MaxView

XPK Start: Sun Apr 19 02:01:42 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-19 02:02:06.633987: 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)
I0419 02:02:06.812478 139181674235712 max_utils.py:273] Attempting to initialize the jax distributed system...
I0419 02:02:15.851423 139181674235712 distributed.py:149] Starting JAX distributed service on [::]:8482
I0419 02:02:15.853702 139181674235712 distributed.py:172] Connecting to JAX distributed service on mt-13-scan-layers-false-c5zwt-slice-job-0-0.mt-13-scan-layers-false-c5zwt:8482
I0419 02:02:17.709814 139181674235712 max_utils.py:284] Jax distributed system initialized!
I0419 02:02:23.637554 139181674235712 max_utils.py:800] System Information: Jax Version: 0.9.2
I0419 02:02:23.637656 139181674235712 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0419 02:02:23.637694 139181674235712 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
I0419 02:02:23.637729 139181674235712 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0419 02:02:24.692476 139181674235712 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0419 02:02:25.501734 139181674235712 checkpointing.py:688] Setting up checkpoint logger...
I0419 02:02:25.501857 139181674235712 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0419 02:02:25.501902 139181674235712 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0419 02:02:25.502145 139181674235712 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 0x7e950ac4c1d0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0419 02:02:29.611223 139181674235712 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0419 02:02:29.611582 139181674235712 checkpoint_manager.py:708] [process=3][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e8fcc7ada00>}, handler_registry=None
I0419 02:02:29.611834 139181674235712 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e8fcc7ada00>` 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`.
I0419 02:02:29.611883 139181674235712 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e80f84a79b0>` 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`.
I0419 02:02:29.611918 139181674235712 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 0x7e8fcc7ada00>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e8fcc7ada00>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e80f84a79b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e80f84a79b0>}).
I0419 02:02:29.612249 139181674235712 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0419 02:02:29.612323 139181674235712 async_checkpointer.py:192] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7e80f84b02c0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0419 02:02:31.144863 139181674235712 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/nnx_xpk_feat_nnx_post_train_fixes_20260418_235007_13_scan_layers_false/checkpoints
I0419 02:02:31.147080 139181674235712 checkpoint_manager.py:929] [process=3][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_20260418_235007/nnx_xpk_feat_nnx_post_train_fixes_20260418_235007_13_scan_layers_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7e8fcc7ac860>
I0419 02:02:31.147190 139181674235712 checkpointing.py:302] Checkpoint manager created!
I0419 02:02:33.256407 139181674235712 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0419 02:02:33.256533 139181674235712 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0419 02:02:41.133269 139181674235712 maxtext_utils.py:1836]  decoder/decoder_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.133392 139181674235712 maxtext_utils.py:1836]  decoder/layers/0/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.133443 139181674235712 maxtext_utils.py:1836]  decoder/layers/0/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.133489 139181674235712 maxtext_utils.py:1836]  decoder/layers/0/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.133526 139181674235712 maxtext_utils.py:1836]  decoder/layers/0/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.133559 139181674235712 maxtext_utils.py:1836]  decoder/layers/0/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.133594 139181674235712 maxtext_utils.py:1836]  decoder/layers/0/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.133629 139181674235712 maxtext_utils.py:1836]  decoder/layers/0/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.133661 139181674235712 maxtext_utils.py:1836]  decoder/layers/0/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.133693 139181674235712 maxtext_utils.py:1836]  decoder/layers/0/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.133727 139181674235712 maxtext_utils.py:1836]  decoder/layers/1/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.133758 139181674235712 maxtext_utils.py:1836]  decoder/layers/1/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.133788 139181674235712 maxtext_utils.py:1836]  decoder/layers/1/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.133818 139181674235712 maxtext_utils.py:1836]  decoder/layers/1/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.133847 139181674235712 maxtext_utils.py:1836]  decoder/layers/1/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.133878 139181674235712 maxtext_utils.py:1836]  decoder/layers/1/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.133908 139181674235712 maxtext_utils.py:1836]  decoder/layers/1/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.133939 139181674235712 maxtext_utils.py:1836]  decoder/layers/1/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.133968 139181674235712 maxtext_utils.py:1836]  decoder/layers/1/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.133997 139181674235712 maxtext_utils.py:1836]  decoder/layers/2/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.134026 139181674235712 maxtext_utils.py:1836]  decoder/layers/2/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.134056 139181674235712 maxtext_utils.py:1836]  decoder/layers/2/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.134084 139181674235712 maxtext_utils.py:1836]  decoder/layers/2/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.134119 139181674235712 maxtext_utils.py:1836]  decoder/layers/2/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.134150 139181674235712 maxtext_utils.py:1836]  decoder/layers/2/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.134180 139181674235712 maxtext_utils.py:1836]  decoder/layers/2/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.134209 139181674235712 maxtext_utils.py:1836]  decoder/layers/2/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.134252 139181674235712 maxtext_utils.py:1836]  decoder/layers/2/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.134288 139181674235712 maxtext_utils.py:1836]  decoder/layers/3/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.134316 139181674235712 maxtext_utils.py:1836]  decoder/layers/3/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.134347 139181674235712 maxtext_utils.py:1836]  decoder/layers/3/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.134376 139181674235712 maxtext_utils.py:1836]  decoder/layers/3/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.134404 139181674235712 maxtext_utils.py:1836]  decoder/layers/3/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.134433 139181674235712 maxtext_utils.py:1836]  decoder/layers/3/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.134462 139181674235712 maxtext_utils.py:1836]  decoder/layers/3/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.134492 139181674235712 maxtext_utils.py:1836]  decoder/layers/3/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.134521 139181674235712 maxtext_utils.py:1836]  decoder/layers/3/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.134550 139181674235712 maxtext_utils.py:1836]  decoder/layers/4/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.134579 139181674235712 maxtext_utils.py:1836]  decoder/layers/4/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.134608 139181674235712 maxtext_utils.py:1836]  decoder/layers/4/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.134635 139181674235712 maxtext_utils.py:1836]  decoder/layers/4/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.134662 139181674235712 maxtext_utils.py:1836]  decoder/layers/4/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.134697 139181674235712 maxtext_utils.py:1836]  decoder/layers/4/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.134730 139181674235712 maxtext_utils.py:1836]  decoder/layers/4/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.134763 139181674235712 maxtext_utils.py:1836]  decoder/layers/4/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.134793 139181674235712 maxtext_utils.py:1836]  decoder/layers/4/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.134821 139181674235712 maxtext_utils.py:1836]  decoder/layers/5/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.134849 139181674235712 maxtext_utils.py:1836]  decoder/layers/5/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.134877 139181674235712 maxtext_utils.py:1836]  decoder/layers/5/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.134904 139181674235712 maxtext_utils.py:1836]  decoder/layers/5/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.134931 139181674235712 maxtext_utils.py:1836]  decoder/layers/5/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.134958 139181674235712 maxtext_utils.py:1836]  decoder/layers/5/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.134986 139181674235712 maxtext_utils.py:1836]  decoder/layers/5/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.135015 139181674235712 maxtext_utils.py:1836]  decoder/layers/5/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135042 139181674235712 maxtext_utils.py:1836]  decoder/layers/5/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135071 139181674235712 maxtext_utils.py:1836]  decoder/layers/6/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.135106 139181674235712 maxtext_utils.py:1836]  decoder/layers/6/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.135137 139181674235712 maxtext_utils.py:1836]  decoder/layers/6/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.135164 139181674235712 maxtext_utils.py:1836]  decoder/layers/6/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.135190 139181674235712 maxtext_utils.py:1836]  decoder/layers/6/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.135218 139181674235712 maxtext_utils.py:1836]  decoder/layers/6/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135276 139181674235712 maxtext_utils.py:1836]  decoder/layers/6/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.135309 139181674235712 maxtext_utils.py:1836]  decoder/layers/6/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135339 139181674235712 maxtext_utils.py:1836]  decoder/layers/6/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135367 139181674235712 maxtext_utils.py:1836]  decoder/layers/7/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.135396 139181674235712 maxtext_utils.py:1836]  decoder/layers/7/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.135424 139181674235712 maxtext_utils.py:1836]  decoder/layers/7/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.135451 139181674235712 maxtext_utils.py:1836]  decoder/layers/7/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.135478 139181674235712 maxtext_utils.py:1836]  decoder/layers/7/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.135506 139181674235712 maxtext_utils.py:1836]  decoder/layers/7/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135535 139181674235712 maxtext_utils.py:1836]  decoder/layers/7/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.135564 139181674235712 maxtext_utils.py:1836]  decoder/layers/7/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135592 139181674235712 maxtext_utils.py:1836]  decoder/layers/7/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135620 139181674235712 maxtext_utils.py:1836]  decoder/layers/8/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.135648 139181674235712 maxtext_utils.py:1836]  decoder/layers/8/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.135676 139181674235712 maxtext_utils.py:1836]  decoder/layers/8/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.135703 139181674235712 maxtext_utils.py:1836]  decoder/layers/8/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.135730 139181674235712 maxtext_utils.py:1836]  decoder/layers/8/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.135758 139181674235712 maxtext_utils.py:1836]  decoder/layers/8/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135786 139181674235712 maxtext_utils.py:1836]  decoder/layers/8/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.135815 139181674235712 maxtext_utils.py:1836]  decoder/layers/8/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135843 139181674235712 maxtext_utils.py:1836]  decoder/layers/8/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.135871 139181674235712 maxtext_utils.py:1836]  decoder/layers/9/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.135900 139181674235712 maxtext_utils.py:1836]  decoder/layers/9/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.135927 139181674235712 maxtext_utils.py:1836]  decoder/layers/9/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.135953 139181674235712 maxtext_utils.py:1836]  decoder/layers/9/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.135981 139181674235712 maxtext_utils.py:1836]  decoder/layers/9/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.136009 139181674235712 maxtext_utils.py:1836]  decoder/layers/9/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136036 139181674235712 maxtext_utils.py:1836]  decoder/layers/9/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.136064 139181674235712 maxtext_utils.py:1836]  decoder/layers/9/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136097 139181674235712 maxtext_utils.py:1836]  decoder/layers/9/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136127 139181674235712 maxtext_utils.py:1836]  decoder/layers/10/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.136155 139181674235712 maxtext_utils.py:1836]  decoder/layers/10/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.136182 139181674235712 maxtext_utils.py:1836]  decoder/layers/10/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.136209 139181674235712 maxtext_utils.py:1836]  decoder/layers/10/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.136246 139181674235712 maxtext_utils.py:1836]  decoder/layers/10/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.136277 139181674235712 maxtext_utils.py:1836]  decoder/layers/10/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136307 139181674235712 maxtext_utils.py:1836]  decoder/layers/10/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.136336 139181674235712 maxtext_utils.py:1836]  decoder/layers/10/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136364 139181674235712 maxtext_utils.py:1836]  decoder/layers/10/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136392 139181674235712 maxtext_utils.py:1836]  decoder/layers/11/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.136421 139181674235712 maxtext_utils.py:1836]  decoder/layers/11/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.136449 139181674235712 maxtext_utils.py:1836]  decoder/layers/11/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.136476 139181674235712 maxtext_utils.py:1836]  decoder/layers/11/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.136503 139181674235712 maxtext_utils.py:1836]  decoder/layers/11/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.136532 139181674235712 maxtext_utils.py:1836]  decoder/layers/11/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136560 139181674235712 maxtext_utils.py:1836]  decoder/layers/11/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.136588 139181674235712 maxtext_utils.py:1836]  decoder/layers/11/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136616 139181674235712 maxtext_utils.py:1836]  decoder/layers/11/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136644 139181674235712 maxtext_utils.py:1836]  decoder/layers/12/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.136671 139181674235712 maxtext_utils.py:1836]  decoder/layers/12/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.136698 139181674235712 maxtext_utils.py:1836]  decoder/layers/12/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.136725 139181674235712 maxtext_utils.py:1836]  decoder/layers/12/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.136752 139181674235712 maxtext_utils.py:1836]  decoder/layers/12/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.136780 139181674235712 maxtext_utils.py:1836]  decoder/layers/12/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136808 139181674235712 maxtext_utils.py:1836]  decoder/layers/12/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.136837 139181674235712 maxtext_utils.py:1836]  decoder/layers/12/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136865 139181674235712 maxtext_utils.py:1836]  decoder/layers/12/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.136893 139181674235712 maxtext_utils.py:1836]  decoder/layers/13/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.136921 139181674235712 maxtext_utils.py:1836]  decoder/layers/13/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.136949 139181674235712 maxtext_utils.py:1836]  decoder/layers/13/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.136976 139181674235712 maxtext_utils.py:1836]  decoder/layers/13/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.137002 139181674235712 maxtext_utils.py:1836]  decoder/layers/13/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.137030 139181674235712 maxtext_utils.py:1836]  decoder/layers/13/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.137058 139181674235712 maxtext_utils.py:1836]  decoder/layers/13/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.137087 139181674235712 maxtext_utils.py:1836]  decoder/layers/13/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.137120 139181674235712 maxtext_utils.py:1836]  decoder/layers/13/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.137147 139181674235712 maxtext_utils.py:1836]  decoder/layers/14/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.137175 139181674235712 maxtext_utils.py:1836]  decoder/layers/14/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.137205 139181674235712 maxtext_utils.py:1836]  decoder/layers/14/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.137243 139181674235712 maxtext_utils.py:1836]  decoder/layers/14/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.137273 139181674235712 maxtext_utils.py:1836]  decoder/layers/14/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.137303 139181674235712 maxtext_utils.py:1836]  decoder/layers/14/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.137331 139181674235712 maxtext_utils.py:1836]  decoder/layers/14/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.137359 139181674235712 maxtext_utils.py:1836]  decoder/layers/14/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.137387 139181674235712 maxtext_utils.py:1836]  decoder/layers/14/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.137415 139181674235712 maxtext_utils.py:1836]  decoder/layers/15/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.137442 139181674235712 maxtext_utils.py:1836]  decoder/layers/15/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0419 02:02:41.137470 139181674235712 maxtext_utils.py:1836]  decoder/layers/15/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0419 02:02:41.137495 139181674235712 maxtext_utils.py:1836]  decoder/layers/15/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.137522 139181674235712 maxtext_utils.py:1836]  decoder/layers/15/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0419 02:02:41.137550 139181674235712 maxtext_utils.py:1836]  decoder/layers/15/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.137578 139181674235712 maxtext_utils.py:1836]  decoder/layers/15/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0419 02:02:41.137605 139181674235712 maxtext_utils.py:1836]  decoder/layers/15/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.137633 139181674235712 maxtext_utils.py:1836]  decoder/layers/15/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0419 02:02:41.137661 139181674235712 maxtext_utils.py:1836]  decoder/logits_dense/kernel/value
    Shape:     float32[2048,32000]
    Physical:  ('fsdp', None)
I0419 02:02:41.137688 139181674235712 maxtext_utils.py:1836]  token_embedder/embedding/value
    Shape:     float32[32000,2048]
    Physical:  (None, 'fsdp')

I0419 02:02:41.632030 139181674235712 nnx_decoders.py:1023] nnx_decoders/y_in Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0419 02:02:41.632130 139181674235712 nnx_decoders.py:1023] nnx_decoders/y_in Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 02:02:41.636719 139181674235712 nnx_decoders.py:1023] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0419 02:02:41.636775 139181674235712 nnx_decoders.py:1023] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 02:02:41.653203 139181674235712 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0419 02:02:41.653274 139181674235712 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 02:02:41.669016 139181674235712 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0419 02:02:41.669075 139181674235712 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 02:02:41.691663 139181674235712 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0419 02:02:41.691732 139181674235712 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 02:02:41.707487 139181674235712 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0419 02:02:41.707545 139181674235712 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 02:02:41.723346 139181674235712 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0419 02:02:41.723404 139181674235712 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 02:02:41.750977 139181674235712 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0419 02:02:41.751050 139181674235712 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 02:02:41.770672 139181674235712 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0419 02:02:41.770736 139181674235712 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0419 02:03:42.271945 139181674235712 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.
I0419 02:03:42.281881 139181674235712 metric_logger.py:301] number parameters: 1.104 billion
I0419 02:04:43.614875 139181674235712 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0419 02:04:43.838442 139181674235712 checkpointing.py:798] Waited 0.22355318069458008 seconds for step 0 to finish before starting checkpointing.
I0419 02:04:43.841624 139181674235712 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0419 02:04:43.843680 139181674235712 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 0
I0419 02:04:43.845143 139181674235712 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/nnx_xpk_feat_nnx_post_train_fixes_20260418_235007_13_scan_layers_false/checkpoints/0.
I0419 02:04:44.651331 139181674235712 signaling_client.py:364] Using JaxDistributedSignalingClient
I0419 02:04:44.652434 139181674235712 jax_array_handlers.py:360] Scheduling D2H of 648 prioritized jax.Array.
I0419 02:04:44.652554 139181674235712 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
I0419 02:04:44.960397 139181674235712 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.313624s
I0419 02:04:44.960567 139181674235712 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 4.162 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.37063050270080566 s) (per-host)
I0419 02:04:44.960621 139181674235712 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.370694s (batch_requests_ready=0.024335s, total_serialization_initiated=0.346288s, others=0.000071s)
I0419 02:04:44.961004 139181674235712 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.375078s (all_items=0.000016s, per_item={'items': '0.00001645'}, temp_paths=0.375061)
I0419 02:04:44.961874 139181674235712 event_tracking.py:125] [process=3] [async] Finished blocking save in 1.12 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/nnx_xpk_feat_nnx_post_train_fixes_20260418_235007_13_scan_layers_false/checkpoints/0.
I0419 02:04:44.962258 139053970945792 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 02:24:44.962202
I0419 02:04:45.014658 139181674235712 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0419 02:04:45.014979 139053954160384 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0419 02:04:45.015132 139181674235712 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/nnx_xpk_feat_nnx_post_train_fixes_20260418_235007_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776564283.8416061, 'wait_for_prev_duration_secs': 6.151199340820312e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776564283.8437197, 'checkpointer_blocking_duration_secs': 1.1187026500701904, 'get_old_steps_start_time': 1776564284.9624474, 'get_old_steps_duration_secs': 3.0994415283203125e-05, 'checkpoint_manager_blocking_start_time': 1776564283.8397295, 'checkpoint_manager_blocking_duration_secs': 1.1753604412078857}
I0419 02:04:45.015379 139181674235712 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0419 02:04:45.015577 139181674235712 max_utils.py:750] 
Memstats: After params initialized:
I0419 02:04:45.015635 139181674235712 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_10(process=3,(2,2,0,0))
I0419 02:04:45.015667 139181674235712 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_11(process=3,(3,2,0,0))
I0419 02:04:45.015694 139181674235712 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_14(process=3,(2,3,0,0))
I0419 02:04:45.015718 139181674235712 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_15(process=3,(3,3,0,0))
I0419 02:04:45.373458 139181674235712 metric_logger.py:196] completed step: 0, seconds: 61.326, TFLOP/s/device: 0.222, Tokens/s/device: 33.395, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52794.566
I0419 02:04:45.528462 139181674235712 metric_logger.py:196] completed step: 1, seconds: 1.756, TFLOP/s/device: 7.738, Tokens/s/device: 1166.348, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52794.566
I0419 02:04:46.043841 139181674235712 metric_logger.py:196] completed step: 2, seconds: 0.045, TFLOP/s/device: 299.765, Tokens/s/device: 45183.780, total_weights: 65536, loss: 10.265, lm_loss: 10.265, perplexity: 28702.133
I0419 02:04:46.179532 139181674235712 metric_logger.py:196] completed step: 3, seconds: 0.427, TFLOP/s/device: 31.793, Tokens/s/device: 4792.179, total_weights: 65536, loss: 9.736, lm_loss: 9.736, perplexity: 16920.369
I0419 02:04:46.457917 139181674235712 metric_logger.py:196] completed step: 4, seconds: 0.218, TFLOP/s/device: 62.426, Tokens/s/device: 9409.559, total_weights: 65536, loss: 9.279, lm_loss: 9.279, perplexity: 10711.665
I0419 02:04:46.479489 139181674235712 metric_logger.py:196] completed step: 5, seconds: 0.135, TFLOP/s/device: 100.402, Tokens/s/device: 15133.713, total_weights: 65536, loss: 8.894, lm_loss: 8.894, perplexity: 7286.382
I0419 02:04:49.192686    2818 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0419 02:04:52.271691 139056142874368 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 648 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/nnx_xpk_feat_nnx_post_train_fixes_20260418_235007_13_scan_layers_false/checkpoints/0/items/array_metadatas/process_3
I0419 02:05:11.176012 139181674235712 metric_logger.py:196] completed step: 6, seconds: 0.279, TFLOP/s/device: 48.666, Tokens/s/device: 7335.427, total_weights: 65536, loss: 8.595, lm_loss: 8.595, perplexity: 5403.828
I0419 02:05:11.313782 139181674235712 metric_logger.py:196] completed step: 7, seconds: 24.572, TFLOP/s/device: 0.553, Tokens/s/device: 83.346, total_weights: 65536, loss: 8.387, lm_loss: 8.387, perplexity: 4389.161
I0419 02:05:11.450058 139181674235712 metric_logger.py:196] completed step: 8, seconds: 0.146, TFLOP/s/device: 93.293, Tokens/s/device: 14062.071, total_weights: 65536, loss: 8.258, lm_loss: 8.258, perplexity: 3856.852
I0419 02:05:11.592388 139181674235712 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0419 02:05:11.596323 139181674235712 checkpointing.py:798] Waited 0.003949165344238281 seconds for step 9 to finish before starting checkpointing.
I0419 02:05:11.600781 139181674235712 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0419 02:05:18.786048 139053970945792 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 46.192 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 34.1960563659668 s) (per-host)
I0419 02:05:18.786177 139053970945792 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 33.823794s.
I0419 02:05:31.518569 139053970945792 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 46.556170s.
I0419 02:05:31.518827 139053954160384 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0419 02:05:31.518964 139053954160384 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0419 02:05:31.519025 139053954160384 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0419 02:05:31.523441 139053954160384 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0419 02:05:31.523634 139181674235712 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0419 02:05:31.523775 139181674235712 checkpoint_manager.py:1452] Waiting for previous save to complete took 19.922994 seconds. If this number is high, consider checkpointing less frequently.
I0419 02:05:31.526607 139181674235712 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 9
I0419 02:05:31.528732 139181674235712 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/nnx_xpk_feat_nnx_post_train_fixes_20260418_235007_13_scan_layers_false/checkpoints/9.
I0419 02:05:32.712308 139181674235712 jax_array_handlers.py:360] Scheduling D2H of 648 prioritized jax.Array.
I0419 02:05:32.712445 139181674235712 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
I0419 02:05:32.899456 139181674235712 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.192856s
I0419 02:05:32.899617 139181674235712 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 6.231 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.24756574630737305 s) (per-host)
I0419 02:05:32.899665 139181674235712 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.247622s (batch_requests_ready=0.023376s, total_serialization_initiated=0.224180s, others=0.000065s)
I0419 02:05:32.900040 139181674235712 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.252191s (all_items=0.000015s, per_item={'items': '0.00001478'}, temp_paths=0.252177)
I0419 02:05:32.900773 139181674235712 event_tracking.py:125] [process=3] [async] Finished blocking save in 1.37 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/nnx_xpk_feat_nnx_post_train_fixes_20260418_235007_13_scan_layers_false/checkpoints/9.
I0419 02:05:32.901086 139053954160384 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 02:25:32.901053
I0419 02:05:32.920103 139181674235712 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0419 02:05:32.920364 139056142874368 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0419 02:05:32.920494 139181674235712 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/nnx_xpk_feat_nnx_post_train_fixes_20260418_235007_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776564311.6007507, 'wait_for_prev_duration_secs': 19.922994375228882, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776564331.5266483, 'checkpointer_blocking_duration_secs': 1.3746037483215332, 'get_old_steps_start_time': 1776564332.9012764, 'get_old_steps_duration_secs': 3.0040740966796875e-05, 'checkpoint_manager_blocking_start_time': 1776564311.5973284, 'checkpoint_manager_blocking_duration_secs': 21.323131322860718}
I0419 02:05:32.920758 139181674235712 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0419 02:05:32.920805 139181674235712 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0419 02:05:40.198013 139053970945792 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 648 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/nnx_xpk_feat_nnx_post_train_fixes_20260418_235007_13_scan_layers_false/checkpoints/9/items/array_metadatas/process_3
I0419 02:06:14.360117 139053954160384 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 37.873 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.708025217056274 s) (per-host)
I0419 02:06:14.360246 139053954160384 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.459028s.
I0419 02:06:25.181849 139053954160384 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 52.280627s.
I0419 02:06:25.182140 139056142874368 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0419 02:06:25.182290 139056142874368 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0419 02:06:25.182341 139056142874368 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0419 02:06:25.183546 139056142874368 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0419 02:06:25.183726 139181674235712 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0419 02:06:25.184020 139181674235712 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0419 02:06:25.184928 139181674235712 metric_logger.py:196] completed step: 9, seconds: 0.137, TFLOP/s/device: 98.921, Tokens/s/device: 14910.486, total_weights: 65536, loss: 8.182, lm_loss: 8.182, perplexity: 3575.032
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Sun Apr 19 02:06:37 UTC 2026
EXIT_CODE=0