MaxView

XPK Start: Wed Apr 22 14:52:23 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
`rope_parameters`'s factor field must be a float >= 1, got 40
`rope_parameters`'s beta_fast field must be a float, got 32
`rope_parameters`'s beta_slow field must be a float, got 1
DeepseekV32Config got `key=rope_scaling` in kwargs but hasn't set it as attribute. For RoPE standardization you need to set `self.rope_parameters` in model's config. 
2026-04-22 14:52:48.290014: 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)
I0422 14:52:48.495763 138678686721856 max_utils.py:273] Attempting to initialize the jax distributed system...
I0422 14:52:57.536643 138678686721856 distributed.py:149] Starting JAX distributed service on [::]:8482
I0422 14:52:57.538949 138678686721856 distributed.py:172] Connecting to JAX distributed service on mt-13-scan-layers-false-ra4oo-slice-job-0-0.mt-13-scan-layers-false-ra4oo:8482
I0422 14:52:58.585638 138678686721856 max_utils.py:284] Jax distributed system initialized!
I0422 14:53:05.076152 138678686721856 max_utils.py:800] System Information: Jax Version: 0.9.2
I0422 14:53:05.076261 138678686721856 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0422 14:53:05.076301 138678686721856 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
I0422 14:53:05.076337 138678686721856 train_utils.py:391] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0422 14:53:05.767247 138678686721856 maxtext_utils.py:1771] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0422 14:53:06.577125 138678686721856 checkpointing.py:688] Setting up checkpoint logger...
I0422 14:53:06.577253 138678686721856 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0422 14:53:06.577299 138678686721856 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0422 14:53:06.577535 138678686721856 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 0x7e1fee3445c0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0422 14:53:09.486691 138678686721856 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0422 14:53:09.486926 138678686721856 checkpoint_manager.py:708] [process=5][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e0bd8739c40>}, handler_registry=None
I0422 14:53:09.487166 138678686721856 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e0bd8739c40>` 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`.
I0422 14:53:09.487214 138678686721856 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e0bd82c4380>` 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`.
I0422 14:53:09.487256 138678686721856 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 0x7e0bd8739c40>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e0bd8739c40>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e0bd82c4380>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e0bd82c4380>}).
I0422 14:53:09.487583 138678686721856 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0422 14:53:09.487686 138678686721856 async_checkpointer.py:192] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7e0bd83e3a60> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0422 14:53:10.623201 138678686721856 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/nnx_xpk_feat_nnx_post_train_fixes_20260422_123906_13_scan_layers_false/checkpoints
I0422 14:53:11.067759 138678686721856 checkpoint_manager.py:929] [process=5][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_20260422_123906/nnx_xpk_feat_nnx_post_train_fixes_20260422_123906_13_scan_layers_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7e0bd86fa630>
I0422 14:53:11.067929 138678686721856 checkpointing.py:302] Checkpoint manager created!
I0422 14:53:13.189698 138678686721856 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0422 14:53:13.189831 138678686721856 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0422 14:53:21.039354 138678686721856 maxtext_utils.py:1889]  decoder/decoder_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.039476 138678686721856 maxtext_utils.py:1889]  decoder/layers/0/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.039527 138678686721856 maxtext_utils.py:1889]  decoder/layers/0/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.039569 138678686721856 maxtext_utils.py:1889]  decoder/layers/0/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.039606 138678686721856 maxtext_utils.py:1889]  decoder/layers/0/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.039641 138678686721856 maxtext_utils.py:1889]  decoder/layers/0/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.039691 138678686721856 maxtext_utils.py:1889]  decoder/layers/0/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.039728 138678686721856 maxtext_utils.py:1889]  decoder/layers/0/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.039762 138678686721856 maxtext_utils.py:1889]  decoder/layers/0/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.039794 138678686721856 maxtext_utils.py:1889]  decoder/layers/0/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.039826 138678686721856 maxtext_utils.py:1889]  decoder/layers/1/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.039859 138678686721856 maxtext_utils.py:1889]  decoder/layers/1/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.039891 138678686721856 maxtext_utils.py:1889]  decoder/layers/1/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.039921 138678686721856 maxtext_utils.py:1889]  decoder/layers/1/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.039954 138678686721856 maxtext_utils.py:1889]  decoder/layers/1/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.039986 138678686721856 maxtext_utils.py:1889]  decoder/layers/1/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040018 138678686721856 maxtext_utils.py:1889]  decoder/layers/1/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.040048 138678686721856 maxtext_utils.py:1889]  decoder/layers/1/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040078 138678686721856 maxtext_utils.py:1889]  decoder/layers/1/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040108 138678686721856 maxtext_utils.py:1889]  decoder/layers/2/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.040138 138678686721856 maxtext_utils.py:1889]  decoder/layers/2/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.040169 138678686721856 maxtext_utils.py:1889]  decoder/layers/2/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.040197 138678686721856 maxtext_utils.py:1889]  decoder/layers/2/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.040225 138678686721856 maxtext_utils.py:1889]  decoder/layers/2/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.040255 138678686721856 maxtext_utils.py:1889]  decoder/layers/2/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040285 138678686721856 maxtext_utils.py:1889]  decoder/layers/2/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.040314 138678686721856 maxtext_utils.py:1889]  decoder/layers/2/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040343 138678686721856 maxtext_utils.py:1889]  decoder/layers/2/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040374 138678686721856 maxtext_utils.py:1889]  decoder/layers/3/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.040404 138678686721856 maxtext_utils.py:1889]  decoder/layers/3/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.040433 138678686721856 maxtext_utils.py:1889]  decoder/layers/3/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.040460 138678686721856 maxtext_utils.py:1889]  decoder/layers/3/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.040488 138678686721856 maxtext_utils.py:1889]  decoder/layers/3/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.040519 138678686721856 maxtext_utils.py:1889]  decoder/layers/3/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040548 138678686721856 maxtext_utils.py:1889]  decoder/layers/3/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.040577 138678686721856 maxtext_utils.py:1889]  decoder/layers/3/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040606 138678686721856 maxtext_utils.py:1889]  decoder/layers/3/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040634 138678686721856 maxtext_utils.py:1889]  decoder/layers/4/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.040674 138678686721856 maxtext_utils.py:1889]  decoder/layers/4/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.040704 138678686721856 maxtext_utils.py:1889]  decoder/layers/4/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.040732 138678686721856 maxtext_utils.py:1889]  decoder/layers/4/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.040759 138678686721856 maxtext_utils.py:1889]  decoder/layers/4/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.040788 138678686721856 maxtext_utils.py:1889]  decoder/layers/4/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040817 138678686721856 maxtext_utils.py:1889]  decoder/layers/4/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.040845 138678686721856 maxtext_utils.py:1889]  decoder/layers/4/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040874 138678686721856 maxtext_utils.py:1889]  decoder/layers/4/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.040901 138678686721856 maxtext_utils.py:1889]  decoder/layers/5/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.040933 138678686721856 maxtext_utils.py:1889]  decoder/layers/5/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.040962 138678686721856 maxtext_utils.py:1889]  decoder/layers/5/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.040990 138678686721856 maxtext_utils.py:1889]  decoder/layers/5/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.041016 138678686721856 maxtext_utils.py:1889]  decoder/layers/5/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.041044 138678686721856 maxtext_utils.py:1889]  decoder/layers/5/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041072 138678686721856 maxtext_utils.py:1889]  decoder/layers/5/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.041100 138678686721856 maxtext_utils.py:1889]  decoder/layers/5/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041128 138678686721856 maxtext_utils.py:1889]  decoder/layers/5/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041156 138678686721856 maxtext_utils.py:1889]  decoder/layers/6/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.041185 138678686721856 maxtext_utils.py:1889]  decoder/layers/6/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.041214 138678686721856 maxtext_utils.py:1889]  decoder/layers/6/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.041243 138678686721856 maxtext_utils.py:1889]  decoder/layers/6/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.041271 138678686721856 maxtext_utils.py:1889]  decoder/layers/6/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.041299 138678686721856 maxtext_utils.py:1889]  decoder/layers/6/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041328 138678686721856 maxtext_utils.py:1889]  decoder/layers/6/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.041356 138678686721856 maxtext_utils.py:1889]  decoder/layers/6/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041384 138678686721856 maxtext_utils.py:1889]  decoder/layers/6/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041412 138678686721856 maxtext_utils.py:1889]  decoder/layers/7/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.041439 138678686721856 maxtext_utils.py:1889]  decoder/layers/7/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.041466 138678686721856 maxtext_utils.py:1889]  decoder/layers/7/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.041492 138678686721856 maxtext_utils.py:1889]  decoder/layers/7/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.041519 138678686721856 maxtext_utils.py:1889]  decoder/layers/7/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.041547 138678686721856 maxtext_utils.py:1889]  decoder/layers/7/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041574 138678686721856 maxtext_utils.py:1889]  decoder/layers/7/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.041602 138678686721856 maxtext_utils.py:1889]  decoder/layers/7/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041630 138678686721856 maxtext_utils.py:1889]  decoder/layers/7/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041667 138678686721856 maxtext_utils.py:1889]  decoder/layers/8/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.041698 138678686721856 maxtext_utils.py:1889]  decoder/layers/8/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.041726 138678686721856 maxtext_utils.py:1889]  decoder/layers/8/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.041753 138678686721856 maxtext_utils.py:1889]  decoder/layers/8/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.041780 138678686721856 maxtext_utils.py:1889]  decoder/layers/8/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.041810 138678686721856 maxtext_utils.py:1889]  decoder/layers/8/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041838 138678686721856 maxtext_utils.py:1889]  decoder/layers/8/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.041866 138678686721856 maxtext_utils.py:1889]  decoder/layers/8/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041893 138678686721856 maxtext_utils.py:1889]  decoder/layers/8/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.041921 138678686721856 maxtext_utils.py:1889]  decoder/layers/9/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.041952 138678686721856 maxtext_utils.py:1889]  decoder/layers/9/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.041980 138678686721856 maxtext_utils.py:1889]  decoder/layers/9/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.042007 138678686721856 maxtext_utils.py:1889]  decoder/layers/9/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.042033 138678686721856 maxtext_utils.py:1889]  decoder/layers/9/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.042060 138678686721856 maxtext_utils.py:1889]  decoder/layers/9/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042088 138678686721856 maxtext_utils.py:1889]  decoder/layers/9/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.042115 138678686721856 maxtext_utils.py:1889]  decoder/layers/9/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042142 138678686721856 maxtext_utils.py:1889]  decoder/layers/9/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042170 138678686721856 maxtext_utils.py:1889]  decoder/layers/10/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.042197 138678686721856 maxtext_utils.py:1889]  decoder/layers/10/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.042225 138678686721856 maxtext_utils.py:1889]  decoder/layers/10/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.042252 138678686721856 maxtext_utils.py:1889]  decoder/layers/10/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.042279 138678686721856 maxtext_utils.py:1889]  decoder/layers/10/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.042308 138678686721856 maxtext_utils.py:1889]  decoder/layers/10/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042336 138678686721856 maxtext_utils.py:1889]  decoder/layers/10/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.042364 138678686721856 maxtext_utils.py:1889]  decoder/layers/10/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042392 138678686721856 maxtext_utils.py:1889]  decoder/layers/10/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042420 138678686721856 maxtext_utils.py:1889]  decoder/layers/11/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.042448 138678686721856 maxtext_utils.py:1889]  decoder/layers/11/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.042475 138678686721856 maxtext_utils.py:1889]  decoder/layers/11/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.042501 138678686721856 maxtext_utils.py:1889]  decoder/layers/11/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.042527 138678686721856 maxtext_utils.py:1889]  decoder/layers/11/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.042556 138678686721856 maxtext_utils.py:1889]  decoder/layers/11/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042583 138678686721856 maxtext_utils.py:1889]  decoder/layers/11/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.042610 138678686721856 maxtext_utils.py:1889]  decoder/layers/11/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042639 138678686721856 maxtext_utils.py:1889]  decoder/layers/11/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042678 138678686721856 maxtext_utils.py:1889]  decoder/layers/12/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.042708 138678686721856 maxtext_utils.py:1889]  decoder/layers/12/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.042736 138678686721856 maxtext_utils.py:1889]  decoder/layers/12/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.042762 138678686721856 maxtext_utils.py:1889]  decoder/layers/12/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.042789 138678686721856 maxtext_utils.py:1889]  decoder/layers/12/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.042817 138678686721856 maxtext_utils.py:1889]  decoder/layers/12/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042845 138678686721856 maxtext_utils.py:1889]  decoder/layers/12/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.042873 138678686721856 maxtext_utils.py:1889]  decoder/layers/12/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042901 138678686721856 maxtext_utils.py:1889]  decoder/layers/12/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.042932 138678686721856 maxtext_utils.py:1889]  decoder/layers/13/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.042960 138678686721856 maxtext_utils.py:1889]  decoder/layers/13/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.042989 138678686721856 maxtext_utils.py:1889]  decoder/layers/13/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.043015 138678686721856 maxtext_utils.py:1889]  decoder/layers/13/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.043041 138678686721856 maxtext_utils.py:1889]  decoder/layers/13/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.043069 138678686721856 maxtext_utils.py:1889]  decoder/layers/13/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.043097 138678686721856 maxtext_utils.py:1889]  decoder/layers/13/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.043124 138678686721856 maxtext_utils.py:1889]  decoder/layers/13/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.043152 138678686721856 maxtext_utils.py:1889]  decoder/layers/13/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.043179 138678686721856 maxtext_utils.py:1889]  decoder/layers/14/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.043207 138678686721856 maxtext_utils.py:1889]  decoder/layers/14/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.043235 138678686721856 maxtext_utils.py:1889]  decoder/layers/14/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.043261 138678686721856 maxtext_utils.py:1889]  decoder/layers/14/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.043287 138678686721856 maxtext_utils.py:1889]  decoder/layers/14/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.043315 138678686721856 maxtext_utils.py:1889]  decoder/layers/14/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.043344 138678686721856 maxtext_utils.py:1889]  decoder/layers/14/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.043371 138678686721856 maxtext_utils.py:1889]  decoder/layers/14/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.043399 138678686721856 maxtext_utils.py:1889]  decoder/layers/14/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.043426 138678686721856 maxtext_utils.py:1889]  decoder/layers/15/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.043454 138678686721856 maxtext_utils.py:1889]  decoder/layers/15/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0422 14:53:21.043482 138678686721856 maxtext_utils.py:1889]  decoder/layers/15/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.043508 138678686721856 maxtext_utils.py:1889]  decoder/layers/15/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.043534 138678686721856 maxtext_utils.py:1889]  decoder/layers/15/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0422 14:53:21.043561 138678686721856 maxtext_utils.py:1889]  decoder/layers/15/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.043590 138678686721856 maxtext_utils.py:1889]  decoder/layers/15/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0422 14:53:21.043617 138678686721856 maxtext_utils.py:1889]  decoder/layers/15/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0422 14:53:21.043645 138678686721856 maxtext_utils.py:1889]  decoder/layers/15/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)

I0422 14:53:21.043684 138678686721856 maxtext_utils.py:1889]  decoder/logits_dense/kernel/value
    Shape:     float32[2048,32000]
    Physical:  ('fsdp', None)
I0422 14:53:21.043713 138678686721856 maxtext_utils.py:1889]  token_embedder/embedding/value
    Shape:     float32[32000,2048]
    Physical:  (None, 'fsdp')
I0422 14:53:21.531760 138678686721856 nnx_decoders.py:1023] nnx_decoders/y_in Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0422 14:53:21.531855 138678686721856 nnx_decoders.py:1023] nnx_decoders/y_in Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 14:53:21.536437 138678686721856 nnx_decoders.py:1023] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0422 14:53:21.536493 138678686721856 nnx_decoders.py:1023] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 14:53:21.553018 138678686721856 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 14:53:21.553083 138678686721856 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 14:53:21.568996 138678686721856 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 14:53:21.569054 138678686721856 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 14:53:21.592194 138678686721856 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0422 14:53:21.592267 138678686721856 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 14:53:21.608227 138678686721856 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0422 14:53:21.608286 138678686721856 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 14:53:21.624290 138678686721856 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0422 14:53:21.624350 138678686721856 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 14:53:21.651358 138678686721856 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0422 14:53:21.651430 138678686721856 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 14:53:21.672235 138678686721856 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0422 14:53:21.672300 138678686721856 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0422 14:54:22.310612 138678686721856 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.
I0422 14:54:22.321821 138678686721856 metric_logger.py:301] number parameters: 1.104 billion
I0422 14:55:22.910902 138678686721856 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0422 14:55:23.136223 138678686721856 checkpointing.py:798] Waited 0.22533535957336426 seconds for step 0 to finish before starting checkpointing.
I0422 14:55:23.139578 138678686721856 checkpoint_manager.py:2009] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 14:55:23.141507 138678686721856 checkpoint_manager.py:1512] [process=5] Saving checkpoint at step 0
I0422 14:55:23.142895 138678686721856 event_tracking.py:70] [process=5] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/nnx_xpk_feat_nnx_post_train_fixes_20260422_123906_13_scan_layers_false/checkpoints/0.
I0422 14:55:23.800494 138678686721856 signaling_client.py:364] Using JaxDistributedSignalingClient
I0422 14:55:23.801628 138678686721856 jax_array_handlers.py:360] Scheduling D2H of 648 prioritized jax.Array.
I0422 14:55:23.801759 138678686721856 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
I0422 14:55:24.125686 138678686721856 base_pytree_checkpoint_handler.py:154] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.329700s
I0422 14:55:24.125856 138678686721856 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/blocking_gbytes_per_sec: 3.985 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3870506286621094 s) (per-host)
I0422 14:55:24.125905 138678686721856 base_pytree_checkpoint_handler.py:768] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.387110s (batch_requests_ready=0.024290s, total_serialization_initiated=0.362753s, others=0.000067s)
I0422 14:55:24.126279 138678686721856 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.391375s (all_items=0.000017s, per_item={'items': '0.00001693'}, temp_paths=0.391358)
I0422 14:55:24.127095 138678686721856 event_tracking.py:125] [process=5] [async] Finished blocking save in 0.99 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/nnx_xpk_feat_nnx_post_train_fixes_20260422_123906_13_scan_layers_false/checkpoints/0.
I0422 14:55:24.127459 138550821975808 async_checkpointer.py:76] [process=5][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 15:15:24.127422
I0422 14:55:24.156265 138678686721856 checkpoint_manager.py:1560] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 14:55:24.156587 138550318188288 async_checkpointer.py:280] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0422 14:55:24.156768 138678686721856 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/nnx_xpk_feat_nnx_post_train_fixes_20260422_123906_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776869723.1395607, 'wait_for_prev_duration_secs': 5.984306335449219e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776869723.1415448, 'checkpointer_blocking_duration_secs': 0.9860692024230957, 'get_old_steps_start_time': 1776869724.1276398, 'get_old_steps_duration_secs': 4.1484832763671875e-05, 'checkpoint_manager_blocking_start_time': 1776869723.1374888, 'checkpoint_manager_blocking_duration_secs': 1.019239902496338}
I0422 14:55:24.157005 138678686721856 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0422 14:55:24.157205 138678686721856 max_utils.py:750] 
Memstats: After params initialized:
I0422 14:55:24.157263 138678686721856 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_18(process=5,(2,4,0,0))
I0422 14:55:24.157296 138678686721856 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_19(process=5,(3,4,0,0))
I0422 14:55:24.157323 138678686721856 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_22(process=5,(2,5,0,0))
I0422 14:55:24.157346 138678686721856 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_23(process=5,(3,5,0,0))
I0422 14:55:24.531672 138678686721856 metric_logger.py:196] completed step: 0, seconds: 60.582, TFLOP/s/device: 0.224, Tokens/s/device: 33.805, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52794.566
I0422 14:55:24.655440 138678686721856 metric_logger.py:196] completed step: 1, seconds: 1.610, TFLOP/s/device: 8.441, Tokens/s/device: 1272.298, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52794.566
I0422 14:55:25.184454 138678686721856 metric_logger.py:196] completed step: 2, seconds: 0.053, TFLOP/s/device: 257.132, Tokens/s/device: 38757.783, total_weights: 65536, loss: 10.265, lm_loss: 10.265, perplexity: 28703.443
I0422 14:55:25.324577 138678686721856 metric_logger.py:196] completed step: 3, seconds: 0.411, TFLOP/s/device: 33.095, Tokens/s/device: 4988.454, total_weights: 65536, loss: 9.736, lm_loss: 9.736, perplexity: 16918.312
I0422 14:55:25.602612 138678686721856 metric_logger.py:196] completed step: 4, seconds: 0.218, TFLOP/s/device: 62.300, Tokens/s/device: 9390.532, total_weights: 65536, loss: 9.279, lm_loss: 9.279, perplexity: 10712.140
I0422 14:55:25.624587 138678686721856 metric_logger.py:196] completed step: 5, seconds: 0.140, TFLOP/s/device: 96.978, Tokens/s/device: 14617.608, total_weights: 65536, loss: 8.894, lm_loss: 8.894, perplexity: 7286.069
I0422 14:55:27.388057    2838 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0422 14:55:29.982092 138550326580992 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 648 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/nnx_xpk_feat_nnx_post_train_fixes_20260422_123906_13_scan_layers_false/checkpoints/0/items/array_metadatas/process_5
I0422 14:55:48.389321 138678686721856 metric_logger.py:196] completed step: 6, seconds: 0.278, TFLOP/s/device: 48.956, Tokens/s/device: 7379.196, total_weights: 65536, loss: 8.595, lm_loss: 8.595, perplexity: 5404.213
I0422 14:55:48.524885 138678686721856 metric_logger.py:196] completed step: 7, seconds: 22.640, TFLOP/s/device: 0.600, Tokens/s/device: 90.458, total_weights: 65536, loss: 8.387, lm_loss: 8.387, perplexity: 4388.456
I0422 14:55:48.661036 138678686721856 metric_logger.py:196] completed step: 8, seconds: 0.147, TFLOP/s/device: 92.322, Tokens/s/device: 13915.785, total_weights: 65536, loss: 8.258, lm_loss: 8.258, perplexity: 3856.630
I0422 14:55:48.803497 138678686721856 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0422 14:55:48.807770 138678686721856 checkpointing.py:798] Waited 0.00428462028503418 seconds for step 9 to finish before starting checkpointing.
I0422 14:55:49.363769 138678686721856 checkpoint_manager.py:2020] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 14:55:57.470356 138550821975808 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/gbytes_per_sec: 46.829 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 33.731523513793945 s) (per-host)
I0422 14:55:57.470488 138550821975808 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 33.342909s.
I0422 14:56:06.524616 138550821975808 async_checkpointer.py:160] [process=5][thread=async_save] Background save thread done. Time taken: 42.397024s.
I0422 14:56:06.524932 138550318188288 async_checkpointer.py:288] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 14:56:06.525055 138550318188288 async_checkpointer.py:298] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0422 14:56:06.525108 138550318188288 checkpoint_manager.py:2137] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0422 14:56:06.527725 138550318188288 checkpoint_manager.py:2146] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0422 14:56:06.527898 138678686721856 checkpoint_manager.py:2032] [process=5][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0422 14:56:06.528031 138678686721856 checkpoint_manager.py:1452] Waiting for previous save to complete took 17.164273 seconds. If this number is high, consider checkpointing less frequently.
I0422 14:56:06.530630 138678686721856 checkpoint_manager.py:1512] [process=5] Saving checkpoint at step 9
I0422 14:56:06.532746 138678686721856 event_tracking.py:70] [process=5] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/nnx_xpk_feat_nnx_post_train_fixes_20260422_123906_13_scan_layers_false/checkpoints/9.
I0422 14:56:07.821984 138678686721856 jax_array_handlers.py:360] Scheduling D2H of 648 prioritized jax.Array.
I0422 14:56:07.822604 138678686721856 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
I0422 14:56:07.950098 138678686721856 base_pytree_checkpoint_handler.py:154] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.133971s
I0422 14:56:07.950260 138678686721856 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/blocking_gbytes_per_sec: 2.192 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.7038896083831787 s) (per-host)
I0422 14:56:07.950320 138678686721856 base_pytree_checkpoint_handler.py:768] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.703965s (batch_requests_ready=0.023287s, total_serialization_initiated=0.680595s, others=0.000083s)
I0422 14:56:07.950766 138678686721856 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.708286s (all_items=0.000016s, per_item={'items': '0.00001597'}, temp_paths=0.708270)
I0422 14:56:07.951797 138678686721856 event_tracking.py:125] [process=5] [async] Finished blocking save in 1.42 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/nnx_xpk_feat_nnx_post_train_fixes_20260422_123906_13_scan_layers_false/checkpoints/9.
I0422 14:56:07.952074 138550326580992 async_checkpointer.py:76] [process=5][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 15:16:07.952049
I0422 14:56:07.976197 138678686721856 checkpoint_manager.py:1560] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 14:56:07.976520 138550318188288 async_checkpointer.py:280] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0422 14:56:07.976702 138678686721856 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/nnx_xpk_feat_nnx_post_train_fixes_20260422_123906_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776869749.363726, 'wait_for_prev_duration_secs': 17.16427254676819, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776869766.530693, 'checkpointer_blocking_duration_secs': 1.421475887298584, 'get_old_steps_start_time': 1776869767.952194, 'get_old_steps_duration_secs': 3.0517578125e-05, 'checkpoint_manager_blocking_start_time': 1776869748.8088186, 'checkpoint_manager_blocking_duration_secs': 19.167845964431763}
I0422 14:56:07.976945 138678686721856 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0422 14:56:07.976992 138678686721856 checkpoint_manager.py:2020] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 14:56:13.090818 138651933792000 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 648 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/nnx_xpk_feat_nnx_post_train_fixes_20260422_123906_13_scan_layers_false/checkpoints/9/items/array_metadatas/process_5
I0422 14:56:49.518583 138550326580992 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/gbytes_per_sec: 37.367 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.27219820022583 s) (per-host)
I0422 14:56:49.518704 138550326580992 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.566556s.
I0422 14:56:58.559231 138550326580992 async_checkpointer.py:160] [process=5][thread=async_save] Background save thread done. Time taken: 50.607067s.
I0422 14:56:58.559505 138550318188288 async_checkpointer.py:288] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 14:56:58.559620 138550318188288 async_checkpointer.py:298] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0422 14:56:58.559683 138550318188288 checkpoint_manager.py:2137] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0422 14:56:58.563431 138550318188288 checkpoint_manager.py:2146] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0422 14:56:58.563623 138678686721856 checkpoint_manager.py:2032] [process=5][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0422 14:56:58.563950 138678686721856 checkpoint_manager.py:2009] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 14:56:58.564945 138678686721856 metric_logger.py:196] completed step: 9, seconds: 0.135, TFLOP/s/device: 100.414, Tokens/s/device: 15135.503, total_weights: 65536, loss: 8.182, lm_loss: 8.182, perplexity: 3574.512
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Wed Apr 22 14:57:08 UTC 2026
EXIT_CODE=0