MaxView

XPK Start: Fri Apr 17 05:39:04 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 05:39:28.652331: 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 05:39:28.828612 137532157138752 max_utils.py:273] Attempting to initialize the jax distributed system...
I0417 05:39:37.867041 137532157138752 distributed.py:149] Starting JAX distributed service on [::]:8482
I0417 05:39:37.869249 137532157138752 distributed.py:172] Connecting to JAX distributed service on mt-13-scan-layers-false-04nh8-slice-job-0-0.mt-13-scan-layers-false-04nh8:8482
I0417 05:39:39.204526 137532157138752 max_utils.py:284] Jax distributed system initialized!
I0417 05:39:45.084510 137532157138752 max_utils.py:800] System Information: Jax Version: 0.9.2
I0417 05:39:45.084609 137532157138752 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0417 05:39:45.084648 137532157138752 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 05:39:45.084685 137532157138752 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0417 05:39:45.755793 137532157138752 maxtext_utils.py:1687] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0417 05:39:46.558304 137532157138752 checkpointing.py:688] Setting up checkpoint logger...
I0417 05:39:46.558423 137532157138752 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0417 05:39:46.558472 137532157138752 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0417 05:39:46.558710 137532157138752 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 0x7d151551c350>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0417 05:39:49.484517 137532157138752 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0417 05:39:49.484749 137532157138752 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 0x7d006c43caa0>}, handler_registry=None
I0417 05:39:49.484998 137532157138752 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d006c43caa0>` 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 05:39:49.485046 137532157138752 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d016877f650>` 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 05:39:49.485081 137532157138752 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 0x7d006c43caa0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d006c43caa0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d016877f650>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d016877f650>}).
I0417 05:39:49.485405 137532157138752 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0417 05:39:49.485477 137532157138752 async_checkpointer.py:192] [process=0][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d0180581940> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0417 05:39:51.533456 137532157138752 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints
I0417 05:39:51.974514 137532157138752 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_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d006c4070b0>
I0417 05:39:51.974683 137532157138752 checkpointing.py:302] Checkpoint manager created!
I0417 05:39:54.090848 137532157138752 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0417 05:39:54.090973 137532157138752 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0417 05:40:02.010828 137532157138752 maxtext_utils.py:1805]  decoder/decoder_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.010983 137532157138752 maxtext_utils.py:1805]  decoder/layers/0/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.011051 137532157138752 maxtext_utils.py:1805]  decoder/layers/0/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.011113 137532157138752 maxtext_utils.py:1805]  decoder/layers/0/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.011163 137532157138752 maxtext_utils.py:1805]  decoder/layers/0/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.011206 137532157138752 maxtext_utils.py:1805]  decoder/layers/0/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.011245 137532157138752 maxtext_utils.py:1805]  decoder/layers/0/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.011283 137532157138752 maxtext_utils.py:1805]  decoder/layers/0/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.011319 137532157138752 maxtext_utils.py:1805]  decoder/layers/0/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.011352 137532157138752 maxtext_utils.py:1805]  decoder/layers/0/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.011386 137532157138752 maxtext_utils.py:1805]  decoder/layers/1/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.011418 137532157138752 maxtext_utils.py:1805]  decoder/layers/1/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.011449 137532157138752 maxtext_utils.py:1805]  decoder/layers/1/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.011480 137532157138752 maxtext_utils.py:1805]  decoder/layers/1/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.011509 137532157138752 maxtext_utils.py:1805]  decoder/layers/1/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.011540 137532157138752 maxtext_utils.py:1805]  decoder/layers/1/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.011570 137532157138752 maxtext_utils.py:1805]  decoder/layers/1/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.011600 137532157138752 maxtext_utils.py:1805]  decoder/layers/1/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.011630 137532157138752 maxtext_utils.py:1805]  decoder/layers/1/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.011660 137532157138752 maxtext_utils.py:1805]  decoder/layers/2/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.011691 137532157138752 maxtext_utils.py:1805]  decoder/layers/2/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.011720 137532157138752 maxtext_utils.py:1805]  decoder/layers/2/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.011749 137532157138752 maxtext_utils.py:1805]  decoder/layers/2/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.011778 137532157138752 maxtext_utils.py:1805]  decoder/layers/2/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.011809 137532157138752 maxtext_utils.py:1805]  decoder/layers/2/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.011861 137532157138752 maxtext_utils.py:1805]  decoder/layers/2/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.011914 137532157138752 maxtext_utils.py:1805]  decoder/layers/2/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.011962 137532157138752 maxtext_utils.py:1805]  decoder/layers/2/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.011996 137532157138752 maxtext_utils.py:1805]  decoder/layers/3/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.012038 137532157138752 maxtext_utils.py:1805]  decoder/layers/3/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.012094 137532157138752 maxtext_utils.py:1805]  decoder/layers/3/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.012135 137532157138752 maxtext_utils.py:1805]  decoder/layers/3/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.012167 137532157138752 maxtext_utils.py:1805]  decoder/layers/3/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.012203 137532157138752 maxtext_utils.py:1805]  decoder/layers/3/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.012234 137532157138752 maxtext_utils.py:1805]  decoder/layers/3/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.012264 137532157138752 maxtext_utils.py:1805]  decoder/layers/3/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.012293 137532157138752 maxtext_utils.py:1805]  decoder/layers/3/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.012322 137532157138752 maxtext_utils.py:1805]  decoder/layers/4/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.012351 137532157138752 maxtext_utils.py:1805]  decoder/layers/4/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.012378 137532157138752 maxtext_utils.py:1805]  decoder/layers/4/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.012406 137532157138752 maxtext_utils.py:1805]  decoder/layers/4/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.012433 137532157138752 maxtext_utils.py:1805]  decoder/layers/4/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.012462 137532157138752 maxtext_utils.py:1805]  decoder/layers/4/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.012490 137532157138752 maxtext_utils.py:1805]  decoder/layers/4/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.012519 137532157138752 maxtext_utils.py:1805]  decoder/layers/4/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.012547 137532157138752 maxtext_utils.py:1805]  decoder/layers/4/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.012575 137532157138752 maxtext_utils.py:1805]  decoder/layers/5/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.012604 137532157138752 maxtext_utils.py:1805]  decoder/layers/5/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.012632 137532157138752 maxtext_utils.py:1805]  decoder/layers/5/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.012660 137532157138752 maxtext_utils.py:1805]  decoder/layers/5/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.012687 137532157138752 maxtext_utils.py:1805]  decoder/layers/5/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.012716 137532157138752 maxtext_utils.py:1805]  decoder/layers/5/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.012744 137532157138752 maxtext_utils.py:1805]  decoder/layers/5/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.012773 137532157138752 maxtext_utils.py:1805]  decoder/layers/5/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.012800 137532157138752 maxtext_utils.py:1805]  decoder/layers/5/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.012830 137532157138752 maxtext_utils.py:1805]  decoder/layers/6/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.012906 137532157138752 maxtext_utils.py:1805]  decoder/layers/6/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.012944 137532157138752 maxtext_utils.py:1805]  decoder/layers/6/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.012975 137532157138752 maxtext_utils.py:1805]  decoder/layers/6/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.013005 137532157138752 maxtext_utils.py:1805]  decoder/layers/6/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.013036 137532157138752 maxtext_utils.py:1805]  decoder/layers/6/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013066 137532157138752 maxtext_utils.py:1805]  decoder/layers/6/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.013101 137532157138752 maxtext_utils.py:1805]  decoder/layers/6/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013131 137532157138752 maxtext_utils.py:1805]  decoder/layers/6/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013159 137532157138752 maxtext_utils.py:1805]  decoder/layers/7/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.013188 137532157138752 maxtext_utils.py:1805]  decoder/layers/7/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.013218 137532157138752 maxtext_utils.py:1805]  decoder/layers/7/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.013246 137532157138752 maxtext_utils.py:1805]  decoder/layers/7/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.013275 137532157138752 maxtext_utils.py:1805]  decoder/layers/7/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.013305 137532157138752 maxtext_utils.py:1805]  decoder/layers/7/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013334 137532157138752 maxtext_utils.py:1805]  decoder/layers/7/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.013363 137532157138752 maxtext_utils.py:1805]  decoder/layers/7/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013391 137532157138752 maxtext_utils.py:1805]  decoder/layers/7/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013419 137532157138752 maxtext_utils.py:1805]  decoder/layers/8/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.013448 137532157138752 maxtext_utils.py:1805]  decoder/layers/8/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.013477 137532157138752 maxtext_utils.py:1805]  decoder/layers/8/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.013504 137532157138752 maxtext_utils.py:1805]  decoder/layers/8/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.013531 137532157138752 maxtext_utils.py:1805]  decoder/layers/8/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.013560 137532157138752 maxtext_utils.py:1805]  decoder/layers/8/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013590 137532157138752 maxtext_utils.py:1805]  decoder/layers/8/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.013619 137532157138752 maxtext_utils.py:1805]  decoder/layers/8/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013647 137532157138752 maxtext_utils.py:1805]  decoder/layers/8/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013676 137532157138752 maxtext_utils.py:1805]  decoder/layers/9/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.013705 137532157138752 maxtext_utils.py:1805]  decoder/layers/9/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.013733 137532157138752 maxtext_utils.py:1805]  decoder/layers/9/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.013760 137532157138752 maxtext_utils.py:1805]  decoder/layers/9/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.013786 137532157138752 maxtext_utils.py:1805]  decoder/layers/9/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.013814 137532157138752 maxtext_utils.py:1805]  decoder/layers/9/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013864 137532157138752 maxtext_utils.py:1805]  decoder/layers/9/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.013912 137532157138752 maxtext_utils.py:1805]  decoder/layers/9/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013954 137532157138752 maxtext_utils.py:1805]  decoder/layers/9/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.013996 137532157138752 maxtext_utils.py:1805]  decoder/layers/10/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.014036 137532157138752 maxtext_utils.py:1805]  decoder/layers/10/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.014068 137532157138752 maxtext_utils.py:1805]  decoder/layers/10/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.014100 137532157138752 maxtext_utils.py:1805]  decoder/layers/10/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.014129 137532157138752 maxtext_utils.py:1805]  decoder/layers/10/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.014159 137532157138752 maxtext_utils.py:1805]  decoder/layers/10/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.014188 137532157138752 maxtext_utils.py:1805]  decoder/layers/10/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.014217 137532157138752 maxtext_utils.py:1805]  decoder/layers/10/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.014245 137532157138752 maxtext_utils.py:1805]  decoder/layers/10/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.014273 137532157138752 maxtext_utils.py:1805]  decoder/layers/11/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.014301 137532157138752 maxtext_utils.py:1805]  decoder/layers/11/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.014329 137532157138752 maxtext_utils.py:1805]  decoder/layers/11/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.014355 137532157138752 maxtext_utils.py:1805]  decoder/layers/11/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.014382 137532157138752 maxtext_utils.py:1805]  decoder/layers/11/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.014410 137532157138752 maxtext_utils.py:1805]  decoder/layers/11/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.014439 137532157138752 maxtext_utils.py:1805]  decoder/layers/11/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.014467 137532157138752 maxtext_utils.py:1805]  decoder/layers/11/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.014495 137532157138752 maxtext_utils.py:1805]  decoder/layers/11/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.014523 137532157138752 maxtext_utils.py:1805]  decoder/layers/12/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.014550 137532157138752 maxtext_utils.py:1805]  decoder/layers/12/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.014577 137532157138752 maxtext_utils.py:1805]  decoder/layers/12/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.014604 137532157138752 maxtext_utils.py:1805]  decoder/layers/12/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.014630 137532157138752 maxtext_utils.py:1805]  decoder/layers/12/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.014658 137532157138752 maxtext_utils.py:1805]  decoder/layers/12/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.014686 137532157138752 maxtext_utils.py:1805]  decoder/layers/12/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.014714 137532157138752 maxtext_utils.py:1805]  decoder/layers/12/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.014741 137532157138752 maxtext_utils.py:1805]  decoder/layers/12/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.014768 137532157138752 maxtext_utils.py:1805]  decoder/layers/13/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.014796 137532157138752 maxtext_utils.py:1805]  decoder/layers/13/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.014852 137532157138752 maxtext_utils.py:1805]  decoder/layers/13/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.014903 137532157138752 maxtext_utils.py:1805]  decoder/layers/13/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.014947 137532157138752 maxtext_utils.py:1805]  decoder/layers/13/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.014993 137532157138752 maxtext_utils.py:1805]  decoder/layers/13/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.015038 137532157138752 maxtext_utils.py:1805]  decoder/layers/13/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.015087 137532157138752 maxtext_utils.py:1805]  decoder/layers/13/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.015121 137532157138752 maxtext_utils.py:1805]  decoder/layers/13/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.015151 137532157138752 maxtext_utils.py:1805]  decoder/layers/14/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.015180 137532157138752 maxtext_utils.py:1805]  decoder/layers/14/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.015209 137532157138752 maxtext_utils.py:1805]  decoder/layers/14/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.015237 137532157138752 maxtext_utils.py:1805]  decoder/layers/14/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.015264 137532157138752 maxtext_utils.py:1805]  decoder/layers/14/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.015293 137532157138752 maxtext_utils.py:1805]  decoder/layers/14/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.015323 137532157138752 maxtext_utils.py:1805]  decoder/layers/14/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.015351 137532157138752 maxtext_utils.py:1805]  decoder/layers/14/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.015380 137532157138752 maxtext_utils.py:1805]  decoder/layers/14/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.015409 137532157138752 maxtext_utils.py:1805]  decoder/layers/15/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.015437 137532157138752 maxtext_utils.py:1805]  decoder/layers/15/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0417 05:40:02.015464 137532157138752 maxtext_utils.py:1805]  decoder/layers/15/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0417 05:40:02.015490 137532157138752 maxtext_utils.py:1805]  decoder/layers/15/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.015516 137532157138752 maxtext_utils.py:1805]  decoder/layers/15/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0417 05:40:02.015544 137532157138752 maxtext_utils.py:1805]  decoder/layers/15/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.015572 137532157138752 maxtext_utils.py:1805]  decoder/layers/15/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0417 05:40:02.015600 137532157138752 maxtext_utils.py:1805]  decoder/layers/15/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.015626 137532157138752 maxtext_utils.py:1805]  decoder/layers/15/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0417 05:40:02.015654 137532157138752 maxtext_utils.py:1805]  decoder/logits_dense/kernel/value
    Shape:     float32[2048,32000]
    Physical:  ('fsdp', None)
I0417 05:40:02.015681 137532157138752 maxtext_utils.py:1805]  token_embedder/embedding/value
    Shape:     float32[32000,2048]
    Physical:  (None, 'fsdp')

I0417 05:40:02.508824 137532157138752 nnx_decoders.py:1023] nnx_decoders/y_in Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0417 05:40:02.508929 137532157138752 nnx_decoders.py:1023] nnx_decoders/y_in Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0417 05:40:02.513525 137532157138752 nnx_decoders.py:1023] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0417 05:40:02.513580 137532157138752 nnx_decoders.py:1023] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0417 05:40:02.529254 137532157138752 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0417 05:40:02.529308 137532157138752 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0417 05:40:02.544466 137532157138752 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0417 05:40:02.544524 137532157138752 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0417 05:40:02.567685 137532157138752 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 05:40:02.567754 137532157138752 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0417 05:40:02.583110 137532157138752 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 05:40:02.583168 137532157138752 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0417 05:40:02.598422 137532157138752 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 05:40:02.598482 137532157138752 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0417 05:40:02.624223 137532157138752 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0417 05:40:02.624286 137532157138752 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0417 05:40:02.643275 137532157138752 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0417 05:40:02.643335 137532157138752 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0417 05:41:02.325525 137532157138752 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 05:41:02.342105 137532157138752 metric_logger.py:289] number parameters: 1.104 billion
I0417 05:42:03.707964 137532157138752 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0417 05:42:03.929331 137532157138752 checkpointing.py:798] Waited 0.22135353088378906 seconds for step 0 to finish before starting checkpointing.
I0417 05:42:03.933134 137532157138752 checkpoint_manager.py:2009] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0417 05:42:03.935154 137532157138752 checkpoint_manager.py:1512] [process=0] Saving checkpoint at step 0
I0417 05:42:03.936557 137532157138752 event_tracking.py:70] [process=0] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0.
I0417 05:42:04.683241 137532157138752 signaling_client.py:364] Using JaxDistributedSignalingClient
I0417 05:42:04.759868 137532157138752 jax_array_handlers.py:360] Scheduling D2H of 648 prioritized jax.Array.
I0417 05:42:04.760026 137532157138752 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 05:42:04.931482 137407777752832 atomicity.py:140] Creating tmp directory gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0
I0417 05:42:05.076717 137532157138752 base_pytree_checkpoint_handler.py:154] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.322480s
I0417 05:42:05.077658 137532157138752 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/blocking_gbytes_per_sec: 3.989 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.38674163818359375 s) (per-host)
I0417 05:42:05.077724 137532157138752 base_pytree_checkpoint_handler.py:768] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.386825s (batch_requests_ready=0.024153s, total_serialization_initiated=0.361827s, others=0.000844s)
I0417 05:42:05.077885 137532157138752 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.391206s (all_items=0.000019s, per_item={'items': '0.00001884'}, temp_paths=0.391187)
I0417 05:42:05.078704 137532157138752 event_tracking.py:125] [process=0] [async] Finished blocking save in 1.14 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0.
I0417 05:42:05.078993 137407786145536 async_checkpointer.py:76] [process=0][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-17 06:02:05.078970
I0417 05:42:05.097548 137532157138752 checkpoint_manager.py:1560] [process=0][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0417 05:42:05.097807 137407252907776 async_checkpointer.py:280] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0417 05:42:05.097975 137532157138752 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776404523.933116, 'wait_for_prev_duration_secs': 5.9604644775390625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776404523.9351919, 'checkpointer_blocking_duration_secs': 1.143902063369751, 'get_old_steps_start_time': 1776404525.0791216, 'get_old_steps_duration_secs': 2.9087066650390625e-05, 'checkpoint_manager_blocking_start_time': 1776404523.9311552, 'checkpoint_manager_blocking_duration_secs': 1.1667821407318115}
I0417 05:42:05.098221 137532157138752 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0417 05:42:05.098431 137532157138752 max_utils.py:750] 
Memstats: After params initialized:
I0417 05:42:05.098482 137532157138752 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_0(process=0,(0,0,0,0))
I0417 05:42:05.098517 137532157138752 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_1(process=0,(1,0,0,0))
I0417 05:42:05.098546 137532157138752 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_4(process=0,(0,1,0,0))
I0417 05:42:05.098572 137532157138752 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_5(process=0,(1,1,0,0))
I0417 05:42:05.532094 137532157138752 metric_logger.py:185] completed step: 0, seconds: 60.828, TFLOP/s/device: 0.223, Tokens/s/device: 33.669, total_weights: 65536, loss: 10.874
I0417 05:42:05.534600 137532157138752 metric_logger.py:269] To see full metrics 'tensorboard --logdir=gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/tensorboard/'
I0417 05:42:06.066105 137532157138752 metric_logger.py:185] completed step: 1, seconds: 1.801, TFLOP/s/device: 7.546, Tokens/s/device: 1137.420, total_weights: 65536, loss: 10.874
I0417 05:42:06.163429 137532157138752 metric_logger.py:185] completed step: 2, seconds: 0.542, TFLOP/s/device: 25.063, Tokens/s/device: 3777.727, total_weights: 65536, loss: 10.265
I0417 05:42:06.298878 137532157138752 metric_logger.py:185] completed step: 3, seconds: 0.040, TFLOP/s/device: 342.677, Tokens/s/device: 51651.955, total_weights: 65536, loss: 9.736
I0417 05:42:06.736252 137407760967424 atomicity.py:140] Creating tmp directory gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0/items
I0417 05:42:06.877654 137407244515072 checkpoint.py:188] Wrote Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776404526463355073, 'commit_timestamp_nsecs': None, 'custom_metadata': {}}, json={"item_handlers": null, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1776404526463355073, "commit_timestamp_nsecs": null, "custom_metadata": {}} to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0/_CHECKPOINT_METADATA
I0417 05:42:08.819813    2764 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0417 05:42:11.360945 137410448344832 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 648 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0/items/array_metadatas/process_0
I0417 05:42:29.840663 137532157138752 metric_logger.py:185] completed step: 4, seconds: 0.099, TFLOP/s/device: 137.616, Tokens/s/device: 20743.022, total_weights: 65536, loss: 9.279
I0417 05:42:29.867687 137532157138752 metric_logger.py:185] completed step: 5, seconds: 0.135, TFLOP/s/device: 100.293, Tokens/s/device: 15117.180, total_weights: 65536, loss: 8.894
I0417 05:42:29.986491 137532157138752 metric_logger.py:185] completed step: 6, seconds: 23.538, TFLOP/s/device: 0.577, Tokens/s/device: 87.009, total_weights: 65536, loss: 8.595
I0417 05:42:30.122278 137532157138752 metric_logger.py:185] completed step: 7, seconds: 0.023, TFLOP/s/device: 591.748, Tokens/s/device: 89194.721, total_weights: 65536, loss: 8.387
I0417 05:42:30.257893 137532157138752 metric_logger.py:185] completed step: 8, seconds: 0.122, TFLOP/s/device: 111.070, Tokens/s/device: 16741.737, total_weights: 65536, loss: 8.258
I0417 05:42:30.400879 137532157138752 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0417 05:42:30.405527 137532157138752 checkpointing.py:798] Waited 0.004658222198486328 seconds for step 9 to finish before starting checkpointing.
I0417 05:42:30.408433 137532157138752 checkpoint_manager.py:2020] [process=0][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0417 05:42:41.748454 137407261300480 base_pytree_checkpoint_handler.py:1282] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 32.952796s (commit=30.728187s, array_metadata_write=2.224610s)
I0417 05:42:41.750568 137407786145536 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/gbytes_per_sec: 42.623 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.059611797332764 s) (per-host)
I0417 05:42:41.750694 137407786145536 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.671624s.
I0417 05:42:42.448076 137407786145536 checkpoint.py:228] Read Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776404526463355073, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} from gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0/_CHECKPOINT_METADATA
I0417 05:42:43.192859 137407244515072 checkpoint.py:247] Updated Metadata={'item_handlers': {'items': 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler'}, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776404526463355073, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0/_CHECKPOINT_METADATA
I0417 05:42:43.560573 137407786145536 ocdbt_utils.py:49] Param validation support for Zarr3 will be added later (b/362328389).
I0417 05:42:43.757790 137407786145536 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.003374s.
I0417 05:42:44.678417 137407786145536 base_pytree_checkpoint_handler.py:1406] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 2.088339s. use_zarr3=True, enable_post_merge_validation=True, directory=gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0/items
I0417 05:42:44.680269 137407786145536 atomicity.py:666] Finalizing gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0/items
I0417 05:42:45.231111 137407786145536 atomicity.py:666] Finalizing gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0
I0417 05:42:46.641682 137407786145536 atomicity.py:847] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0`.
I0417 05:42:46.642564 137407786145536 event_tracking.py:138] [process=0] [async] Finished save (blocking + background) in 42.71 seconds @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/0
I0417 05:42:46.644049 137407786145536 async_checkpointer.py:160] [process=0][thread=async_save] Background save thread done. Time taken: 41.564976s.
I0417 05:42:46.644236 137407252907776 async_checkpointer.py:288] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0417 05:42:46.644356 137407252907776 async_checkpointer.py:298] [process=0][thread=save_finalize] No errors found in background save thread=async_save.
I0417 05:42:46.644421 137407252907776 checkpoint_manager.py:2137] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0417 05:42:46.648494 137407252907776 checkpoint_manager.py:2146] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0417 05:42:46.648680 137532157138752 checkpoint_manager.py:2032] [process=0][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0417 05:42:46.648814 137532157138752 checkpoint_manager.py:1452] Waiting for previous save to complete took 16.240381 seconds. If this number is high, consider checkpointing less frequently.
I0417 05:42:46.651333 137532157138752 checkpoint_manager.py:1512] [process=0] Saving checkpoint at step 9
I0417 05:42:46.653533 137532157138752 event_tracking.py:70] [process=0] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/9.
I0417 05:42:47.429554 137532157138752 jax_array_handlers.py:360] Scheduling D2H of 648 prioritized jax.Array.
I0417 05:42:47.429695 137532157138752 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 05:42:47.569671 137532157138752 base_pytree_checkpoint_handler.py:154] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.145764s
I0417 05:42:47.570555 137532157138752 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/blocking_gbytes_per_sec: 7.355 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.20972251892089844 s) (per-host)
I0417 05:42:47.570626 137532157138752 base_pytree_checkpoint_handler.py:768] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.209810s (batch_requests_ready=0.023565s, total_serialization_initiated=0.185440s, others=0.000805s)
I0417 05:42:47.570798 137532157138752 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.213818s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.213800)
I0417 05:42:47.571563 137532157138752 event_tracking.py:125] [process=0] [async] Finished blocking save in 0.92 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/9.
I0417 05:42:47.571967 137407252907776 async_checkpointer.py:76] [process=0][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-17 06:02:47.571920
I0417 05:42:47.575083 137532157138752 checkpoint_manager.py:1560] [process=0][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0417 05:42:47.575395 137407261300480 async_checkpointer.py:280] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0417 05:42:47.575557 137532157138752 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776404550.4083986, 'wait_for_prev_duration_secs': 16.24038076400757, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776404566.6513734, 'checkpointer_blocking_duration_secs': 0.9207420349121094, 'get_old_steps_start_time': 1776404567.5721402, 'get_old_steps_duration_secs': 3.0279159545898438e-05, 'checkpoint_manager_blocking_start_time': 1776404550.4065647, 'checkpoint_manager_blocking_duration_secs': 17.168959856033325}
I0417 05:42:47.575749 137532157138752 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0417 05:42:47.575795 137532157138752 checkpoint_manager.py:2020] [process=0][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0417 05:42:48.039529 137407786145536 atomicity.py:140] Creating tmp directory gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/9
I0417 05:42:49.769886 137407269693184 atomicity.py:140] Creating tmp directory gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/9/items
I0417 05:42:54.372400 137408319330048 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 648 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/9/items/array_metadatas/process_0
I0417 05:43:32.595571 137407814956800 base_pytree_checkpoint_handler.py:1282] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 41.629917s (commit=39.359983s, array_metadata_write=2.269934s)
I0417 05:43:32.597688 137407252907776 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/gbytes_per_sec: 34.918 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 45.23685121536255 s) (per-host)
I0417 05:43:32.597756 137407252907776 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 45.025667s.
I0417 05:43:34.634273 137407252907776 ocdbt_utils.py:49] Param validation support for Zarr3 will be added later (b/362328389).
I0417 05:43:34.725715 137407252907776 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.003167s.
I0417 05:43:35.796730 137407252907776 base_pytree_checkpoint_handler.py:1406] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 2.316100s. use_zarr3=True, enable_post_merge_validation=True, directory=gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/9/items
I0417 05:43:35.798620 137407252907776 atomicity.py:666] Finalizing gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/9/items
I0417 05:43:36.751890 137407252907776 atomicity.py:666] Finalizing gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/9
I0417 05:43:38.227747 137407252907776 atomicity.py:847] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/9`.
I0417 05:43:38.228579 137407252907776 event_tracking.py:138] [process=0] [async] Finished save (blocking + background) in 51.58 seconds @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/nnx_xpk_feat_nnx_post_train_fixes_20260417_023710_13_scan_layers_false/checkpoints/9
I0417 05:43:38.230037 137407252907776 async_checkpointer.py:160] [process=0][thread=async_save] Background save thread done. Time taken: 50.657946s.
I0417 05:43:38.230258 137407261300480 async_checkpointer.py:288] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0417 05:43:38.230375 137407261300480 async_checkpointer.py:298] [process=0][thread=save_finalize] No errors found in background save thread=async_save.
I0417 05:43:38.230440 137407261300480 checkpoint_manager.py:2137] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0417 05:43:38.232289 137407261300480 checkpoint_manager.py:2146] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0417 05:43:38.232476 137532157138752 checkpoint_manager.py:2032] [process=0][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0417 05:43:38.232741 137532157138752 checkpoint_manager.py:2009] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0417 05:43:38.233445 137532157138752 metric_logger.py:185] completed step: 9, seconds: 0.135, TFLOP/s/device: 100.705, Tokens/s/device: 15179.366, 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 05:43:50 UTC 2026
EXIT_CODE=0