MaxView

← Back to run

Log Summary

XPK Start: Mon Apr 20 21:32:49 UTC 2026
2026-04-20 21:32:53.239050: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:467] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
E0000 00:00:1776720773.251929      10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1776720773.255626      10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
W0000 00:00:1776720773.267108      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776720773.267130      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776720773.267132      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776720773.267134      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
2026-04-20 21:33:12.453413: 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)
I0420 21:33:12.977396 136850738403136 max_utils.py:273] Attempting to initialize the jax distributed system...
INFO:2026-04-20 21:33:22,019:jax._src.distributed:140: Starting JAX distributed service on [::]:8482
I0420 21:33:22.019452 136850738403136 distributed.py:140] Starting JAX distributed service on [::]:8482
INFO:2026-04-20 21:33:22,021:jax._src.distributed:157: Connecting to JAX distributed service on mt-13-scan-layers-false-90ady-slice-job-0-0.mt-13-scan-layers-false-90ady:8482
I0420 21:33:22.021845 136850738403136 distributed.py:157] Connecting to JAX distributed service on mt-13-scan-layers-false-90ady-slice-job-0-0.mt-13-scan-layers-false-90ady:8482
I0420 21:33:23.020287 136850738403136 max_utils.py:284] Jax distributed system initialized!
I0420 21:33:29.897709 136850738403136 max_utils.py:800] System Information: Jax Version: 0.8.1
I0420 21:33:29.897813 136850738403136 max_utils.py:801] System Information: Jaxlib Version: 0.8.1
I0420 21:33:29.897853 136850738403136 max_utils.py:802] System Information: Jax Backend: PJRT C API
TFRT TPU v6 lite
Built on Nov 12 2025 14:16:36 (1762985796) cl/831091709
I0420 21:33:29.897888 136850738403136 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0420 21:33:30.507970 136850738403136 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 21:33:31.658732 136850738403136 checkpointing.py:688] Setting up checkpoint logger...
I0420 21:33:31.658860 136850738403136 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0420 21:33:31.658908 136850738403136 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None
I0420 21:33:31.659179 136850738403136 base_pytree_checkpoint_handler.py:415] 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 0x7c7684d790a0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0420 21:33:34.627605 136850738403136 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0420 21:33:34.627846 136850738403136 checkpoint_manager.py:709] [process=0][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c61d8584fe0>}, handler_registry=None
I0420 21:33:34.628098 136850738403136 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c61d8584fe0>` 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`.
I0420 21:33:34.628148 136850738403136 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c61d86518b0>` 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`.
I0420 21:33:34.628184 136850738403136 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 0x7c61d8584fe0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c61d8584fe0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c61d86518b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c61d86518b0>}).
I0420 21:33:34.628510 136850738403136 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33
I0420 21:33:34.628593 136850738403136 async_checkpointer.py:177] [process=0][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7c61d86600e0> timeout: 600 secs and primary_host=0 for async checkpoint writes
I0420 21:33:35.731837 136850738403136 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints
I0420 21:33:35.737056 136850738403136 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_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7c61d8651e50>
I0420 21:33:35.737184 136850738403136 checkpointing.py:302] Checkpoint manager created!
I0420 21:33:38.141231 136850738403136 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0420 21:33:38.141359 136850738403136 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0420 21:33:48.195737 136850738403136 maxtext_utils.py:1749]  decoder/decoder_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.195860 136850738403136 maxtext_utils.py:1749]  decoder/layers_0/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.195918 136850738403136 maxtext_utils.py:1749]  decoder/layers_0/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.195966 136850738403136 maxtext_utils.py:1749]  decoder/layers_0/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.196004 136850738403136 maxtext_utils.py:1749]  decoder/layers_0/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.196038 136850738403136 maxtext_utils.py:1749]  decoder/layers_0/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.196076 136850738403136 maxtext_utils.py:1749]  decoder/layers_0/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.196127 136850738403136 maxtext_utils.py:1749]  decoder/layers_0/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.196163 136850738403136 maxtext_utils.py:1749]  decoder/layers_0/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.196197 136850738403136 maxtext_utils.py:1749]  decoder/layers_0/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.196230 136850738403136 maxtext_utils.py:1749]  decoder/layers_1/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.196264 136850738403136 maxtext_utils.py:1749]  decoder/layers_1/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.196297 136850738403136 maxtext_utils.py:1749]  decoder/layers_1/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.196327 136850738403136 maxtext_utils.py:1749]  decoder/layers_1/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.196356 136850738403136 maxtext_utils.py:1749]  decoder/layers_1/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.196389 136850738403136 maxtext_utils.py:1749]  decoder/layers_1/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.196423 136850738403136 maxtext_utils.py:1749]  decoder/layers_1/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.196456 136850738403136 maxtext_utils.py:1749]  decoder/layers_1/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.196488 136850738403136 maxtext_utils.py:1749]  decoder/layers_1/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.196521 136850738403136 maxtext_utils.py:1749]  decoder/layers_10/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.196552 136850738403136 maxtext_utils.py:1749]  decoder/layers_10/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.196584 136850738403136 maxtext_utils.py:1749]  decoder/layers_10/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.196613 136850738403136 maxtext_utils.py:1749]  decoder/layers_10/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.196642 136850738403136 maxtext_utils.py:1749]  decoder/layers_10/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.196673 136850738403136 maxtext_utils.py:1749]  decoder/layers_10/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.196704 136850738403136 maxtext_utils.py:1749]  decoder/layers_10/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.196735 136850738403136 maxtext_utils.py:1749]  decoder/layers_10/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.196768 136850738403136 maxtext_utils.py:1749]  decoder/layers_10/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.196800 136850738403136 maxtext_utils.py:1749]  decoder/layers_11/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.196833 136850738403136 maxtext_utils.py:1749]  decoder/layers_11/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.196866 136850738403136 maxtext_utils.py:1749]  decoder/layers_11/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.196896 136850738403136 maxtext_utils.py:1749]  decoder/layers_11/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.196929 136850738403136 maxtext_utils.py:1749]  decoder/layers_11/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.196963 136850738403136 maxtext_utils.py:1749]  decoder/layers_11/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.196995 136850738403136 maxtext_utils.py:1749]  decoder/layers_11/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.197026 136850738403136 maxtext_utils.py:1749]  decoder/layers_11/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197056 136850738403136 maxtext_utils.py:1749]  decoder/layers_11/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197099 136850738403136 maxtext_utils.py:1749]  decoder/layers_12/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.197132 136850738403136 maxtext_utils.py:1749]  decoder/layers_12/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.197162 136850738403136 maxtext_utils.py:1749]  decoder/layers_12/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.197190 136850738403136 maxtext_utils.py:1749]  decoder/layers_12/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.197219 136850738403136 maxtext_utils.py:1749]  decoder/layers_12/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.197249 136850738403136 maxtext_utils.py:1749]  decoder/layers_12/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197279 136850738403136 maxtext_utils.py:1749]  decoder/layers_12/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.197309 136850738403136 maxtext_utils.py:1749]  decoder/layers_12/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197339 136850738403136 maxtext_utils.py:1749]  decoder/layers_12/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197369 136850738403136 maxtext_utils.py:1749]  decoder/layers_13/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.197399 136850738403136 maxtext_utils.py:1749]  decoder/layers_13/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.197428 136850738403136 maxtext_utils.py:1749]  decoder/layers_13/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.197455 136850738403136 maxtext_utils.py:1749]  decoder/layers_13/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.197488 136850738403136 maxtext_utils.py:1749]  decoder/layers_13/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.197517 136850738403136 maxtext_utils.py:1749]  decoder/layers_13/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197545 136850738403136 maxtext_utils.py:1749]  decoder/layers_13/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.197576 136850738403136 maxtext_utils.py:1749]  decoder/layers_13/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197605 136850738403136 maxtext_utils.py:1749]  decoder/layers_13/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197635 136850738403136 maxtext_utils.py:1749]  decoder/layers_14/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.197664 136850738403136 maxtext_utils.py:1749]  decoder/layers_14/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.197694 136850738403136 maxtext_utils.py:1749]  decoder/layers_14/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.197722 136850738403136 maxtext_utils.py:1749]  decoder/layers_14/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.197749 136850738403136 maxtext_utils.py:1749]  decoder/layers_14/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.197779 136850738403136 maxtext_utils.py:1749]  decoder/layers_14/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197809 136850738403136 maxtext_utils.py:1749]  decoder/layers_14/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.197839 136850738403136 maxtext_utils.py:1749]  decoder/layers_14/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197869 136850738403136 maxtext_utils.py:1749]  decoder/layers_14/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.197899 136850738403136 maxtext_utils.py:1749]  decoder/layers_15/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.197933 136850738403136 maxtext_utils.py:1749]  decoder/layers_15/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.197963 136850738403136 maxtext_utils.py:1749]  decoder/layers_15/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.197991 136850738403136 maxtext_utils.py:1749]  decoder/layers_15/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.198019 136850738403136 maxtext_utils.py:1749]  decoder/layers_15/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.198048 136850738403136 maxtext_utils.py:1749]  decoder/layers_15/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198101 136850738403136 maxtext_utils.py:1749]  decoder/layers_15/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.198140 136850738403136 maxtext_utils.py:1749]  decoder/layers_15/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198172 136850738403136 maxtext_utils.py:1749]  decoder/layers_15/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198202 136850738403136 maxtext_utils.py:1749]  decoder/layers_2/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.198232 136850738403136 maxtext_utils.py:1749]  decoder/layers_2/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.198262 136850738403136 maxtext_utils.py:1749]  decoder/layers_2/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.198289 136850738403136 maxtext_utils.py:1749]  decoder/layers_2/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.198317 136850738403136 maxtext_utils.py:1749]  decoder/layers_2/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.198348 136850738403136 maxtext_utils.py:1749]  decoder/layers_2/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198380 136850738403136 maxtext_utils.py:1749]  decoder/layers_2/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.198411 136850738403136 maxtext_utils.py:1749]  decoder/layers_2/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198441 136850738403136 maxtext_utils.py:1749]  decoder/layers_2/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198471 136850738403136 maxtext_utils.py:1749]  decoder/layers_3/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.198501 136850738403136 maxtext_utils.py:1749]  decoder/layers_3/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.198531 136850738403136 maxtext_utils.py:1749]  decoder/layers_3/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.198559 136850738403136 maxtext_utils.py:1749]  decoder/layers_3/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.198586 136850738403136 maxtext_utils.py:1749]  decoder/layers_3/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.198615 136850738403136 maxtext_utils.py:1749]  decoder/layers_3/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198645 136850738403136 maxtext_utils.py:1749]  decoder/layers_3/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.198675 136850738403136 maxtext_utils.py:1749]  decoder/layers_3/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198705 136850738403136 maxtext_utils.py:1749]  decoder/layers_3/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198734 136850738403136 maxtext_utils.py:1749]  decoder/layers_4/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.198763 136850738403136 maxtext_utils.py:1749]  decoder/layers_4/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.198792 136850738403136 maxtext_utils.py:1749]  decoder/layers_4/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.198820 136850738403136 maxtext_utils.py:1749]  decoder/layers_4/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.198848 136850738403136 maxtext_utils.py:1749]  decoder/layers_4/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.198880 136850738403136 maxtext_utils.py:1749]  decoder/layers_4/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198917 136850738403136 maxtext_utils.py:1749]  decoder/layers_4/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.198949 136850738403136 maxtext_utils.py:1749]  decoder/layers_4/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.198979 136850738403136 maxtext_utils.py:1749]  decoder/layers_4/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.199009 136850738403136 maxtext_utils.py:1749]  decoder/layers_5/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.199039 136850738403136 maxtext_utils.py:1749]  decoder/layers_5/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.199069 136850738403136 maxtext_utils.py:1749]  decoder/layers_5/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.199119 136850738403136 maxtext_utils.py:1749]  decoder/layers_5/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.199150 136850738403136 maxtext_utils.py:1749]  decoder/layers_5/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.199181 136850738403136 maxtext_utils.py:1749]  decoder/layers_5/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.199211 136850738403136 maxtext_utils.py:1749]  decoder/layers_5/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.199241 136850738403136 maxtext_utils.py:1749]  decoder/layers_5/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.199272 136850738403136 maxtext_utils.py:1749]  decoder/layers_5/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.199301 136850738403136 maxtext_utils.py:1749]  decoder/layers_6/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.199331 136850738403136 maxtext_utils.py:1749]  decoder/layers_6/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.199361 136850738403136 maxtext_utils.py:1749]  decoder/layers_6/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.199388 136850738403136 maxtext_utils.py:1749]  decoder/layers_6/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.199416 136850738403136 maxtext_utils.py:1749]  decoder/layers_6/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.199446 136850738403136 maxtext_utils.py:1749]  decoder/layers_6/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.199476 136850738403136 maxtext_utils.py:1749]  decoder/layers_6/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.199506 136850738403136 maxtext_utils.py:1749]  decoder/layers_6/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.199536 136850738403136 maxtext_utils.py:1749]  decoder/layers_6/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.199576 136850738403136 maxtext_utils.py:1749]  decoder/layers_7/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.199607 136850738403136 maxtext_utils.py:1749]  decoder/layers_7/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.199636 136850738403136 maxtext_utils.py:1749]  decoder/layers_7/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.199664 136850738403136 maxtext_utils.py:1749]  decoder/layers_7/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.199691 136850738403136 maxtext_utils.py:1749]  decoder/layers_7/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.199720 136850738403136 maxtext_utils.py:1749]  decoder/layers_7/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.199750 136850738403136 maxtext_utils.py:1749]  decoder/layers_7/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.199779 136850738403136 maxtext_utils.py:1749]  decoder/layers_7/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.199809 136850738403136 maxtext_utils.py:1749]  decoder/layers_7/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.199837 136850738403136 maxtext_utils.py:1749]  decoder/layers_8/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.199867 136850738403136 maxtext_utils.py:1749]  decoder/layers_8/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.199897 136850738403136 maxtext_utils.py:1749]  decoder/layers_8/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.199928 136850738403136 maxtext_utils.py:1749]  decoder/layers_8/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.199955 136850738403136 maxtext_utils.py:1749]  decoder/layers_8/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.199985 136850738403136 maxtext_utils.py:1749]  decoder/layers_8/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.200015 136850738403136 maxtext_utils.py:1749]  decoder/layers_8/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.200045 136850738403136 maxtext_utils.py:1749]  decoder/layers_8/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.200074 136850738403136 maxtext_utils.py:1749]  decoder/layers_8/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.200116 136850738403136 maxtext_utils.py:1749]  decoder/layers_9/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.200146 136850738403136 maxtext_utils.py:1749]  decoder/layers_9/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0420 21:33:48.200175 136850738403136 maxtext_utils.py:1749]  decoder/layers_9/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:48.200202 136850738403136 maxtext_utils.py:1749]  decoder/layers_9/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.200229 136850738403136 maxtext_utils.py:1749]  decoder/layers_9/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0420 21:33:48.200258 136850738403136 maxtext_utils.py:1749]  decoder/layers_9/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.200288 136850738403136 maxtext_utils.py:1749]  decoder/layers_9/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0420 21:33:48.200317 136850738403136 maxtext_utils.py:1749]  decoder/layers_9/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.200347 136850738403136 maxtext_utils.py:1749]  decoder/layers_9/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0420 21:33:48.200376 136850738403136 maxtext_utils.py:1749]  decoder/logits_dense/kernel/value
    Shape:     float32[2048,32000]
    Physical:  ('fsdp', None)

I0420 21:33:48.200405 136850738403136 maxtext_utils.py:1749]  token_embedder/embedding/value
    Shape:     float32[32000,2048]
    Physical:  (None, 'fsdp')
I0420 21:33:49.073495 136850738403136 nnx_decoders.py:770] nnx_decoders/y_in Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0420 21:33:49.073590 136850738403136 nnx_decoders.py:770] nnx_decoders/y_in Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 21:33:49.078527 136850738403136 nnx_decoders.py:770] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0420 21:33:49.078582 136850738403136 nnx_decoders.py:770] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 21:33:49.094155 136850738403136 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 21:33:49.094212 136850738403136 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 21:33:49.109200 136850738403136 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 21:33:49.109256 136850738403136 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 21:33:49.133232 136850738403136 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 21:33:49.133303 136850738403136 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 21:33:49.148612 136850738403136 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 21:33:49.148676 136850738403136 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 21:33:49.163693 136850738403136 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 21:33:49.163760 136850738403136 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 21:33:49.194477 136850738403136 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0420 21:33:49.194549 136850738403136 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 21:33:49.213866 136850738403136 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0420 21:33:49.213930 136850738403136 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0420 21:34:50.038318 136850738403136 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.
I0420 21:34:50.055647 136850738403136 metric_logger.py:289] number parameters: 1.104 billion
I0420 21:35:54.316525 136850738403136 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0420 21:35:54.588252 136850738403136 checkpointing.py:798] Waited 0.2717113494873047 seconds for step 0 to finish before starting checkpointing.
I0420 21:35:54.591607 136850738403136 checkpoint_manager.py:2013] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 21:35:54.593360 136850738403136 checkpoint_manager.py:1518] [process=0] Saving checkpoint at step 0
I0420 21:35:54.596307 136850738403136 async_checkpointer.py:452] [process=0] Started async saving checkpoint to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0.
I0420 21:35:55.350585 136850738403136 signaling_client.py:364] Using JaxDistributedSignalingClient
I0420 21:35:55.622782 136726475413248 atomicity.py:137] Creating tmp directory gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0
I0420 21:35:55.710978 136850738403136 jax_array_handlers.py:358] Scheduling D2H of 648 prioritized jax.Array.
I0420 21:35:55.711158 136850738403136 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
I0420 21:35:56.030188 136850738403136 base_pytree_checkpoint_handler.py:153] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.327334s
I0420 21:35:56.031094 136850738403136 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.293 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.6726229190826416 s) (per-host)
I0420 21:35:56.031164 136850738403136 base_pytree_checkpoint_handler.py:737] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.672725s (batch_requests_ready=0.302981s, total_serialization_initiated=0.368924s, others=0.000820s)
I0420 21:35:56.031296 136850738403136 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.676916s (all_items=0.000019s, per_item={'items': '0.00001884'}, temp_paths=0.676897)
I0420 21:35:56.032248 136727045854976 async_checkpointer.py:79] [process=0][thread=async_save] Background save thread started.
I0420 21:35:56.032353 136850738403136 async_checkpointer.py:561] Finished blocking save. Time taken: 1.438921s. Continuing background save to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0.
I0420 21:35:56.427642 136850738403136 checkpoint_manager.py:1566] [process=0][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 21:35:56.428067 136725988898560 async_checkpointer.py:265] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0420 21:35:56.428253 136850738403136 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776720954.591589, 'wait_for_prev_duration_secs': 6.175041198730469e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776720954.5934007, 'checkpointer_blocking_duration_secs': 1.4390618801116943, 'get_old_steps_start_time': 1776720956.0324848, 'get_old_steps_duration_secs': 2.4557113647460938e-05, 'checkpoint_manager_blocking_start_time': 1776720954.589671, 'checkpoint_manager_blocking_duration_secs': 1.8385398387908936}
I0420 21:35:56.428499 136850738403136 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0420 21:35:56.428739 136850738403136 max_utils.py:750] 
Memstats: After params initialized:
I0420 21:35:56.428798 136850738403136 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_0(process=0,(0,0,0,0))
I0420 21:35:56.428837 136850738403136 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_1(process=0,(1,0,0,0))
I0420 21:35:56.428865 136850738403136 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_4(process=0,(0,1,0,0))
I0420 21:35:56.428888 136850738403136 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_5(process=0,(1,1,0,0))
I0420 21:35:56.791604 136850738403136 metric_logger.py:185] completed step: 0, seconds: 63.667, TFLOP/s/device: 0.213, Tokens/s/device: 32.167, total_weights: 65536, loss: 10.874
I0420 21:35:56.794446 136850738403136 metric_logger.py:269] To see full metrics 'tensorboard --logdir=gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/tensorboard/'
I0420 21:35:57.306341 136850738403136 metric_logger.py:185] completed step: 1, seconds: 2.448, TFLOP/s/device: 5.551, Tokens/s/device: 836.709, total_weights: 65536, loss: 10.874
I0420 21:35:57.390221 136727062640384 atomicity.py:137] Creating tmp directory gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0/items
I0420 21:35:57.405364 136850738403136 metric_logger.py:185] completed step: 2, seconds: 0.526, TFLOP/s/device: 25.812, Tokens/s/device: 3890.607, total_weights: 65536, loss: 10.265
I0420 21:35:57.540064 136726522713856 checkpoint.py:188] Wrote Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776720957132807504, 'commit_timestamp_nsecs': None, 'custom_metadata': {}}, json={"item_handlers": null, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1776720957132807504, "commit_timestamp_nsecs": null, "custom_metadata": {}} to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0/_CHECKPOINT_METADATA
I0420 21:35:57.541978 136850738403136 metric_logger.py:185] completed step: 3, seconds: 0.039, TFLOP/s/device: 351.652, Tokens/s/device: 53004.814, total_weights: 65536, loss: 9.736
I0420 21:36:19.424246 136850738403136 metric_logger.py:185] completed step: 4, seconds: 0.099, TFLOP/s/device: 136.565, Tokens/s/device: 20584.570, total_weights: 65536, loss: 9.279
I0420 21:36:19.454796    2743 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0420 21:36:19.616767 136850738403136 metric_logger.py:185] completed step: 5, seconds: 0.137, TFLOP/s/device: 99.411, Tokens/s/device: 14984.233, total_weights: 65536, loss: 8.894
I0420 21:36:19.790682 136850738403136 metric_logger.py:185] completed step: 6, seconds: 21.967, TFLOP/s/device: 0.619, Tokens/s/device: 93.229, total_weights: 65536, loss: 8.595
I0420 21:36:19.976954 136850738403136 metric_logger.py:185] completed step: 7, seconds: 0.136, TFLOP/s/device: 99.828, Tokens/s/device: 15047.096, total_weights: 65536, loss: 8.387
I0420 21:36:20.021143 136850738403136 metric_logger.py:185] completed step: 8, seconds: 0.177, TFLOP/s/device: 76.781, Tokens/s/device: 11573.237, total_weights: 65536, loss: 8.258
I0420 21:36:20.149922 136850738403136 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0420 21:36:20.154326 136850738403136 checkpointing.py:798] Waited 0.004412412643432617 seconds for step 9 to finish before starting checkpointing.
I0420 21:36:20.159245 136850738403136 checkpoint_manager.py:2024] [process=0][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 21:36:21.370651 136729730209536 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_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0/items/array_metadatas/process_0
I0420 21:36:53.357548 136729721816832 base_pytree_checkpoint_handler.py:1232] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 33.924765s (commit=32.270681s, array_metadata_write=1.654084s)
I0420 21:36:53.359634 136727045854976 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/gbytes_per_sec: 27.234 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 58.00112581253052 s) (per-host)
I0420 21:36:53.359755 136727045854976 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 57.327447s.
I0420 21:36:54.050441 136727045854976 checkpoint.py:228] Read Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776720957132807504, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} from gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0/_CHECKPOINT_METADATA
I0420 21:36:54.809069 136726522713856 checkpoint.py:247] Updated Metadata={'item_handlers': {'items': 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler'}, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776720957132807504, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0/_CHECKPOINT_METADATA
I0420 21:36:55.019810 136727045854976 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.002635s.
I0420 21:36:55.697252 136727045854976 ocdbt_utils.py:49] Param validation support for Zarr3 will be added later (b/362328389).
I0420 21:36:57.076926 136727045854976 base_pytree_checkpoint_handler.py:1356] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 2.904073s. use_zarr3=True, enable_post_merge_validation=True, directory=gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0/items
I0420 21:36:57.079410 136727045854976 atomicity.py:608] Finalizing gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0/items
I0420 21:36:57.504239 136727045854976 atomicity.py:608] Finalizing gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0
I0420 21:36:58.798094 136727045854976 atomicity.py:790] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0`.
I0420 21:36:58.798901 136727045854976 async_checkpointer.py:420] Finished async_save (blocking + background). Time taken: 64.205484s. directory=gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/0
I0420 21:36:58.800482 136727045854976 async_checkpointer.py:144] [process=0][thread=async_save] Background save thread done. Time taken: 62.768174s.
I0420 21:36:58.800729 136725988898560 async_checkpointer.py:273] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 21:36:58.800873 136725988898560 async_checkpointer.py:283] [process=0][thread=save_finalize] No errors found in background save thread=async_save.
I0420 21:36:58.800952 136725988898560 checkpoint_manager.py:2133] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0420 21:36:58.802700 136725988898560 checkpoint_manager.py:2142] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0420 21:36:58.802905 136850738403136 checkpoint_manager.py:2036] [process=0][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0420 21:36:58.803036 136850738403136 checkpoint_manager.py:1458] Waiting for previous save to complete took 38.643798 seconds. If this number is high, consider checkpointing less frequently.
I0420 21:36:58.805293 136850738403136 checkpoint_manager.py:1518] [process=0] Saving checkpoint at step 9
I0420 21:36:58.808882 136850738403136 async_checkpointer.py:452] [process=0] Started async saving checkpoint to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/9.
I0420 21:36:59.748158 136727045854976 atomicity.py:137] Creating tmp directory gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/9
I0420 21:36:59.856855 136850738403136 jax_array_handlers.py:358] Scheduling D2H of 648 prioritized jax.Array.
I0420 21:36:59.856988 136850738403136 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
I0420 21:37:00.002866 136850738403136 base_pytree_checkpoint_handler.py:153] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.153963s
I0420 21:37:00.003763 136850738403136 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.145 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.49044013023376465 s) (per-host)
I0420 21:37:00.003828 136850738403136 base_pytree_checkpoint_handler.py:737] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.490520s (batch_requests_ready=0.294717s, total_serialization_initiated=0.195004s, others=0.000799s)
I0420 21:37:00.003962 136850738403136 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.494702s (all_items=0.000018s, per_item={'items': '0.00001788'}, temp_paths=0.494684)
I0420 21:37:00.004937 136727011096320 async_checkpointer.py:79] [process=0][thread=async_save] Background save thread started.
I0420 21:37:00.005111 136850738403136 async_checkpointer.py:561] Finished blocking save. Time taken: 1.199733s. Continuing background save to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/9.
I0420 21:37:00.029528 136850738403136 checkpoint_manager.py:1566] [process=0][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 21:37:00.029795 136725988898560 async_checkpointer.py:265] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0420 21:37:00.029966 136850738403136 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776720980.1592124, 'wait_for_prev_duration_secs': 38.643797874450684, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776721018.8053339, 'checkpointer_blocking_duration_secs': 1.1998884677886963, 'get_old_steps_start_time': 1776721020.0052416, 'get_old_steps_duration_secs': 3.147125244140625e-05, 'checkpoint_manager_blocking_start_time': 1776720980.1553648, 'checkpoint_manager_blocking_duration_secs': 39.87456798553467}
I0420 21:37:00.030175 136850738403136 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0420 21:37:00.030225 136850738403136 checkpoint_manager.py:2024] [process=0][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 21:37:01.559156 136727054247680 atomicity.py:137] Creating tmp directory gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/9/items
I0420 21:37:04.914111 136729721816832 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_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/9/items/array_metadatas/process_0
I0420 21:37:46.667546 136727549114112 base_pytree_checkpoint_handler.py:1232] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 43.944826s (commit=41.679937s, array_metadata_write=2.264889s)
I0420 21:37:46.669560 136727011096320 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/gbytes_per_sec: 33.497 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 47.15620231628418 s) (per-host)
I0420 21:37:46.669694 136727011096320 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 46.664640s.
I0420 21:37:48.321582 136727011096320 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.003656s.
I0420 21:37:48.782610 136727011096320 ocdbt_utils.py:49] Param validation support for Zarr3 will be added later (b/362328389).
I0420 21:37:50.314962 136727011096320 base_pytree_checkpoint_handler.py:1356] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 2.818037s. use_zarr3=True, enable_post_merge_validation=True, directory=gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/9/items
I0420 21:37:50.316739 136727011096320 atomicity.py:608] Finalizing gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/9/items
I0420 21:37:50.760923 136727011096320 atomicity.py:608] Finalizing gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/9
I0420 21:37:52.351622 136727011096320 atomicity.py:790] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/9`.
I0420 21:37:52.352474 136727011096320 async_checkpointer.py:420] Finished async_save (blocking + background). Time taken: 53.547112s. directory=gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/nnx_xpk_feat_nnx_set_defaults_true_20260420_190404_13_scan_layers_false/checkpoints/9
I0420 21:37:52.353874 136727011096320 async_checkpointer.py:144] [process=0][thread=async_save] Background save thread done. Time taken: 52.348819s.
I0420 21:37:52.354115 136725988898560 async_checkpointer.py:273] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 21:37:52.354228 136725988898560 async_checkpointer.py:283] [process=0][thread=save_finalize] No errors found in background save thread=async_save.
I0420 21:37:52.354297 136725988898560 checkpoint_manager.py:2133] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0420 21:37:52.356074 136725988898560 checkpoint_manager.py:2142] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0420 21:37:52.356262 136850738403136 checkpoint_manager.py:2036] [process=0][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0420 21:37:52.356591 136850738403136 checkpoint_manager.py:2013] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 21:37:52.357218 136850738403136 metric_logger.py:185] completed step: 9, seconds: 0.171, TFLOP/s/device: 79.282, Tokens/s/device: 11950.192, 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: Mon Apr 20 21:38:05 UTC 2026
EXIT_CODE=0