MaxView

XPK Start: Thu Apr 23 14:58:26 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. 
2026-04-23 14:58:51.753011: 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)
I0423 14:58:51.968037 140078499977024 max_utils.py:273] Attempting to initialize the jax distributed system...
I0423 14:59:01.008135 140078499977024 distributed.py:149] Starting JAX distributed service on [::]:8482
I0423 14:59:01.010507 140078499977024 distributed.py:172] Connecting to JAX distributed service on mt-13-scan-layers-false-ln8uv-slice-job-0-0.mt-13-scan-layers-false-ln8uv:8482
I0423 14:59:01.802176 140078499977024 max_utils.py:284] Jax distributed system initialized!
I0423 14:59:08.083616 140078499977024 max_utils.py:800] System Information: Jax Version: 0.9.2
I0423 14:59:08.083742 140078499977024 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0423 14:59:08.083786 140078499977024 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
I0423 14:59:08.083824 140078499977024 train_utils.py:391] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0423 14:59:08.776582 140078499977024 maxtext_utils.py:1771] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0423 14:59:09.636248 140078499977024 checkpointing.py:688] Setting up checkpoint logger...
I0423 14:59:09.636381 140078499977024 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0423 14:59:09.636427 140078499977024 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0423 14:59:09.636688 140078499977024 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 0x7f65d9653bf0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0423 14:59:12.513150 140078499977024 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0423 14:59:12.513394 140078499977024 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 0x7f51ac465e50>}, handler_registry=None
I0423 14:59:12.513636 140078499977024 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7f51ac465e50>` 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`.
I0423 14:59:12.513698 140078499977024 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f51ac323740>` 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`.
I0423 14:59:12.513736 140078499977024 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 0x7f51ac465e50>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7f51ac465e50>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f51ac323740>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f51ac323740>}).
I0423 14:59:12.514070 140078499977024 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0423 14:59:12.514142 140078499977024 async_checkpointer.py:192] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7f50cc6be520> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0423 14:59:14.159556 140078499977024 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260423_124541/nnx_xpk_feat_nnx_post_train_fixes_20260423_124541_13_scan_layers_false/checkpoints
I0423 14:59:14.553393 140078499977024 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_20260423_124541/nnx_xpk_feat_nnx_post_train_fixes_20260423_124541_13_scan_layers_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7f51ac323590>
I0423 14:59:14.553570 140078499977024 checkpointing.py:302] Checkpoint manager created!
I0423 14:59:16.727405 140078499977024 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0423 14:59:16.727535 140078499977024 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0423 14:59:24.727742 140078499977024 maxtext_utils.py:1889]  decoder/decoder_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.727868 140078499977024 maxtext_utils.py:1889]  decoder/layers/0/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.727924 140078499977024 maxtext_utils.py:1889]  decoder/layers/0/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.727974 140078499977024 maxtext_utils.py:1889]  decoder/layers/0/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.728015 140078499977024 maxtext_utils.py:1889]  decoder/layers/0/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.728053 140078499977024 maxtext_utils.py:1889]  decoder/layers/0/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.728091 140078499977024 maxtext_utils.py:1889]  decoder/layers/0/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.728130 140078499977024 maxtext_utils.py:1889]  decoder/layers/0/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.728166 140078499977024 maxtext_utils.py:1889]  decoder/layers/0/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.728199 140078499977024 maxtext_utils.py:1889]  decoder/layers/0/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.728231 140078499977024 maxtext_utils.py:1889]  decoder/layers/1/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.728263 140078499977024 maxtext_utils.py:1889]  decoder/layers/1/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.728295 140078499977024 maxtext_utils.py:1889]  decoder/layers/1/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.728327 140078499977024 maxtext_utils.py:1889]  decoder/layers/1/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.728357 140078499977024 maxtext_utils.py:1889]  decoder/layers/1/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.728390 140078499977024 maxtext_utils.py:1889]  decoder/layers/1/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.728422 140078499977024 maxtext_utils.py:1889]  decoder/layers/1/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.728453 140078499977024 maxtext_utils.py:1889]  decoder/layers/1/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.728484 140078499977024 maxtext_utils.py:1889]  decoder/layers/1/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.728516 140078499977024 maxtext_utils.py:1889]  decoder/layers/2/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.728548 140078499977024 maxtext_utils.py:1889]  decoder/layers/2/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.728579 140078499977024 maxtext_utils.py:1889]  decoder/layers/2/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.728610 140078499977024 maxtext_utils.py:1889]  decoder/layers/2/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.728639 140078499977024 maxtext_utils.py:1889]  decoder/layers/2/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.728684 140078499977024 maxtext_utils.py:1889]  decoder/layers/2/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.728717 140078499977024 maxtext_utils.py:1889]  decoder/layers/2/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.728749 140078499977024 maxtext_utils.py:1889]  decoder/layers/2/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.728779 140078499977024 maxtext_utils.py:1889]  decoder/layers/2/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.728812 140078499977024 maxtext_utils.py:1889]  decoder/layers/3/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.728844 140078499977024 maxtext_utils.py:1889]  decoder/layers/3/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.728875 140078499977024 maxtext_utils.py:1889]  decoder/layers/3/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.728904 140078499977024 maxtext_utils.py:1889]  decoder/layers/3/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.728933 140078499977024 maxtext_utils.py:1889]  decoder/layers/3/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.728970 140078499977024 maxtext_utils.py:1889]  decoder/layers/3/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729001 140078499977024 maxtext_utils.py:1889]  decoder/layers/3/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.729031 140078499977024 maxtext_utils.py:1889]  decoder/layers/3/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729061 140078499977024 maxtext_utils.py:1889]  decoder/layers/3/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729091 140078499977024 maxtext_utils.py:1889]  decoder/layers/4/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.729120 140078499977024 maxtext_utils.py:1889]  decoder/layers/4/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.729149 140078499977024 maxtext_utils.py:1889]  decoder/layers/4/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.729177 140078499977024 maxtext_utils.py:1889]  decoder/layers/4/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.729206 140078499977024 maxtext_utils.py:1889]  decoder/layers/4/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.729235 140078499977024 maxtext_utils.py:1889]  decoder/layers/4/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729266 140078499977024 maxtext_utils.py:1889]  decoder/layers/4/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.729296 140078499977024 maxtext_utils.py:1889]  decoder/layers/4/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729327 140078499977024 maxtext_utils.py:1889]  decoder/layers/4/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729358 140078499977024 maxtext_utils.py:1889]  decoder/layers/5/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.729387 140078499977024 maxtext_utils.py:1889]  decoder/layers/5/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.729418 140078499977024 maxtext_utils.py:1889]  decoder/layers/5/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.729445 140078499977024 maxtext_utils.py:1889]  decoder/layers/5/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.729473 140078499977024 maxtext_utils.py:1889]  decoder/layers/5/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.729503 140078499977024 maxtext_utils.py:1889]  decoder/layers/5/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729533 140078499977024 maxtext_utils.py:1889]  decoder/layers/5/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.729562 140078499977024 maxtext_utils.py:1889]  decoder/layers/5/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729592 140078499977024 maxtext_utils.py:1889]  decoder/layers/5/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729621 140078499977024 maxtext_utils.py:1889]  decoder/layers/6/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.729660 140078499977024 maxtext_utils.py:1889]  decoder/layers/6/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.729694 140078499977024 maxtext_utils.py:1889]  decoder/layers/6/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.729722 140078499977024 maxtext_utils.py:1889]  decoder/layers/6/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.729750 140078499977024 maxtext_utils.py:1889]  decoder/layers/6/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.729781 140078499977024 maxtext_utils.py:1889]  decoder/layers/6/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729811 140078499977024 maxtext_utils.py:1889]  decoder/layers/6/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.729859 140078499977024 maxtext_utils.py:1889]  decoder/layers/6/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729892 140078499977024 maxtext_utils.py:1889]  decoder/layers/6/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.729922 140078499977024 maxtext_utils.py:1889]  decoder/layers/7/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.729951 140078499977024 maxtext_utils.py:1889]  decoder/layers/7/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.729987 140078499977024 maxtext_utils.py:1889]  decoder/layers/7/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.730021 140078499977024 maxtext_utils.py:1889]  decoder/layers/7/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.730050 140078499977024 maxtext_utils.py:1889]  decoder/layers/7/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.730080 140078499977024 maxtext_utils.py:1889]  decoder/layers/7/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730110 140078499977024 maxtext_utils.py:1889]  decoder/layers/7/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.730140 140078499977024 maxtext_utils.py:1889]  decoder/layers/7/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730170 140078499977024 maxtext_utils.py:1889]  decoder/layers/7/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730199 140078499977024 maxtext_utils.py:1889]  decoder/layers/8/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.730229 140078499977024 maxtext_utils.py:1889]  decoder/layers/8/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.730259 140078499977024 maxtext_utils.py:1889]  decoder/layers/8/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.730286 140078499977024 maxtext_utils.py:1889]  decoder/layers/8/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.730314 140078499977024 maxtext_utils.py:1889]  decoder/layers/8/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.730343 140078499977024 maxtext_utils.py:1889]  decoder/layers/8/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730374 140078499977024 maxtext_utils.py:1889]  decoder/layers/8/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.730404 140078499977024 maxtext_utils.py:1889]  decoder/layers/8/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730434 140078499977024 maxtext_utils.py:1889]  decoder/layers/8/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730464 140078499977024 maxtext_utils.py:1889]  decoder/layers/9/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.730494 140078499977024 maxtext_utils.py:1889]  decoder/layers/9/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.730524 140078499977024 maxtext_utils.py:1889]  decoder/layers/9/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.730551 140078499977024 maxtext_utils.py:1889]  decoder/layers/9/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.730579 140078499977024 maxtext_utils.py:1889]  decoder/layers/9/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.730609 140078499977024 maxtext_utils.py:1889]  decoder/layers/9/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730639 140078499977024 maxtext_utils.py:1889]  decoder/layers/9/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.730680 140078499977024 maxtext_utils.py:1889]  decoder/layers/9/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730711 140078499977024 maxtext_utils.py:1889]  decoder/layers/9/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730743 140078499977024 maxtext_utils.py:1889]  decoder/layers/10/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.730774 140078499977024 maxtext_utils.py:1889]  decoder/layers/10/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.730803 140078499977024 maxtext_utils.py:1889]  decoder/layers/10/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.730831 140078499977024 maxtext_utils.py:1889]  decoder/layers/10/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.730859 140078499977024 maxtext_utils.py:1889]  decoder/layers/10/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.730889 140078499977024 maxtext_utils.py:1889]  decoder/layers/10/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730919 140078499977024 maxtext_utils.py:1889]  decoder/layers/10/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.730948 140078499977024 maxtext_utils.py:1889]  decoder/layers/10/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.730983 140078499977024 maxtext_utils.py:1889]  decoder/layers/10/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731013 140078499977024 maxtext_utils.py:1889]  decoder/layers/11/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.731042 140078499977024 maxtext_utils.py:1889]  decoder/layers/11/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.731072 140078499977024 maxtext_utils.py:1889]  decoder/layers/11/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.731100 140078499977024 maxtext_utils.py:1889]  decoder/layers/11/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.731127 140078499977024 maxtext_utils.py:1889]  decoder/layers/11/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.731158 140078499977024 maxtext_utils.py:1889]  decoder/layers/11/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731187 140078499977024 maxtext_utils.py:1889]  decoder/layers/11/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.731216 140078499977024 maxtext_utils.py:1889]  decoder/layers/11/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731246 140078499977024 maxtext_utils.py:1889]  decoder/layers/11/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731275 140078499977024 maxtext_utils.py:1889]  decoder/layers/12/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.731305 140078499977024 maxtext_utils.py:1889]  decoder/layers/12/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.731334 140078499977024 maxtext_utils.py:1889]  decoder/layers/12/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.731361 140078499977024 maxtext_utils.py:1889]  decoder/layers/12/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.731389 140078499977024 maxtext_utils.py:1889]  decoder/layers/12/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.731418 140078499977024 maxtext_utils.py:1889]  decoder/layers/12/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731448 140078499977024 maxtext_utils.py:1889]  decoder/layers/12/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.731477 140078499977024 maxtext_utils.py:1889]  decoder/layers/12/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731507 140078499977024 maxtext_utils.py:1889]  decoder/layers/12/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731536 140078499977024 maxtext_utils.py:1889]  decoder/layers/13/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.731565 140078499977024 maxtext_utils.py:1889]  decoder/layers/13/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.731595 140078499977024 maxtext_utils.py:1889]  decoder/layers/13/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.731623 140078499977024 maxtext_utils.py:1889]  decoder/layers/13/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.731664 140078499977024 maxtext_utils.py:1889]  decoder/layers/13/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.731698 140078499977024 maxtext_utils.py:1889]  decoder/layers/13/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731728 140078499977024 maxtext_utils.py:1889]  decoder/layers/13/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.731758 140078499977024 maxtext_utils.py:1889]  decoder/layers/13/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731787 140078499977024 maxtext_utils.py:1889]  decoder/layers/13/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731817 140078499977024 maxtext_utils.py:1889]  decoder/layers/14/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.731847 140078499977024 maxtext_utils.py:1889]  decoder/layers/14/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.731877 140078499977024 maxtext_utils.py:1889]  decoder/layers/14/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.731905 140078499977024 maxtext_utils.py:1889]  decoder/layers/14/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.731933 140078499977024 maxtext_utils.py:1889]  decoder/layers/14/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.731966 140078499977024 maxtext_utils.py:1889]  decoder/layers/14/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.731997 140078499977024 maxtext_utils.py:1889]  decoder/layers/14/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.732026 140078499977024 maxtext_utils.py:1889]  decoder/layers/14/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.732056 140078499977024 maxtext_utils.py:1889]  decoder/layers/14/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.732085 140078499977024 maxtext_utils.py:1889]  decoder/layers/15/mlp/wi_0/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.732115 140078499977024 maxtext_utils.py:1889]  decoder/layers/15/mlp/wi_1/kernel/value
    Shape:     float32[2048,7168]
    Physical:  ('fsdp', None)
I0423 14:59:24.732143 140078499977024 maxtext_utils.py:1889]  decoder/layers/15/mlp/wo/kernel/value
    Shape:     float32[7168,2048]
    Physical:  (None, 'fsdp')
I0423 14:59:24.732171 140078499977024 maxtext_utils.py:1889]  decoder/layers/15/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.732200 140078499977024 maxtext_utils.py:1889]  decoder/layers/15/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0423 14:59:24.732230 140078499977024 maxtext_utils.py:1889]  decoder/layers/15/self_attention/key/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.732259 140078499977024 maxtext_utils.py:1889]  decoder/layers/15/self_attention/out/kernel/value
    Shape:     float32[16,128,2048]
    Physical:  (None, None, 'fsdp')
I0423 14:59:24.732288 140078499977024 maxtext_utils.py:1889]  decoder/layers/15/self_attention/query/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.732317 140078499977024 maxtext_utils.py:1889]  decoder/layers/15/self_attention/value/kernel/value
    Shape:     float32[2048,16,128]
    Physical:  ('fsdp', None, None)
I0423 14:59:24.732347 140078499977024 maxtext_utils.py:1889]  decoder/logits_dense/kernel/value
    Shape:     float32[2048,32000]
    Physical:  ('fsdp', None)
I0423 14:59:24.732377 140078499977024 maxtext_utils.py:1889]  token_embedder/embedding/value
    Shape:     float32[32000,2048]
    Physical:  (None, 'fsdp')

I0423 14:59:25.235250 140078499977024 nnx_decoders.py:1023] nnx_decoders/y_in Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0423 14:59:25.235349 140078499977024 nnx_decoders.py:1023] nnx_decoders/y_in Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 14:59:25.240712 140078499977024 nnx_decoders.py:1023] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0423 14:59:25.240817 140078499977024 nnx_decoders.py:1023] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 14:59:25.261646 140078499977024 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0423 14:59:25.261729 140078499977024 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 14:59:25.278061 140078499977024 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0423 14:59:25.278116 140078499977024 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 14:59:25.301537 140078499977024 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 14:59:25.301604 140078499977024 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 14:59:25.317899 140078499977024 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 14:59:25.317966 140078499977024 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 14:59:25.334256 140078499977024 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 14:59:25.334317 140078499977024 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 14:59:25.362859 140078499977024 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0423 14:59:25.362924 140078499977024 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 14:59:25.382603 140078499977024 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0423 14:59:25.382680 140078499977024 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0423 15:00:26.502668 140078499977024 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.
I0423 15:00:26.513393 140078499977024 metric_logger.py:301] number parameters: 1.104 billion
I0423 15:01:27.693546 140078499977024 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0423 15:01:27.919262 140078499977024 checkpointing.py:798] Waited 0.22572970390319824 seconds for step 0 to finish before starting checkpointing.
I0423 15:01:27.922608 140078499977024 checkpoint_manager.py:2009] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 15:01:27.924577 140078499977024 checkpoint_manager.py:1512] [process=5] Saving checkpoint at step 0
I0423 15:01:27.925995 140078499977024 event_tracking.py:70] [process=5] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260423_124541/nnx_xpk_feat_nnx_post_train_fixes_20260423_124541_13_scan_layers_false/checkpoints/0.
I0423 15:01:28.331008 140078499977024 signaling_client.py:364] Using JaxDistributedSignalingClient
I0423 15:01:28.332135 140078499977024 jax_array_handlers.py:360] Scheduling D2H of 648 prioritized jax.Array.
I0423 15:01:28.332245 140078499977024 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
I0423 15:01:28.657518 140078499977024 base_pytree_checkpoint_handler.py:154] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.331399s
I0423 15:01:28.657710 140078499977024 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/blocking_gbytes_per_sec: 3.944 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3911135196685791 s) (per-host)
I0423 15:01:28.657761 140078499977024 base_pytree_checkpoint_handler.py:768] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.391176s (batch_requests_ready=0.025417s, total_serialization_initiated=0.365689s, others=0.000070s)
I0423 15:01:28.658146 140078499977024 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.395905s (all_items=0.000018s, per_item={'items': '0.00001788'}, temp_paths=0.395887)
I0423 15:01:28.659002 140078499977024 event_tracking.py:125] [process=5] [async] Finished blocking save in 0.73 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260423_124541/nnx_xpk_feat_nnx_post_train_fixes_20260423_124541_13_scan_layers_false/checkpoints/0.
I0423 15:01:28.659357 139951356786432 async_checkpointer.py:76] [process=5][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 15:21:28.659319
I0423 15:01:28.681929 140078499977024 checkpoint_manager.py:1560] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 15:01:28.682230 139951326349056 async_checkpointer.py:280] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0423 15:01:28.682389 140078499977024 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260423_124541/nnx_xpk_feat_nnx_post_train_fixes_20260423_124541_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776956487.9225905, 'wait_for_prev_duration_secs': 7.796287536621094e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776956487.9246163, 'checkpointer_blocking_duration_secs': 0.7348949909210205, 'get_old_steps_start_time': 1776956488.659535, 'get_old_steps_duration_secs': 2.9325485229492188e-05, 'checkpoint_manager_blocking_start_time': 1776956487.9205244, 'checkpoint_manager_blocking_duration_secs': 0.7618248462677002}
I0423 15:01:28.682628 140078499977024 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0423 15:01:28.682840 140078499977024 max_utils.py:750] 
Memstats: After params initialized:
I0423 15:01:28.682895 140078499977024 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_18(process=5,(2,4,0,0))
I0423 15:01:28.682928 140078499977024 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_19(process=5,(3,4,0,0))
I0423 15:01:28.682963 140078499977024 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_22(process=5,(2,5,0,0))
I0423 15:01:28.682991 140078499977024 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_23(process=5,(3,5,0,0))
I0423 15:01:29.077739 140078499977024 metric_logger.py:196] completed step: 0, seconds: 61.173, TFLOP/s/device: 0.222, Tokens/s/device: 33.479, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52794.566
I0423 15:01:29.213729 140078499977024 metric_logger.py:196] completed step: 1, seconds: 1.373, TFLOP/s/device: 9.893, Tokens/s/device: 1491.194, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52794.566
I0423 15:01:29.747378 140078499977024 metric_logger.py:196] completed step: 2, seconds: 0.056, TFLOP/s/device: 243.349, Tokens/s/device: 36680.159, total_weights: 65536, loss: 10.265, lm_loss: 10.265, perplexity: 28706.648
I0423 15:01:29.886144 140078499977024 metric_logger.py:196] completed step: 3, seconds: 0.425, TFLOP/s/device: 31.941, Tokens/s/device: 4814.474, total_weights: 65536, loss: 9.736, lm_loss: 9.736, perplexity: 16920.650
I0423 15:01:30.161267 140078499977024 metric_logger.py:196] completed step: 4, seconds: 0.218, TFLOP/s/device: 62.427, Tokens/s/device: 9409.689, total_weights: 65536, loss: 9.279, lm_loss: 9.279, perplexity: 10711.495
I0423 15:01:30.184226 140078499977024 metric_logger.py:196] completed step: 5, seconds: 0.139, TFLOP/s/device: 97.963, Tokens/s/device: 14766.107, total_weights: 65536, loss: 8.894, lm_loss: 8.894, perplexity: 7285.978
I0423 15:01:31.667377    2812 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0423 15:01:33.887871 139951334741760 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_20260423_124541/nnx_xpk_feat_nnx_post_train_fixes_20260423_124541_13_scan_layers_false/checkpoints/0/items/array_metadatas/process_5
I0423 15:01:52.746886 140078499977024 metric_logger.py:196] completed step: 6, seconds: 0.275, TFLOP/s/device: 49.424, Tokens/s/device: 7449.711, total_weights: 65536, loss: 8.595, lm_loss: 8.595, perplexity: 5404.313
I0423 15:01:52.882663 140078499977024 metric_logger.py:196] completed step: 7, seconds: 22.439, TFLOP/s/device: 0.606, Tokens/s/device: 91.269, total_weights: 65536, loss: 8.387, lm_loss: 8.387, perplexity: 4389.132
I0423 15:01:53.018008 140078499977024 metric_logger.py:196] completed step: 8, seconds: 0.147, TFLOP/s/device: 92.284, Tokens/s/device: 13910.020, total_weights: 65536, loss: 8.258, lm_loss: 8.258, perplexity: 3856.806
I0423 15:01:53.161137 140078499977024 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0423 15:01:53.165379 140078499977024 checkpointing.py:798] Waited 0.004250049591064453 seconds for step 9 to finish before starting checkpointing.
I0423 15:01:53.168615 140078499977024 checkpoint_manager.py:2020] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 15:02:02.218989 139951356786432 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/gbytes_per_sec: 46.524 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 33.952330350875854 s) (per-host)
I0423 15:02:02.219124 139951356786432 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 33.559651s.
I0423 15:02:11.287164 139951356786432 async_checkpointer.py:160] [process=5][thread=async_save] Background save thread done. Time taken: 42.627676s.
I0423 15:02:11.287465 139951326349056 async_checkpointer.py:288] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 15:02:11.287584 139951326349056 async_checkpointer.py:298] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0423 15:02:11.287636 139951326349056 checkpoint_manager.py:2137] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0423 15:02:11.290417 139951326349056 checkpoint_manager.py:2146] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0423 15:02:11.290601 140078499977024 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.
W0423 15:02:11.290744 140078499977024 checkpoint_manager.py:1452] Waiting for previous save to complete took 18.122128 seconds. If this number is high, consider checkpointing less frequently.
I0423 15:02:11.293358 140078499977024 checkpoint_manager.py:1512] [process=5] Saving checkpoint at step 9
I0423 15:02:11.295627 140078499977024 event_tracking.py:70] [process=5] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260423_124541/nnx_xpk_feat_nnx_post_train_fixes_20260423_124541_13_scan_layers_false/checkpoints/9.
I0423 15:02:12.599553 140078499977024 jax_array_handlers.py:360] Scheduling D2H of 648 prioritized jax.Array.
I0423 15:02:12.599756 140078499977024 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
I0423 15:02:12.724618 140078499977024 base_pytree_checkpoint_handler.py:154] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.131411s
I0423 15:02:12.724788 140078499977024 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/blocking_gbytes_per_sec: 2.151 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.7171933650970459 s) (per-host)
I0423 15:02:12.724838 140078499977024 base_pytree_checkpoint_handler.py:768] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.717252s (batch_requests_ready=0.024899s, total_serialization_initiated=0.692287s, others=0.000065s)
I0423 15:02:12.725224 140078499977024 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.721679s (all_items=0.000016s, per_item={'items': '0.00001645'}, temp_paths=0.721663)
I0423 15:02:12.725973 140078499977024 event_tracking.py:125] [process=5] [async] Finished blocking save in 1.43 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260423_124541/nnx_xpk_feat_nnx_post_train_fixes_20260423_124541_13_scan_layers_false/checkpoints/9.
I0423 15:02:12.726334 139951326349056 async_checkpointer.py:76] [process=5][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 15:22:12.726298
I0423 15:02:12.742974 140078499977024 checkpoint_manager.py:1560] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 15:02:12.743305 139950821463808 async_checkpointer.py:280] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0423 15:02:12.743472 140078499977024 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260423_124541/nnx_xpk_feat_nnx_post_train_fixes_20260423_124541_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776956513.1685863, 'wait_for_prev_duration_secs': 18.122127532958984, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776956531.2934008, 'checkpointer_blocking_duration_secs': 1.4330365657806396, 'get_old_steps_start_time': 1776956532.7264612, 'get_old_steps_duration_secs': 3.0517578125e-05, 'checkpoint_manager_blocking_start_time': 1776956513.1664424, 'checkpoint_manager_blocking_duration_secs': 19.576995849609375}
I0423 15:02:12.743727 140078499977024 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0423 15:02:12.743777 140078499977024 checkpoint_manager.py:2020] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 15:02:17.959216 139951334741760 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_20260423_124541/nnx_xpk_feat_nnx_post_train_fixes_20260423_124541_13_scan_layers_false/checkpoints/9/items/array_metadatas/process_5
I0423 15:02:54.118032 139951326349056 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/gbytes_per_sec: 37.511 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.11039662361145 s) (per-host)
I0423 15:02:54.118165 139951326349056 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.391048s.
I0423 15:03:05.177930 139951326349056 async_checkpointer.py:160] [process=5][thread=async_save] Background save thread done. Time taken: 52.450799s.
I0423 15:03:05.178216 139950821463808 async_checkpointer.py:288] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 15:03:05.178336 139950821463808 async_checkpointer.py:298] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0423 15:03:05.178387 139950821463808 checkpoint_manager.py:2137] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0423 15:03:05.180258 139950821463808 checkpoint_manager.py:2146] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0423 15:03:05.180458 140078499977024 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.
I0423 15:03:05.180803 140078499977024 checkpoint_manager.py:2009] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 15:03:05.181680 140078499977024 metric_logger.py:196] completed step: 9, seconds: 0.135, TFLOP/s/device: 100.598, Tokens/s/device: 15163.294, total_weights: 65536, loss: 8.182, lm_loss: 8.182, perplexity: 3574.803
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Thu Apr 23 15:03:17 UTC 2026
EXIT_CODE=0