XPK Start: Sun Apr 19 04:53:02 UTC 2026 2026-04-19 04:53:06.437941: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:467] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered WARNING: All log messages before absl::InitializeLog() is called are written to STDERR E0000 00:00:1776574386.450831 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776574386.454512 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776574386.466036 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776574386.466057 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776574386.466059 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776574386.466061 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-19 04:53:25.572824: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303) I0419 04:53:26.102316 135447319566144 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-19 04:53:35,143:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0419 04:53:35.143900 135447319566144 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-19 04:53:35,146:jax._src.distributed:157: Connecting to JAX distributed service on mt-16-zarr3-false-rs-z7006-slice-job-0-0.mt-16-zarr3-false-rs-z7006:8482 I0419 04:53:35.146147 135447319566144 distributed.py:157] Connecting to JAX distributed service on mt-16-zarr3-false-rs-z7006-slice-job-0-0.mt-16-zarr3-false-rs-z7006:8482 I0419 04:53:36.710688 135447319566144 max_utils.py:284] Jax distributed system initialized! I0419 04:53:43.557512 135447319566144 max_utils.py:800] System Information: Jax Version: 0.8.1 I0419 04:53:43.557622 135447319566144 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0419 04:53:43.557673 135447319566144 max_utils.py:802] System Information: Jax Backend: PJRT C API TFRT TPU v6 lite Built on Nov 12 2025 14:16:36 (1762985796) cl/831091709 I0419 04:53:43.557720 135447319566144 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0419 04:53:44.171194 135447319566144 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 04:53:44.171856 135447319566144 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 04:53:44.172047 135447319566144 checkpointing.py:688] Setting up checkpoint logger... I0419 04:53:44.172102 135447319566144 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=False I0419 04:53:44.172150 135447319566144 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0419 04:53:44.172500 135447319566144 base_pytree_checkpoint_handler.py:415] Created BasePyTreeCheckpointHandler: use_ocdbt=True, use_zarr3=False, pytree_metadata_options=PyTreeMetadataOptions(support_rich_types=False), array_metadata_store=<orbax.checkpoint._src.metadata.array_metadata_store.Store object at 0x7b2fc77786e0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0419 04:53:45.544154 135447319566144 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0419 04:53:45.544502 135447319566144 checkpoint_manager.py:709] [process=0][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b1b8063db20>}, handler_registry=None I0419 04:53:45.544739 135447319566144 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b1b8063db20>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`. I0419 04:53:45.544788 135447319566144 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b1b8025fef0>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`. I0419 04:53:45.544839 135447319566144 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 0x7b1b8063db20>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b1b8063db20>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b1b8025fef0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b1b8025fef0>}). I0419 04:53:45.545373 135447319566144 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0419 04:53:45.545453 135447319566144 async_checkpointer.py:177] [process=0][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7b1b80408180> timeout: 600 secs and primary_host=0 for async checkpoint writes I0419 04:53:49.119044 135326232213248 checkpoint.py:228] Read Metadata={'item_handlers': {'items': 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler'}, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776574259581887249, 'commit_timestamp_nsecs': 1776574307530321616, 'custom_metadata': {}} from gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/9/_CHECKPOINT_METADATA I0419 04:53:49.368428 135326232213248 checkpoint_manager.py:1877] Missing metrics for step 9 I0419 04:53:50.211412 135447319566144 checkpoint_manager.py:1818] Found 3 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints I0419 04:53:51.097779 135447319566144 checkpoint_manager.py:929] [process=0][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=5), 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/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7b1b80210170> I0419 04:53:51.097985 135447319566144 checkpointing.py:302] Checkpoint manager created! I0419 04:53:51.972929 135447319566144 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0419 04:53:51.973040 135447319566144 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 04:53:52.344734 135447319566144 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 04:53:52.344838 135447319566144 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 04:53:52.360440 135447319566144 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 04:53:52.360499 135447319566144 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 04:53:52.383297 135447319566144 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 04:53:52.383367 135447319566144 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:53:52.399189 135447319566144 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 04:53:52.399252 135447319566144 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:53:52.414888 135447319566144 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 04:53:52.414948 135447319566144 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:53:52.439777 135447319566144 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0419 04:53:52.439861 135447319566144 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:53:52.459348 135447319566144 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0419 04:53:52.459420 135447319566144 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0419 04:53:52.651638 135447319566144 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0419 04:53:52.651750 135447319566144 checkpointing.py:582] restoring from this run's directory step 9 I0419 04:53:53.567038 135447319566144 checkpointer.py:304] Restoring checkpoint from gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/9. I0419 04:53:56.071594 1426 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0419 04:54:02.700982 135447319566144 jax_array_handlers.py:837] [process=0] /jax/orbax/read/worker/io/requested throughput: 238.188 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 6.63456392288208 s) (per-host) I0419 04:54:02.716767 135447319566144 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/read/gbytes_per_sec: 224.273 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 7.056540012359619 s) (per-host) I0419 04:54:04.188863 135447319566144 checkpointer.py:318] Finished restoring checkpoint in 11.29 seconds from gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/9. I0419 04:54:04.189046 135447319566144 standard_logger.py:34] {'step': 9, 'event_type': 'restore', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints', 'checkpointer_start_time': 1776574432.8942454, 'checkpointer_duration_secs': 11.294768571853638, 'checkpoint_manager_start_time': 1776574432.6520088, 'checkpoint_manager_duration_secs': 11.537007808685303} fsdp: 32 I0419 04:54:04.593873 135447319566144 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0419 04:54:04.594006 135447319566144 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0419 04:54:04.594061 135447319566144 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0419 04:54:04.594129 135447319566144 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: PartitionSpec('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0419 04:54:04.594192 135447319566144 maxtext_utils.py:1740] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0419 04:54:04.594234 135447319566144 maxtext_utils.py:1740] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0419 04:54:04.594290 135447319566144 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/key/kernel Shape: float32[2048,16,16,128] Logical: PartitionSpec('embed', 'layers', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None, None) I0419 04:54:04.594344 135447319566144 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/out/kernel Shape: float32[16,16,128,2048] Logical: PartitionSpec('heads', 'layers', 'kv', 'embed') Physical: (None, None, None, 'fsdp') I0419 04:54:04.594387 135447319566144 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/query/kernel Shape: float32[2048,16,16,128] Logical: PartitionSpec('embed', 'layers', 'q_heads', 'kv') Physical: ('fsdp', None, None, None) I0419 04:54:04.594426 135447319566144 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/value/kernel Shape: float32[2048,16,16,128] Logical: PartitionSpec('embed', 'layers', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None, None) I0419 04:54:04.594473 135447319566144 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0419 04:54:04.594519 135447319566144 maxtext_utils.py:1740] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0419 04:54:05.108603 135447319566144 train.py:158] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 04:54:05.108695 135447319566144 train.py:158] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0419 04:54:05.123724 135447319566144 train.py:165] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 04:54:05.123783 135447319566144 train.py:165] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0419 04:54:15.903426 135447319566144 max_utils.py:791] Total memory size: 1.5 GB, Output size: 0.4 GB, Temp size: 1.1 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0419 04:54:15.909597 135447319566144 metric_logger.py:289] number parameters: 1.104 billion I0419 04:54:27.395862 135447319566144 checkpointing.py:794] Waiting for step 10 to finish before checkpoint... I0419 04:54:27.548498 135447319566144 checkpointing.py:798] Waited 0.15261554718017578 seconds for step 10 to finish before starting checkpointing. I0419 04:54:27.550914 135447319566144 checkpoint_manager.py:2013] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 04:54:27.552589 135447319566144 checkpoint_manager.py:1518] [process=0] Saving checkpoint at step 10 I0419 04:54:27.555335 135447319566144 async_checkpointer.py:452] [process=0] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10. I0419 04:54:28.711905 135447319566144 signaling_client.py:364] Using JaxDistributedSignalingClient I0419 04:54:29.004106 135447319566144 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 04:54:29.004212 135447319566144 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0419 04:54:29.251313 135447319566144 base_pytree_checkpoint_handler.py:153] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.248254s I0419 04:54:29.252229 135447319566144 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.899 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5321474075317383 s) (per-host) I0419 04:54:29.252301 135447319566144 base_pytree_checkpoint_handler.py:737] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.532239s (batch_requests_ready=0.280574s, total_serialization_initiated=0.250827s, others=0.000838s) I0419 04:54:29.252379 135447319566144 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.536381s (all_items=0.000019s, per_item={'items': '0.00001884'}, temp_paths=0.536363) I0419 04:54:29.253254 135323505895168 async_checkpointer.py:79] [process=0][thread=async_save] Background save thread started. I0419 04:54:29.253412 135447319566144 async_checkpointer.py:561] Finished blocking save. Time taken: 1.700753s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10. I0419 04:54:29.396281 135321912067840 atomicity.py:137] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10 I0419 04:54:29.667338 135447319566144 checkpoint_manager.py:1566] [process=0][thread=MainThread][step=10] Starting CheckpointManager Save Finalize thread=save_finalize I0419 04:54:29.667719 135321391208192 async_checkpointer.py:265] [process=0][thread=save_finalize] Waiting for background save thread=async_save. I0419 04:54:29.667862 135447319566144 standard_logger.py:34] {'step': 10, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776574467.5508964, 'wait_for_prev_duration_secs': 6.198883056640625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776574467.5526268, 'checkpointer_blocking_duration_secs': 1.7008919715881348, 'get_old_steps_start_time': 1776574469.2535384, 'get_old_steps_duration_secs': 2.5033950805664062e-05, 'checkpoint_manager_blocking_start_time': 1776574467.5490813, 'checkpoint_manager_blocking_duration_secs': 2.1187429428100586} I0419 04:54:29.668025 135447319566144 checkpointing.py:409] Started an asynchronous checkpoint save for step 10 I0419 04:54:29.668080 135447319566144 max_utils.py:750] Memstats: After params initialized: I0419 04:54:29.668147 135447319566144 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_0(process=0,(0,0,0,0)) I0419 04:54:29.668183 135447319566144 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_1(process=0,(1,0,0,0)) I0419 04:54:29.668210 135447319566144 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_4(process=0,(0,1,0,0)) I0419 04:54:29.668236 135447319566144 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_5(process=0,(1,1,0,0)) I0419 04:54:29.960444 135447319566144 metric_logger.py:185] completed step: 10, seconds: 11.127, TFLOP/s/device: 1.221, Tokens/s/device: 184.065, total_weights: 65536, loss: 6.833 I0419 04:54:30.105937 135447319566144 metric_logger.py:185] completed step: 11, seconds: 2.563, TFLOP/s/device: 5.301, Tokens/s/device: 799.058, total_weights: 65536, loss: 6.528 I0419 04:54:30.253611 135447319566144 metric_logger.py:185] completed step: 12, seconds: 0.010, TFLOP/s/device: 1417.393, Tokens/s/device: 213644.899, total_weights: 65536, loss: 6.265 I0419 04:54:30.401142 135447319566144 metric_logger.py:185] completed step: 13, seconds: 0.147, TFLOP/s/device: 92.636, Tokens/s/device: 13963.033, total_weights: 65536, loss: 6.045 I0419 04:54:30.409988 135447319566144 checkpointing.py:794] Waiting for step 15 to finish before checkpoint... I0419 04:54:30.695963 135447319566144 checkpointing.py:798] Waited 0.28584909439086914 seconds for step 15 to finish before starting checkpointing. I0419 04:54:30.698974 135447319566144 checkpoint_manager.py:2024] [process=0][thread=MainThread][step=10][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 04:54:32.318711 135326190249728 atomicity.py:137] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10/items I0419 04:54:32.462023 135323364161280 checkpoint.py:188] Wrote Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776574471629153570, 'commit_timestamp_nsecs': None, 'custom_metadata': {}}, json={"item_handlers": null, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1776574471629153570, "commit_timestamp_nsecs": null, "custom_metadata": {}} to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10/_CHECKPOINT_METADATA I0419 04:54:35.485001 135325108115200 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10/items/array_metadatas/process_0 I0419 04:55:14.904704 135323514287872 base_pytree_checkpoint_handler.py:1232] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 41.061270s (commit=39.534473s, array_metadata_write=1.526798s) I0419 04:55:14.906211 135323505895168 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/gbytes_per_sec: 34.201 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 46.18609023094177 s) (per-host) I0419 04:55:14.906375 135323505895168 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 45.653007s. I0419 04:55:16.397689 135323364161280 checkpoint.py:247] Updated Metadata={'item_handlers': {'items': 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler'}, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776574471629153570, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10/_CHECKPOINT_METADATA I0419 04:55:17.221562 135323505895168 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000383s. I0419 04:55:18.569352 135323505895168 base_pytree_checkpoint_handler.py:1356] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 2.757968s. use_zarr3=False, enable_post_merge_validation=True, directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10/items I0419 04:55:18.571669 135323505895168 atomicity.py:608] Finalizing gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10/items I0419 04:55:18.983858 135323505895168 atomicity.py:608] Finalizing gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10 I0419 04:55:20.652923 135323505895168 atomicity.py:790] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10`. I0419 04:55:20.654011 135323505895168 async_checkpointer.py:420] Finished async_save (blocking + background). Time taken: 53.101352s. directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/10 I0419 04:55:20.655441 135323505895168 async_checkpointer.py:144] [process=0][thread=async_save] Background save thread done. Time taken: 51.402071s. I0419 04:55:20.655636 135321391208192 async_checkpointer.py:273] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 04:55:20.655739 135321391208192 async_checkpointer.py:283] [process=0][thread=save_finalize] No errors found in background save thread=async_save. I0419 04:55:20.655872 135321391208192 checkpoint_manager.py:2133] [process=0][thread=save_finalize][step=10] CheckpointManager Save Finalize is syncing with other hosts... I0419 04:55:20.657459 135321391208192 checkpoint_manager.py:2142] [process=0][thread=save_finalize][step=10] CheckpointManager Save Finalize is done on all hosts. I0419 04:55:20.657711 135447319566144 checkpoint_manager.py:2036] [process=0][thread=MainThread][step=10][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=10. W0419 04:55:20.657898 135447319566144 checkpoint_manager.py:1458] Waiting for previous save to complete took 49.958933 seconds. If this number is high, consider checkpointing less frequently. I0419 04:55:20.659743 135447319566144 checkpoint_manager.py:1518] [process=0] Saving checkpoint at step 15 I0419 04:55:20.663851 135447319566144 async_checkpointer.py:452] [process=0] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/15. I0419 04:55:22.066863 135447319566144 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 04:55:22.067001 135447319566144 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0419 04:55:22.115146 135447319566144 base_pytree_checkpoint_handler.py:153] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.049669s I0419 04:55:22.118703 135447319566144 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.108 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.30199766159057617 s) (per-host) I0419 04:55:22.118836 135447319566144 base_pytree_checkpoint_handler.py:737] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.302162s (batch_requests_ready=0.244888s, total_serialization_initiated=0.053812s, others=0.003462s) I0419 04:55:22.118982 135447319566144 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.306830s (all_items=0.000023s, per_item={'items': '0.00002265'}, temp_paths=0.306808) I0419 04:55:22.120301 135321391208192 async_checkpointer.py:79] [process=0][thread=async_save] Background save thread started. I0419 04:55:22.120519 135447319566144 async_checkpointer.py:561] Finished blocking save. Time taken: 1.460657s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/15. I0419 04:55:22.486863 135323505895168 atomicity.py:137] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/15 I0419 04:55:22.556488 135447319566144 checkpoint_manager.py:1566] [process=0][thread=MainThread][step=15] Starting CheckpointManager Save Finalize thread=save_finalize I0419 04:55:22.556989 135323514287872 async_checkpointer.py:265] [process=0][thread=save_finalize] Waiting for background save thread=async_save. I0419 04:55:22.557241 135447319566144 standard_logger.py:34] {'step': 15, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776574470.6989303, 'wait_for_prev_duration_secs': 49.95893335342407, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776574520.6598213, 'checkpointer_blocking_duration_secs': 1.4608535766601562, 'get_old_steps_start_time': 1776574522.1207001, 'get_old_steps_duration_secs': 3.838539123535156e-05, 'checkpoint_manager_blocking_start_time': 1776574470.6965902, 'checkpoint_manager_blocking_duration_secs': 51.86059594154358} I0419 04:55:22.557388 135447319566144 checkpointing.py:409] Started an asynchronous checkpoint save for step 15 I0419 04:55:41.117401 135447319566144 metric_logger.py:185] completed step: 14, seconds: 0.147, TFLOP/s/device: 92.192, Tokens/s/device: 13896.145, total_weights: 65536, loss: 5.864 I0419 04:55:41.126085 135447319566144 metric_logger.py:185] completed step: 15, seconds: 0.147, TFLOP/s/device: 92.205, Tokens/s/device: 13898.126, total_weights: 65536, loss: 5.719 I0419 04:55:41.272126 135447319566144 metric_logger.py:185] completed step: 16, seconds: 70.715, TFLOP/s/device: 0.192, Tokens/s/device: 28.961, total_weights: 65536, loss: 5.605 I0419 04:55:41.419377 135447319566144 metric_logger.py:185] completed step: 17, seconds: 0.007, TFLOP/s/device: 2001.050, Tokens/s/device: 301620.029, total_weights: 65536, loss: 5.517 I0419 04:55:41.567186 135447319566144 metric_logger.py:185] completed step: 18, seconds: 0.147, TFLOP/s/device: 92.378, Tokens/s/device: 13924.206, total_weights: 65536, loss: 5.448 I0419 04:55:41.714165 135447319566144 checkpointing.py:794] Waiting for step 19 to finish before checkpoint... I0419 04:55:41.714827 135447319566144 checkpointing.py:798] Waited 0.0006654262542724609 seconds for step 19 to finish before starting checkpointing. I0419 04:55:41.716761 135447319566144 checkpoint_manager.py:2024] [process=0][thread=MainThread][step=15][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 04:55:42.709317 135323355768576 atomicity.py:137] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/15/items I0419 04:55:45.933741 135325116507904 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/15/items/array_metadatas/process_0 I0419 04:56:27.088756 135325108115200 base_pytree_checkpoint_handler.py:1232] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 42.829082s (commit=40.880412s, array_metadata_write=1.948670s) I0419 04:56:27.089787 135321391208192 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/gbytes_per_sec: 24.200 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 65.27309799194336 s) (per-host) I0419 04:56:27.089884 135321391208192 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 64.969425s. I0419 04:56:29.357005 135321391208192 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000273s. I0419 04:56:29.767999 135321391208192 base_pytree_checkpoint_handler.py:1356] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 1.443830s. use_zarr3=False, enable_post_merge_validation=True, directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/15/items I0419 04:56:29.770105 135321391208192 atomicity.py:608] Finalizing gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/15/items I0419 04:56:31.070652 135321391208192 atomicity.py:608] Finalizing gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/15 I0419 04:56:32.833545 135321391208192 atomicity.py:790] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/15`. I0419 04:56:32.834217 135321391208192 async_checkpointer.py:420] Finished async_save (blocking + background). Time taken: 72.174379s. directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/15 I0419 04:56:32.835583 135321391208192 async_checkpointer.py:144] [process=0][thread=async_save] Background save thread done. Time taken: 70.715118s. I0419 04:56:32.835777 135323514287872 async_checkpointer.py:273] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 04:56:32.835926 135323514287872 async_checkpointer.py:283] [process=0][thread=save_finalize] No errors found in background save thread=async_save. I0419 04:56:32.835994 135323514287872 checkpoint_manager.py:2133] [process=0][thread=save_finalize][step=15] CheckpointManager Save Finalize is syncing with other hosts... I0419 04:56:32.837947 135323514287872 checkpoint_manager.py:2142] [process=0][thread=save_finalize][step=15] CheckpointManager Save Finalize is done on all hosts. I0419 04:56:32.838156 135447319566144 checkpoint_manager.py:2036] [process=0][thread=MainThread][step=15][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=15. W0419 04:56:32.838290 135447319566144 checkpoint_manager.py:1458] Waiting for previous save to complete took 51.121542 seconds. If this number is high, consider checkpointing less frequently. I0419 04:56:32.839710 135447319566144 checkpoint_manager.py:1518] [process=0] Saving checkpoint at step 19 I0419 04:56:32.843227 135447319566144 async_checkpointer.py:452] [process=0] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/19. I0419 04:56:33.787292 135321391208192 atomicity.py:137] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/19 I0419 04:56:33.788024 135447319566144 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 04:56:33.788134 135447319566144 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0419 04:56:33.813848 135447319566144 base_pytree_checkpoint_handler.py:153] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.027274s I0419 04:56:33.814675 135447319566144 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.712 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2700634002685547 s) (per-host) I0419 04:56:33.814742 135447319566144 base_pytree_checkpoint_handler.py:737] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.270149s (batch_requests_ready=0.239954s, total_serialization_initiated=0.029437s, others=0.000757s) I0419 04:56:33.814840 135447319566144 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.274046s (all_items=0.000017s, per_item={'items': '0.00001693'}, temp_paths=0.274029) I0419 04:56:33.815671 135323505895168 async_checkpointer.py:79] [process=0][thread=async_save] Background save thread started. I0419 04:56:33.815829 135447319566144 async_checkpointer.py:561] Finished blocking save. Time taken: 0.976033s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/19. I0419 04:56:34.700878 135447319566144 checkpoint_manager.py:1566] [process=0][thread=MainThread][step=19] Starting CheckpointManager Save Finalize thread=save_finalize I0419 04:56:34.701250 135323514287872 async_checkpointer.py:265] [process=0][thread=save_finalize] Waiting for background save thread=async_save. I0419 04:56:34.701423 135447319566144 standard_logger.py:34] {'step': 19, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776574541.7167313, 'wait_for_prev_duration_secs': 51.12154173851013, 'time_between_consecutive_saves_sec': 21.059204816818237, 'checkpointer_blocking_start_time': 1776574592.8397481, 'checkpointer_blocking_duration_secs': 0.9761910438537598, 'get_old_steps_start_time': 1776574593.8159544, 'get_old_steps_duration_secs': 2.5033950805664062e-05, 'checkpoint_manager_blocking_start_time': 1776574541.7150533, 'checkpoint_manager_blocking_duration_secs': 52.986334562301636} I0419 04:56:34.701536 135447319566144 checkpointing.py:409] Started an asynchronous checkpoint save for step 19 I0419 04:56:34.701579 135447319566144 checkpoint_manager.py:2024] [process=0][thread=MainThread][step=19][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 04:56:35.507341 135325116507904 atomicity.py:137] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/19/items I0419 04:56:38.894909 135325108115200 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/19/items/array_metadatas/process_0 I0419 04:57:17.432846 135324596422400 base_pytree_checkpoint_handler.py:1232] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 40.837664s (commit=39.271892s, array_metadata_write=1.565772s) I0419 04:57:17.433934 135323505895168 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/gbytes_per_sec: 35.990 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.88929200172424 s) (per-host) I0419 04:57:17.434046 135323505895168 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.618277s. I0419 04:57:19.362459 135323505895168 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000214s. I0419 04:57:19.820883 135323505895168 base_pytree_checkpoint_handler.py:1356] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 1.478898s. use_zarr3=False, enable_post_merge_validation=True, directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/19/items I0419 04:57:19.827317 135323505895168 atomicity.py:608] Finalizing gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/19/items I0419 04:57:20.247862 135323505895168 atomicity.py:608] Finalizing gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/19 I0419 04:57:21.932525 135323505895168 atomicity.py:790] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/19`. I0419 04:57:21.933235 135323505895168 async_checkpointer.py:420] Finished async_save (blocking + background). Time taken: 49.093474s. directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_16_zarr3_false/checkpoints/19 I0419 04:57:21.934546 135323505895168 async_checkpointer.py:144] [process=0][thread=async_save] Background save thread done. Time taken: 48.118776s. I0419 04:57:21.934741 135323514287872 async_checkpointer.py:273] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 04:57:21.934881 135323514287872 async_checkpointer.py:283] [process=0][thread=save_finalize] No errors found in background save thread=async_save. I0419 04:57:21.934935 135323514287872 checkpoint_manager.py:2133] [process=0][thread=save_finalize][step=19] CheckpointManager Save Finalize is syncing with other hosts... I0419 04:57:21.936587 135323514287872 checkpoint_manager.py:2142] [process=0][thread=save_finalize][step=19] CheckpointManager Save Finalize is done on all hosts. I0419 04:57:21.936749 135447319566144 checkpoint_manager.py:2036] [process=0][thread=MainThread][step=19][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=19. I0419 04:57:21.936901 135447319566144 checkpoint_manager.py:2013] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 04:57:21.937517 135447319566144 metric_logger.py:185] completed step: 19, seconds: 0.147, TFLOP/s/device: 92.242, Tokens/s/device: 13903.692, total_weights: 65536, loss: 5.391 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Sun Apr 19 04:57:33 UTC 2026 EXIT_CODE=0