XPK Start: Sun Apr 19 04:29:13 UTC 2026 2026-04-19 04:29:17.140380: 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 00:00:1776572957.153105 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776572957.156856 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776572957.168109 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776572957.168126 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776572957.168128 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776572957.168130 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-19 04:29:36.360938: 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 04:29:36.880279 139515374733120 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-19 04:29:45,921:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0419 04:29:45.921483 139515374733120 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-19 04:29:45,923:jax._src.distributed:157: Connecting to JAX distributed service on mt-15-ocdbt-false-sv-vf62j-slice-job-0-0.mt-15-ocdbt-false-sv-vf62j:8482 I0419 04:29:45.923821 139515374733120 distributed.py:157] Connecting to JAX distributed service on mt-15-ocdbt-false-sv-vf62j-slice-job-0-0.mt-15-ocdbt-false-sv-vf62j:8482 I0419 04:29:47.997142 139515374733120 max_utils.py:284] Jax distributed system initialized! I0419 04:29:54.740306 139515374733120 max_utils.py:800] System Information: Jax Version: 0.8.1 I0419 04:29:54.740411 139515374733120 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0419 04:29:54.740452 139515374733120 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 04:29:54.740488 139515374733120 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0419 04:29:55.354502 139515374733120 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 04:29:55.355114 139515374733120 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 04:29:55.355302 139515374733120 checkpointing.py:688] Setting up checkpoint logger... I0419 04:29:55.355354 139515374733120 checkpointing.py:234] Creating checkpoint manager with ocdbt=False and zarr3=True I0419 04:29:55.355398 139515374733120 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0419 04:29:55.355738 139515374733120 base_pytree_checkpoint_handler.py:415] Created BasePyTreeCheckpointHandler: use_ocdbt=False, use_zarr3=True, pytree_metadata_options=PyTreeMetadataOptions(support_rich_types=False), array_metadata_store=<orbax.checkpoint._src.metadata.array_metadata_store.Store object at 0x7ee2f2775280>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0419 04:29:59.055627 139515374733120 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0419 04:29:59.055881 139515374733120 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 0x7ee1a4782600>}, handler_registry=None I0419 04:29:59.056117 139515374733120 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ee1a4782600>` 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 04:29:59.056166 139515374733120 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ecfda05b200>` 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 04:29:59.056206 139515374733120 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 0x7ee1a4782600>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ee1a4782600>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ecfda05b200>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ecfda05b200>}). I0419 04:29:59.056713 139515374733120 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0419 04:29:59.056792 139515374733120 async_checkpointer.py:177] [process=0][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7ecf0bf98180> timeout: 600 secs and primary_host=0 for async checkpoint writes I0419 04:30:00.418020 139515374733120 checkpoint_manager.py:1818] Found 0 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_15_ocdbt_false/checkpoints I0419 04:30:01.283266 139515374733120 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_15_ocdbt_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7ecfd8c7ee70> I0419 04:30:01.283439 139515374733120 checkpointing.py:302] Checkpoint manager created! I0419 04:30:02.157083 139515374733120 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0419 04:30:02.157187 139515374733120 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 04:30:02.539914 139515374733120 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 04:30:02.540013 139515374733120 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 04:30:02.555825 139515374733120 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 04:30:02.555886 139515374733120 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 04:30:02.578501 139515374733120 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 04:30:02.578569 139515374733120 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:30:02.594357 139515374733120 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 04:30:02.594422 139515374733120 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:30:02.610330 139515374733120 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 04:30:02.610397 139515374733120 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:30:02.635039 139515374733120 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0419 04:30:02.635108 139515374733120 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:30:02.654615 139515374733120 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0419 04:30:02.654684 139515374733120 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0419 04:30:02.849043 139515374733120 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0419 04:30:02.849157 139515374733120 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0419 04:30:04.270354 139515374733120 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0419 04:30:04.270494 139515374733120 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 04:30:04.270551 139515374733120 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 04:30:04.270612 139515374733120 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 04:30:04.270665 139515374733120 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 04:30:04.270705 139515374733120 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 04:30:04.270758 139515374733120 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 04:30:04.270837 139515374733120 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 04:30:04.270884 139515374733120 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 04:30:04.270924 139515374733120 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 04:30:04.270976 139515374733120 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0419 04:30:04.271028 139515374733120 maxtext_utils.py:1740] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0419 04:30:04.768703 139515374733120 train.py:158] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 04:30:04.768796 139515374733120 train.py:158] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0419 04:30:04.783604 139515374733120 train.py:165] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 04:30:04.783666 139515374733120 train.py:165] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0419 04:30:15.360528 139515374733120 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 04:30:15.366652 139515374733120 metric_logger.py:289] number parameters: 1.104 billion I0419 04:30:26.967602 139515374733120 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0419 04:30:27.121993 139515374733120 checkpointing.py:798] Waited 0.1543750762939453 seconds for step 0 to finish before starting checkpointing. I0419 04:30:27.124539 139515374733120 checkpoint_manager.py:2013] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 04:30:27.126423 139515374733120 checkpoint_manager.py:1518] [process=0] Saving checkpoint at step 0 I0419 04:30:27.129231 139515374733120 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_15_ocdbt_false/checkpoints/0. I0419 04:30:27.843502 139515374733120 signaling_client.py:364] Using JaxDistributedSignalingClient I0419 04:30:28.090334 139387945850624 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_15_ocdbt_false/checkpoints/0 I0419 04:30:28.531117 139515374733120 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 04:30:28.531218 139515374733120 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 04:30:28.773749 139515374733120 base_pytree_checkpoint_handler.py:153] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.243850s I0419 04:30:28.776146 139515374733120 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.670 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.9235568046569824 s) (per-host) I0419 04:30:28.776232 139515374733120 base_pytree_checkpoint_handler.py:737] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.923663s (batch_requests_ready=0.673603s, total_serialization_initiated=0.247750s, others=0.002311s) I0419 04:30:28.776315 139515374733120 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.927633s (all_items=0.000018s, per_item={'items': '0.00001812'}, temp_paths=0.927615) I0419 04:30:28.777258 139389027985152 async_checkpointer.py:79] [process=0][thread=async_save] Background save thread started. I0419 04:30:28.777420 139515374733120 async_checkpointer.py:561] Finished blocking save. Time taken: 1.650927s. 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_15_ocdbt_false/checkpoints/0. I0419 04:30:28.833698 139515374733120 checkpoint_manager.py:1566] [process=0][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0419 04:30:28.834047 139387459335936 async_checkpointer.py:265] [process=0][thread=save_finalize] Waiting for background save thread=async_save. I0419 04:30:28.834200 139515374733120 standard_logger.py:34] {'step': 0, '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_15_ocdbt_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776573027.124521, 'wait_for_prev_duration_secs': 6.461143493652344e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776573027.1264625, 'checkpointer_blocking_duration_secs': 1.6510624885559082, 'get_old_steps_start_time': 1776573028.7775452, 'get_old_steps_duration_secs': 2.6226043701171875e-05, 'checkpoint_manager_blocking_start_time': 1776573027.1226032, 'checkpoint_manager_blocking_duration_secs': 1.7115576267242432} I0419 04:30:28.834307 139515374733120 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0419 04:30:28.834365 139515374733120 max_utils.py:750] Memstats: After params initialized: I0419 04:30:28.834420 139515374733120 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_0(process=0,(0,0,0,0)) I0419 04:30:28.834454 139515374733120 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_1(process=0,(1,0,0,0)) I0419 04:30:28.834481 139515374733120 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_4(process=0,(0,1,0,0)) I0419 04:30:28.834505 139515374733120 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_5(process=0,(1,1,0,0)) I0419 04:30:29.141167 139515374733120 metric_logger.py:185] completed step: 0, seconds: 11.255, TFLOP/s/device: 1.207, Tokens/s/device: 181.969, total_weights: 65536, loss: 10.874 I0419 04:30:29.143977 139515374733120 metric_logger.py:269] To see full metrics 'tensorboard --logdir=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_15_ocdbt_false/tensorboard/' I0419 04:30:29.555433 139515374733120 metric_logger.py:185] completed step: 1, seconds: 2.172, TFLOP/s/device: 6.256, Tokens/s/device: 942.945, total_weights: 65536, loss: 10.874 I0419 04:30:29.701478 139515374733120 metric_logger.py:185] completed step: 2, seconds: 0.415, TFLOP/s/device: 32.741, Tokens/s/device: 4935.071, total_weights: 65536, loss: 10.560 I0419 04:30:29.849061 139515374733120 metric_logger.py:185] completed step: 3, seconds: 0.008, TFLOP/s/device: 1605.854, Tokens/s/device: 242051.767, total_weights: 65536, loss: 9.980 I0419 04:30:29.858070 139515374733120 checkpointing.py:794] Waiting for step 5 to finish before checkpoint... I0419 04:30:30.143642 139515374733120 checkpointing.py:798] Waited 0.28545498847961426 seconds for step 5 to finish before starting checkpointing. I0419 04:30:30.146568 139515374733120 checkpoint_manager.py:2024] [process=0][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 04:30:31.163414 139389061555968 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_15_ocdbt_false/checkpoints/0/items I0419 04:30:31.347836 139388491990784 checkpoint.py:188] Wrote Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776573030916258378, 'commit_timestamp_nsecs': None, 'custom_metadata': {}}, json={"item_handlers": null, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1776573030916258378, "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_15_ocdbt_false/checkpoints/0/_CHECKPOINT_METADATA I0419 04:30:33.133829 2620 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0419 04:30:35.956881 139392265996032 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_15_ocdbt_false/checkpoints/0/items/array_metadatas/process_0 I0419 04:30:43.405851 139392257603328 base_pytree_checkpoint_handler.py:1232] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 10.280358s (commit=9.327860s, array_metadata_write=0.952498s) I0419 04:30:43.407292 139389027985152 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/gbytes_per_sec: 101.551 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 15.55466365814209 s) (per-host) I0419 04:30:43.407462 139389027985152 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 14.630087s. I0419 04:30:44.068325 139389027985152 checkpoint.py:228] Read Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776573030916258378, 'commit_timestamp_nsecs': None, '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_15_ocdbt_false/checkpoints/0/_CHECKPOINT_METADATA I0419 04:30:45.221611 139388491990784 checkpoint.py:247] Updated Metadata={'item_handlers': {'items': 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler'}, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776573030916258378, '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_15_ocdbt_false/checkpoints/0/_CHECKPOINT_METADATA W0419 04:30:45.283596 139389027985152 ocdbt_utils.py:168] [process=0][thread=async_save] Skipping merge of OCDBT checkpoints: No per-process OCDBT checkpoint subdirs found in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_15_ocdbt_false/checkpoints/0/items, I0419 04:30:46.158737 139389027985152 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000393s. I0419 04:30:46.159000 139389027985152 base_pytree_checkpoint_handler.py:1356] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 1.525310s. use_zarr3=True, 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_15_ocdbt_false/checkpoints/0/items I0419 04:30:46.160853 139389027985152 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_15_ocdbt_false/checkpoints/0/items I0419 04:30:46.568151 139389027985152 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_15_ocdbt_false/checkpoints/0 I0419 04:30:47.807549 139389027985152 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_15_ocdbt_false/checkpoints/0`. I0419 04:30:47.808652 139389027985152 async_checkpointer.py:420] Finished async_save (blocking + background). Time taken: 20.682158s. directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_15_ocdbt_false/checkpoints/0 I0419 04:30:47.810107 139389027985152 async_checkpointer.py:144] [process=0][thread=async_save] Background save thread done. Time taken: 19.032726s. I0419 04:30:47.810327 139387459335936 async_checkpointer.py:273] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 04:30:47.810442 139387459335936 async_checkpointer.py:283] [process=0][thread=save_finalize] No errors found in background save thread=async_save. I0419 04:30:47.810547 139387459335936 checkpoint_manager.py:2133] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0419 04:30:47.812454 139387459335936 checkpoint_manager.py:2142] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0419 04:30:47.812697 139515374733120 checkpoint_manager.py:2036] [process=0][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0419 04:30:47.812819 139515374733120 checkpoint_manager.py:1458] Waiting for previous save to complete took 17.666272 seconds. If this number is high, consider checkpointing less frequently. I0419 04:30:47.814878 139515374733120 checkpoint_manager.py:1518] [process=0] Saving checkpoint at step 5 I0419 04:30:47.819024 139515374733120 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_15_ocdbt_false/checkpoints/5. I0419 04:30:48.768939 139389027985152 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_15_ocdbt_false/checkpoints/5 I0419 04:30:48.784821 139515374733120 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 04:30:48.784965 139515374733120 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 04:30:48.827518 139515374733120 base_pytree_checkpoint_handler.py:153] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.043982s I0419 04:30:48.828583 139515374733120 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.183 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.29763245582580566 s) (per-host) I0419 04:30:48.828695 139515374733120 base_pytree_checkpoint_handler.py:737] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.297771s (batch_requests_ready=0.248729s, total_serialization_initiated=0.048079s, others=0.000963s) I0419 04:30:48.828841 139515374733120 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.302733s (all_items=0.000025s, per_item={'items': '0.00002503'}, temp_paths=0.302708) I0419 04:30:48.829953 139383139190528 async_checkpointer.py:79] [process=0][thread=async_save] Background save thread started. I0419 04:30:48.830159 139515374733120 async_checkpointer.py:561] Finished blocking save. Time taken: 1.015161s. 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_15_ocdbt_false/checkpoints/5. I0419 04:30:49.706575 139515374733120 checkpoint_manager.py:1566] [process=0][thread=MainThread][step=5] Starting CheckpointManager Save Finalize thread=save_finalize I0419 04:30:49.707086 139387459335936 async_checkpointer.py:265] [process=0][thread=save_finalize] Waiting for background save thread=async_save. I0419 04:30:49.707340 139515374733120 standard_logger.py:34] {'step': 5, '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_15_ocdbt_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776573030.1465154, 'wait_for_prev_duration_secs': 17.666271686553955, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776573047.8149633, 'checkpointer_blocking_duration_secs': 1.015348196029663, 'get_old_steps_start_time': 1776573048.8303342, 'get_old_steps_duration_secs': 4.506111145019531e-05, 'checkpoint_manager_blocking_start_time': 1776573030.1442635, 'checkpoint_manager_blocking_duration_secs': 19.56302309036255} I0419 04:30:49.707494 139515374733120 checkpointing.py:409] Started an asynchronous checkpoint save for step 5 I0419 04:31:10.021277 139515374733120 metric_logger.py:185] completed step: 4, seconds: 0.147, TFLOP/s/device: 92.338, Tokens/s/device: 13918.149, total_weights: 65536, loss: 9.460 I0419 04:31:10.029587 139515374733120 metric_logger.py:185] completed step: 5, seconds: 0.148, TFLOP/s/device: 91.908, Tokens/s/device: 13853.376, total_weights: 65536, loss: 8.959 I0419 04:31:10.175211 139515374733120 metric_logger.py:185] completed step: 6, seconds: 40.170, TFLOP/s/device: 0.338, Tokens/s/device: 50.983, total_weights: 65536, loss: 8.469 I0419 04:31:10.322827 139515374733120 metric_logger.py:185] completed step: 7, seconds: 0.006, TFLOP/s/device: 2100.669, Tokens/s/device: 316635.745, total_weights: 65536, loss: 8.003 I0419 04:31:10.470264 139515374733120 metric_logger.py:185] completed step: 8, seconds: 0.147, TFLOP/s/device: 92.142, Tokens/s/device: 13888.701, total_weights: 65536, loss: 7.572 I0419 04:31:10.617556 139515374733120 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0419 04:31:10.618247 139515374733120 checkpointing.py:798] Waited 0.0007078647613525391 seconds for step 9 to finish before starting checkpointing. I0419 04:31:10.620308 139515374733120 checkpoint_manager.py:2024] [process=0][thread=MainThread][step=5][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 04:31:11.002118 139383650883328 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_15_ocdbt_false/checkpoints/5/items I0419 04:31:14.856845 139383164368640 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_15_ocdbt_false/checkpoints/5/items/array_metadatas/process_0 I0419 04:31:22.631854 139383155975936 base_pytree_checkpoint_handler.py:1232] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 10.457677s (commit=9.493181s, array_metadata_write=0.964496s) I0419 04:31:22.632941 139383139190528 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/gbytes_per_sec: 46.320 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 34.10196661949158 s) (per-host) I0419 04:31:22.633064 139383139190528 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 33.802961s. W0419 04:31:24.023961 139383139190528 ocdbt_utils.py:168] [process=0][thread=async_save] Skipping merge of OCDBT checkpoints: No per-process OCDBT checkpoint subdirs found in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_15_ocdbt_false/checkpoints/5/items, I0419 04:31:24.886641 139383139190528 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000243s. I0419 04:31:24.886844 139383139190528 base_pytree_checkpoint_handler.py:1356] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 1.482253s. use_zarr3=True, 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_15_ocdbt_false/checkpoints/5/items I0419 04:31:24.888438 139383139190528 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_15_ocdbt_false/checkpoints/5/items I0419 04:31:25.277491 139383139190528 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_15_ocdbt_false/checkpoints/5 I0419 04:31:26.522652 139383139190528 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_15_ocdbt_false/checkpoints/5`. I0419 04:31:26.523473 139383139190528 async_checkpointer.py:420] Finished async_save (blocking + background). Time taken: 38.708481s. directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_15_ocdbt_false/checkpoints/5 I0419 04:31:26.524647 139383139190528 async_checkpointer.py:144] [process=0][thread=async_save] Background save thread done. Time taken: 37.694543s. I0419 04:31:26.524858 139387459335936 async_checkpointer.py:273] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 04:31:26.524973 139387459335936 async_checkpointer.py:283] [process=0][thread=save_finalize] No errors found in background save thread=async_save. I0419 04:31:26.525039 139387459335936 checkpoint_manager.py:2133] [process=0][thread=save_finalize][step=5] CheckpointManager Save Finalize is syncing with other hosts... I0419 04:31:26.528044 139387459335936 checkpoint_manager.py:2142] [process=0][thread=save_finalize][step=5] CheckpointManager Save Finalize is done on all hosts. I0419 04:31:26.528192 139515374733120 checkpoint_manager.py:2036] [process=0][thread=MainThread][step=5][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=5. W0419 04:31:26.528264 139515374733120 checkpoint_manager.py:1458] Waiting for previous save to complete took 15.907977 seconds. If this number is high, consider checkpointing less frequently. I0419 04:31:26.529926 139515374733120 checkpoint_manager.py:1518] [process=0] Saving checkpoint at step 9 I0419 04:31:26.533263 139515374733120 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_15_ocdbt_false/checkpoints/9. I0419 04:31:27.469981 139381545363200 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_15_ocdbt_false/checkpoints/9 I0419 04:31:27.487773 139515374733120 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 04:31:27.487886 139515374733120 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 04:31:27.514620 139515374733120 base_pytree_checkpoint_handler.py:153] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.028020s I0419 04:31:27.515449 139515374733120 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.511 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2798910140991211 s) (per-host) I0419 04:31:27.515518 139515374733120 base_pytree_checkpoint_handler.py:737] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.279979s (batch_requests_ready=0.248990s, total_serialization_initiated=0.030210s, others=0.000779s) I0419 04:31:27.515593 139515374733120 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.283884s (all_items=0.000017s, per_item={'items': '0.00001717'}, temp_paths=0.283867) I0419 04:31:27.516454 139384774981376 async_checkpointer.py:79] [process=0][thread=async_save] Background save thread started. I0419 04:31:27.516556 139515374733120 async_checkpointer.py:561] Finished blocking save. Time taken: 0.986557s. 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_15_ocdbt_false/checkpoints/9. I0419 04:31:27.987075 139515374733120 checkpoint_manager.py:1566] [process=0][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0419 04:31:27.987432 139387459335936 async_checkpointer.py:265] [process=0][thread=save_finalize] Waiting for background save thread=async_save. I0419 04:31:27.987606 139515374733120 standard_logger.py:34] {'step': 9, '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_15_ocdbt_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776573070.6202753, 'wait_for_prev_duration_secs': 15.907976627349854, 'time_between_consecutive_saves_sec': 22.80775475502014, 'checkpointer_blocking_start_time': 1776573086.5299687, 'checkpointer_blocking_duration_secs': 0.9866974353790283, 'get_old_steps_start_time': 1776573087.516681, 'get_old_steps_duration_secs': 2.6226043701171875e-05, 'checkpoint_manager_blocking_start_time': 1776573070.618511, 'checkpoint_manager_blocking_duration_secs': 17.369060516357422} I0419 04:31:27.987712 139515374733120 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0419 04:31:27.987756 139515374733120 checkpoint_manager.py:2024] [process=0][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 04:31:30.045613 139385269847808 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_15_ocdbt_false/checkpoints/9/items I0419 04:31:32.772382 139385278240512 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_15_ocdbt_false/checkpoints/9/items/array_metadatas/process_0 I0419 04:31:41.351116 139385261455104 base_pytree_checkpoint_handler.py:1232] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 10.193512s (commit=9.147323s, array_metadata_write=1.046189s) I0419 04:31:41.352222 139384774981376 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/gbytes_per_sec: 111.896 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 14.116629123687744 s) (per-host) I0419 04:31:41.352343 139384774981376 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 13.835827s. W0419 04:31:43.279341 139384774981376 ocdbt_utils.py:168] [process=0][thread=async_save] Skipping merge of OCDBT checkpoints: No per-process OCDBT checkpoint subdirs found in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_15_ocdbt_false/checkpoints/9/items, I0419 04:31:44.117093 139384774981376 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000209s. I0419 04:31:44.117272 139384774981376 base_pytree_checkpoint_handler.py:1356] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 1.489229s. use_zarr3=True, 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_15_ocdbt_false/checkpoints/9/items I0419 04:31:44.119699 139384774981376 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_15_ocdbt_false/checkpoints/9/items I0419 04:31:44.531855 139384774981376 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_15_ocdbt_false/checkpoints/9 I0419 04:31:45.814203 139384774981376 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_15_ocdbt_false/checkpoints/9`. I0419 04:31:45.815025 139384774981376 async_checkpointer.py:420] Finished async_save (blocking + background). Time taken: 19.285031s. directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_15_ocdbt_false/checkpoints/9 I0419 04:31:45.816359 139384774981376 async_checkpointer.py:144] [process=0][thread=async_save] Background save thread done. Time taken: 18.299842s. I0419 04:31:45.816560 139387459335936 async_checkpointer.py:273] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 04:31:45.816681 139387459335936 async_checkpointer.py:283] [process=0][thread=save_finalize] No errors found in background save thread=async_save. I0419 04:31:45.816735 139387459335936 checkpoint_manager.py:2133] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0419 04:31:45.818922 139387459335936 checkpoint_manager.py:2142] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0419 04:31:45.819082 139515374733120 checkpoint_manager.py:2036] [process=0][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0419 04:31:45.819228 139515374733120 checkpoint_manager.py:2013] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 04:31:45.819835 139515374733120 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.437, Tokens/s/device: 13933.110, total_weights: 65536, loss: 7.181 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Sun Apr 19 04:31:58 UTC 2026 EXIT_CODE=0