638603818816 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.190236s (all_items=0.000016s, per_item={'items': '0.00001574'}, temp_paths=0.190220)
I0417 01:38:34.272760 133638603818816 event_tracking.py:125] [process=1] [async] Finished blocking save in 0.88 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_003418/nnx_xpk_feat_nnx_post_train_fixes_20260417_003418_13_scan_layers_false/checkpoints/9.
I0417 01:38:34.273137 133514276099840 async_checkpointer.py:76] [process=1][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-17 01:58:34.273101
I0417 01:38:34.296830 133638603818816 checkpoint_manager.py:1560] [process=1][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0417 01:38:34.297146 133514242529024 async_checkpointer.py:280] [process=1][thread=save_finalize] Waiting for background save thread=async_save.
I0417 01:38:34.297307 133638603818816 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_003418/nnx_xpk_feat_nnx_post_train_fixes_20260417_003418_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776389895.3467643, 'wait_for_prev_duration_secs': 18.03973627090454, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776389913.3891492, 'checkpointer_blocking_duration_secs': 0.8841359615325928, 'get_old_steps_start_time': 1776389914.2733085, 'get_old_steps_duration_secs': 2.9802322387695312e-05, 'checkpoint_manager_blocking_start_time': 1776389895.344286, 'checkpoint_manager_blocking_duration_secs': 18.952987909317017}
I0417 01:38:34.297558 133638603818816 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0417 01:38:34.297605 133638603818816 checkpoint_manager.py:2020] [process=1][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0417 01:38:40.394062 133516968228608 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 648 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_003418/nnx_xpk_feat_nnx_post_train_fixes_20260417_003418_13_scan_layers_false/checkpoints/9/items/array_metadatas/process_1
I0417 01:39:17.076742 133514276099840 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/gbytes_per_sec: 36.743 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.990822076797485 s) (per-host)
I0417 01:39:17.076875 133514276099840 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.803623s.
I0417 01:39:26.120266 133514276099840 async_checkpointer.py:160] [process=1][thread=async_save] Background save thread done. Time taken: 51.846998s.
I0417 01:39:26.120579 133514242529024 async_checkpointer.py:288] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save.
I0417 01:39:26.120704 133514242529024 async_checkpointer.py:298] [process=1][thread=save_finalize] No errors found in background save thread=async_save.
I0417 01:39:26.120755 133514242529024 checkpoint_manager.py:2137] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0417 01:39:26.122479 133514242529024 checkpoint_manager.py:2146] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0417 01:39:26.122669 133638603818816 checkpoint_manager.py:2032] [process=1][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0417 01:39:26.122978 133638603818816 checkpoint_manager.py:2009] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0417 01:39:26.123674 133638603818816 metric_logger.py:185] completed step: 9, seconds: 0.136, TFLOP/s/device: 99.926, Tokens/s/device: 15061.925, total_weights: 65536, loss: 8.182
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Fri Apr 17 01:39:36 UTC 2026
EXIT_CODE=0