XPK Start: Sun Apr 19 09:28:26 UTC 2026 2026-04-19 09:28:30.278036: 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 00:00:1776590910.291949 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776590910.295746 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776590910.307347 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776590910.307368 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776590910.307371 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776590910.307373 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-19 09:28:49.333000: 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 09:28:49.850825 132768589182784 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-19 09:28:58,892:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0419 09:28:58.892251 132768589182784 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-19 09:28:58,894:jax._src.distributed:157: Connecting to JAX distributed service on mt-09-pdb-lt-1-kp25i-slice-job-0-0.mt-09-pdb-lt-1-kp25i:8482 I0419 09:28:58.894773 132768589182784 distributed.py:157] Connecting to JAX distributed service on mt-09-pdb-lt-1-kp25i-slice-job-0-0.mt-09-pdb-lt-1-kp25i:8482 I0419 09:29:00.356144 132768589182784 max_utils.py:284] Jax distributed system initialized! I0419 09:29:07.412309 132768589182784 max_utils.py:800] System Information: Jax Version: 0.8.1 I0419 09:29:07.412412 132768589182784 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0419 09:29:07.412453 132768589182784 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 09:29:07.412487 132768589182784 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0419 09:29:08.055097 132768589182784 maxtext_utils.py:1398] Num_devices: 32, shape (1, 1, 1, 8, 1, 1, 1, 1, 4, 1, 1, 1, 1) I0419 09:29:08.055386 132768589182784 checkpointing.py:677] Setting up checkpoint logger... I0419 09:29:08.055443 132768589182784 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0419 09:29:08.055489 132768589182784 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0419 09:29:08.055830 132768589182784 base_pytree_checkpoint_handler.py:415] 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 0x78c00f378ec0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0419 09:29:10.876352 132768589182784 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0419 09:29:10.876699 132768589182784 checkpoint_manager.py:709] [process=4][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78acf2715dc0>}, handler_registry=None I0419 09:29:10.876949 132768589182784 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78acf2715dc0>` 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 09:29:10.876999 132768589182784 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78acf1c478f0>` 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 09:29:10.877043 132768589182784 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 0x78acf2715dc0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78acf2715dc0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78acf1c478f0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78acf1c478f0>}). I0419 09:29:10.879694 132768589182784 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0419 09:29:10.879779 132768589182784 async_checkpointer.py:177] [process=4][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x78bff7ffb2e0> timeout: 600 secs and primary_host=0 for async checkpoint writes I0419 09:29:12.132756 132768589182784 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_09_pdb_lt_1/checkpoints I0419 09:29:12.246526 132768589182784 checkpoint_manager.py:929] [process=4][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/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_09_pdb_lt_1/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x78acf271a6c0> I0419 09:29:12.246699 132768589182784 checkpointing.py:301] Checkpoint manager created! I0419 09:29:13.128988 132768589182784 nnx_wrappers.py:453] Unknown Logical: bfloat16[8,2048,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0419 09:29:13.129096 132768589182784 nnx_wrappers.py:453] Unknown Physical: bfloat16[8,2048,2048]....................................... ('fsdp', None, 'tensor'). I0419 09:29:13.495903 132768589182784 attentions.py:1084] attentions/inputs_q Logical: bfloat16[8,2048,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 09:29:13.496018 132768589182784 attentions.py:1084] attentions/inputs_q Physical: bfloat16[8,2048,2048]....................................... ('fsdp', None, 'tensor'). I0419 09:29:13.511671 132768589182784 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[8,2048,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 09:29:13.511730 132768589182784 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[8,2048,2048]....................................... ('fsdp', None, 'tensor'). I0419 09:29:13.534908 132768589182784 attentions.py:1150] attentions/query Logical: bfloat16[8,2048,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 09:29:13.534988 132768589182784 attentions.py:1150] attentions/query Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0419 09:29:13.550856 132768589182784 attentions.py:1151] attentions/key Logical: bfloat16[8,2048,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 09:29:13.550921 132768589182784 attentions.py:1151] attentions/key Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0419 09:29:13.567087 132768589182784 attentions.py:1152] attentions/value Logical: bfloat16[8,2048,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 09:29:13.567166 132768589182784 attentions.py:1152] attentions/value Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0419 09:29:13.591304 132768589182784 attentions.py:1193] attentions/out Logical: bfloat16[8,2048,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0419 09:29:13.591379 132768589182784 attentions.py:1193] attentions/out Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0419 09:29:13.610917 132768589182784 linears.py:525] linears/x Logical: bfloat16[8,2048,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0419 09:29:13.610993 132768589182784 linears.py:525] linears/x Physical: bfloat16[8,2048,7168]....................................... ('fsdp', None, 'tensor'). I0419 09:29:13.799375 132768589182784 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0419 09:29:13.799484 132768589182784 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 8 tensor: 4 I0419 09:29:15.219687 132768589182784 maxtext_utils.py:1501] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: ('tensor',) I0419 09:29:15.219843 132768589182784 maxtext_utils.py:1501] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, 'tensor') I0419 09:29:15.219905 132768589182784 maxtext_utils.py:1501] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, 'tensor') I0419 09:29:15.219984 132768589182784 maxtext_utils.py:1501] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: PartitionSpec('mlp', 'layers', 'embed') Physical: ('tensor', None, 'fsdp') I0419 09:29:15.220042 132768589182784 maxtext_utils.py:1501] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: ('tensor', None) I0419 09:29:15.220083 132768589182784 maxtext_utils.py:1501] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: ('tensor', None) I0419 09:29:15.220136 132768589182784 maxtext_utils.py:1501] 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, 'tensor', None) I0419 09:29:15.220189 132768589182784 maxtext_utils.py:1501] params/params/decoder/layers/self_attention/out/kernel Shape: float32[16,16,128,2048] Logical: PartitionSpec('heads', 'layers', 'kv', 'embed') Physical: ('tensor', None, None, 'fsdp') I0419 09:29:15.220229 132768589182784 maxtext_utils.py:1501] params/params/decoder/layers/self_attention/query/kernel Shape: float32[2048,16,16,128] Logical: PartitionSpec('embed', 'layers', 'q_heads', 'kv') Physical: ('fsdp', None, 'tensor', None) I0419 09:29:15.220265 132768589182784 maxtext_utils.py:1501] 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, 'tensor', None) I0419 09:29:15.220314 132768589182784 maxtext_utils.py:1501] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', 'tensor') I0419 09:29:15.220361 132768589182784 maxtext_utils.py:1501] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: ('tensor', 'fsdp') I0419 09:29:15.719546 132768589182784 train.py:156] train/xent Logical: float32[8,2048]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0419 09:29:15.719641 132768589182784 train.py:156] train/xent Physical: float32[8,2048]............................................. ('fsdp', None). I0419 09:29:15.734635 132768589182784 train.py:163] train/z_loss Logical: float32[8,2048]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0419 09:29:15.734699 132768589182784 train.py:163] train/z_loss Physical: float32[8,2048]............................................. ('fsdp', None). I0419 09:29:25.694892 132768589182784 max_utils.py:791] Total memory size: 0.9 GB, Output size: 0.4 GB, Temp size: 0.5 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0419 09:29:25.695719 132768589182784 metric_logger.py:289] number parameters: 1.104 billion I0419 09:29:27.197933 132768589182784 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0419 09:29:35.795394 132768589182784 checkpointing.py:776] Waited 8.597440242767334 seconds for step 0 to finish before starting checkpointing. I0419 09:29:35.797804 132768589182784 checkpoint_manager.py:2013] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 09:29:35.799387 132768589182784 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 0 I0419 09:29:35.802145 132768589182784 async_checkpointer.py:452] [process=4] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_09_pdb_lt_1/checkpoints/0. I0419 09:29:36.784004 132768589182784 signaling_client.py:364] Using JaxDistributedSignalingClient I0419 09:29:36.785047 132768589182784 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 09:29:36.785106 132768589182784 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 09:29:37.010799 132768589182784 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.226889s I0419 09:29:37.010984 132768589182784 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.119 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.49451541900634766 s) (per-host) I0419 09:29:37.011040 132768589182784 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.494581s (batch_requests_ready=0.264154s, total_serialization_initiated=0.230353s, others=0.000075s) I0419 09:29:37.011130 132768589182784 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.498425s (all_items=0.000019s, per_item={'items': '0.00001931'}, temp_paths=0.498406) I0419 09:29:37.012094 132645409244928 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started. I0419 09:29:37.012264 132768589182784 async_checkpointer.py:561] Finished blocking save. Time taken: 1.212808s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_09_pdb_lt_1/checkpoints/0. I0419 09:29:37.468683 132768589182784 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0419 09:29:37.469077 132642204804864 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save. I0419 09:29:37.469244 132768589182784 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_09_pdb_lt_1/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776590975.7977855, 'wait_for_prev_duration_secs': 6.365776062011719e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776590975.7994256, 'checkpointer_blocking_duration_secs': 1.212944507598877, 'get_old_steps_start_time': 1776590977.0123906, 'get_old_steps_duration_secs': 2.7179718017578125e-05, 'checkpoint_manager_blocking_start_time': 1776590975.7959487, 'checkpoint_manager_blocking_duration_secs': 1.6732556819915771} I0419 09:29:37.469354 132768589182784 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0419 09:29:37.469405 132768589182784 max_utils.py:750] Memstats: After params initialized: I0419 09:29:37.469460 132768589182784 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_16(process=4,(0,4,0,0)) I0419 09:29:37.469492 132768589182784 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_17(process=4,(1,4,0,0)) I0419 09:29:37.469519 132768589182784 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_20(process=4,(0,5,0,0)) I0419 09:29:37.469544 132768589182784 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_21(process=4,(1,5,0,0)) I0419 09:29:37.731328 132768589182784 metric_logger.py:185] completed step: 0, seconds: 1.502, TFLOP/s/device: 2.261, Tokens/s/device: 340.854, total_weights: 16384, loss: 10.862 I0419 09:29:37.816350 132768589182784 metric_logger.py:185] completed step: 1, seconds: 10.532, TFLOP/s/device: 0.323, Tokens/s/device: 48.613, total_weights: 16384, loss: 10.862 I0419 09:29:38.225035 132768589182784 metric_logger.py:185] completed step: 2, seconds: 0.027, TFLOP/s/device: 126.911, Tokens/s/device: 19129.460, total_weights: 16384, loss: 9.763 I0419 09:29:38.283632 132768589182784 metric_logger.py:185] completed step: 3, seconds: 0.409, TFLOP/s/device: 8.306, Tokens/s/device: 1251.962, total_weights: 16384, loss: 8.823 I0419 09:29:38.401161 132768589182784 metric_logger.py:185] completed step: 4, seconds: 0.064, TFLOP/s/device: 53.277, Tokens/s/device: 8030.491, total_weights: 16384, loss: 7.947 I0419 09:29:38.406820 132768589182784 metric_logger.py:185] completed step: 5, seconds: 0.058, TFLOP/s/device: 58.085, Tokens/s/device: 8755.130, total_weights: 16384, loss: 7.212 I0419 09:29:50.873291 132768589182784 metric_logger.py:185] completed step: 6, seconds: 0.118, TFLOP/s/device: 28.773, Tokens/s/device: 4337.035, total_weights: 16384, loss: 6.652 I0419 09:29:50.932012 132768589182784 metric_logger.py:185] completed step: 7, seconds: 12.409, TFLOP/s/device: 0.274, Tokens/s/device: 41.262, total_weights: 16384, loss: 6.269 I0419 09:29:50.990602 132768589182784 metric_logger.py:185] completed step: 8, seconds: 0.063, TFLOP/s/device: 53.684, Tokens/s/device: 8091.791, total_weights: 16384, loss: 6.031 I0419 09:29:51.048779 132768589182784 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0419 09:29:51.049480 132768589182784 checkpointing.py:776] Waited 0.000713348388671875 seconds for step 9 to finish before starting checkpointing. I0419 09:29:51.051279 132768589182784 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 09:29:53.038786 2419 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0419 09:29:55.464222 132645417637632 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_09_pdb_lt_1/checkpoints/0/items/array_metadatas/process_4 I0419 09:30:26.784060 132645409244928 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 31.424 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 50.26755928993225 s) (per-host) I0419 09:30:26.784162 132645409244928 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 49.771957s. I0419 09:30:33.798840 132645409244928 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 56.786618s. I0419 09:30:33.799171 132642204804864 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 09:30:33.799293 132642204804864 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save. I0419 09:30:33.799343 132642204804864 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0419 09:30:33.800849 132642204804864 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0419 09:30:33.801056 132768589182784 checkpoint_manager.py:2036] [process=4][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0419 09:30:33.801191 132768589182784 checkpoint_manager.py:1458] Waiting for previous save to complete took 42.749915 seconds. If this number is high, consider checkpointing less frequently. I0419 09:30:33.803011 132768589182784 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 9 I0419 09:30:33.806576 132768589182784 async_checkpointer.py:452] [process=4] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_09_pdb_lt_1/checkpoints/9. I0419 09:30:34.777199 132768589182784 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 09:30:34.777297 132768589182784 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 09:30:34.822800 132768589182784 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.046738s I0419 09:30:34.822974 132768589182784 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.048 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3055582046508789 s) (per-host) I0419 09:30:34.823026 132768589182784 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.305619s (batch_requests_ready=0.256966s, total_serialization_initiated=0.048586s, others=0.000067s) I0419 09:30:34.823115 132768589182784 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.309830s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.309815) I0419 09:30:34.823989 132645417637632 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started. I0419 09:30:34.824146 132768589182784 async_checkpointer.py:561] Finished blocking save. Time taken: 1.021055s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_09_pdb_lt_1/checkpoints/9. I0419 09:30:34.826072 132768589182784 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0419 09:30:34.826292 132642204804864 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save. I0419 09:30:34.826413 132768589182784 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_09_pdb_lt_1/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776590991.0512507, 'wait_for_prev_duration_secs': 42.74991488456726, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776591033.803057, 'checkpointer_blocking_duration_secs': 1.0211966037750244, 'get_old_steps_start_time': 1776591034.8242745, 'get_old_steps_duration_secs': 2.47955322265625e-05, 'checkpoint_manager_blocking_start_time': 1776590991.0497017, 'checkpoint_manager_blocking_duration_secs': 43.77667737007141} I0419 09:30:34.826516 132768589182784 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0419 09:30:34.826559 132768589182784 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 09:30:41.019838 132615793227520 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_09_pdb_lt_1/checkpoints/9/items/array_metadatas/process_4 I0419 09:31:15.972914 132645417637632 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 38.103 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.45545291900635 s) (per-host) I0419 09:31:15.973043 132645417637632 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.148942s. I0419 09:31:24.532696 132645417637632 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 49.708581s. I0419 09:31:24.532909 132642204804864 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 09:31:24.532989 132642204804864 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save. I0419 09:31:24.533043 132642204804864 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0419 09:31:24.535694 132642204804864 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0419 09:31:24.535869 132768589182784 checkpoint_manager.py:2036] [process=4][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0419 09:31:24.536031 132768589182784 checkpoint_manager.py:2013] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 09:31:24.536723 132768589182784 metric_logger.py:185] completed step: 9, seconds: 0.059, TFLOP/s/device: 57.971, Tokens/s/device: 8737.947, total_weights: 16384, loss: 5.892 Per train step: Total TFLOPs: 3.40 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Sun Apr 19 09:31:37 UTC 2026 EXIT_CODE=0