XPK Start: Sun Apr 19 09:41:34 UTC 2026 2026-04-19 09:41:38.941365: 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 00:00:1776591698.954388 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776591698.958124 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776591698.969599 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776591698.969617 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776591698.969620 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776591698.969622 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-19 09:41:58.120740: 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 09:41:58.636385 138224638232384 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-19 09:42:07,677:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0419 09:42:07.677348 138224638232384 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-19 09:42:07,679:jax._src.distributed:157: Connecting to JAX distributed service on mt-13-scan-layers-true-ruw2u-slice-job-0-0.mt-13-scan-layers-true-ruw2u:8482 I0419 09:42:07.679664 138224638232384 distributed.py:157] Connecting to JAX distributed service on mt-13-scan-layers-true-ruw2u-slice-job-0-0.mt-13-scan-layers-true-ruw2u:8482 I0419 09:42:08.734265 138224638232384 max_utils.py:284] Jax distributed system initialized! I0419 09:42:15.280697 138224638232384 max_utils.py:800] System Information: Jax Version: 0.8.1 I0419 09:42:15.280809 138224638232384 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0419 09:42:15.280849 138224638232384 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 09:42:15.280886 138224638232384 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0419 09:42:15.903925 138224638232384 maxtext_utils.py:1398] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 09:42:15.904248 138224638232384 checkpointing.py:677] Setting up checkpoint logger... I0419 09:42:15.904303 138224638232384 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0419 09:42:15.904347 138224638232384 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0419 09:42:15.904699 138224638232384 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 0x7db65d975520>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0419 09:42:18.751225 138224638232384 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0419 09:42:18.751567 138224638232384 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 0x7db5292a2360>}, handler_registry=None I0419 09:42:18.751799 138224638232384 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7db5292a2360>` 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 09:42:18.751851 138224638232384 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7da2a85b88c0>` 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 09:42:18.751889 138224638232384 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 0x7db5292a2360>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7db5292a2360>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7da2a85b88c0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7da2a85b88c0>}). I0419 09:42:18.754598 138224638232384 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0419 09:42:18.754679 138224638232384 async_checkpointer.py:177] [process=4][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7db636bb72e0> timeout: 600 secs and primary_host=0 for async checkpoint writes I0419 09:42:20.401386 138224638232384 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_13_scan_layers_true/checkpoints I0419 09:42:20.847598 138224638232384 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_13_scan_layers_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7da2a85b8260> I0419 09:42:20.847774 138224638232384 checkpointing.py:301] Checkpoint manager created! I0419 09:42:21.741744 138224638232384 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0419 09:42:21.741847 138224638232384 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 09:42:22.114616 138224638232384 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 09:42:22.114712 138224638232384 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 09:42:22.130949 138224638232384 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 09:42:22.131010 138224638232384 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 09:42:22.154473 138224638232384 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 09:42:22.154537 138224638232384 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 09:42:22.170545 138224638232384 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 09:42:22.170610 138224638232384 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 09:42:22.186503 138224638232384 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 09:42:22.186566 138224638232384 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 09:42:22.210530 138224638232384 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0419 09:42:22.210602 138224638232384 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 09:42:22.230231 138224638232384 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0419 09:42:22.230293 138224638232384 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0419 09:42:22.421719 138224638232384 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0419 09:42:22.421828 138224638232384 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0419 09:42:23.867273 138224638232384 maxtext_utils.py:1501] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0419 09:42:23.867406 138224638232384 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, None) I0419 09:42:23.867462 138224638232384 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, None) I0419 09:42:23.867521 138224638232384 maxtext_utils.py:1501] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: PartitionSpec('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0419 09:42:23.867574 138224638232384 maxtext_utils.py:1501] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0419 09:42:23.867614 138224638232384 maxtext_utils.py:1501] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0419 09:42:23.867667 138224638232384 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, None, None) I0419 09:42:23.867719 138224638232384 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: (None, None, None, 'fsdp') I0419 09:42:23.867759 138224638232384 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, None, None) I0419 09:42:23.867803 138224638232384 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, None, None) I0419 09:42:23.867851 138224638232384 maxtext_utils.py:1501] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0419 09:42:23.867897 138224638232384 maxtext_utils.py:1501] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0419 09:42:24.376904 138224638232384 train.py:156] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 09:42:24.377021 138224638232384 train.py:156] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0419 09:42:24.392088 138224638232384 train.py:163] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 09:42:24.392149 138224638232384 train.py:163] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0419 09:42:35.261396 138224638232384 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 09:42:35.262241 138224638232384 metric_logger.py:289] number parameters: 1.104 billion I0419 09:42:46.211571 138224638232384 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0419 09:42:46.365304 138224638232384 checkpointing.py:776] Waited 0.15371441841125488 seconds for step 0 to finish before starting checkpointing. I0419 09:42:46.367674 138224638232384 checkpoint_manager.py:2013] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 09:42:46.369817 138224638232384 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 0 I0419 09:42:46.372581 138224638232384 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_13_scan_layers_true/checkpoints/0. I0419 09:42:47.417847 138224638232384 signaling_client.py:364] Using JaxDistributedSignalingClient I0419 09:42:47.418983 138224638232384 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 09:42:47.419039 138224638232384 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 09:42:47.670439 138224638232384 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.252682s I0419 09:42:47.670607 138224638232384 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.598 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.965146541595459 s) (per-host) I0419 09:42:47.670659 138224638232384 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.965207s (batch_requests_ready=0.708920s, total_serialization_initiated=0.256220s, others=0.000068s) I0419 09:42:47.670748 138224638232384 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.969368s (all_items=0.000019s, per_item={'items': '0.00001884'}, temp_paths=0.969349) I0419 09:42:47.671750 138105436743424 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started. I0419 09:42:47.671881 138224638232384 async_checkpointer.py:561] Finished blocking save. Time taken: 1.301993s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_13_scan_layers_true/checkpoints/0. I0419 09:42:47.674269 138224638232384 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0419 09:42:47.674550 138099532039936 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save. I0419 09:42:47.674703 138224638232384 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_13_scan_layers_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776591766.3676548, 'wait_for_prev_duration_secs': 6.699562072753906e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776591766.3698556, 'checkpointer_blocking_duration_secs': 1.302215814590454, 'get_old_steps_start_time': 1776591767.6720932, 'get_old_steps_duration_secs': 2.5987625122070312e-05, 'checkpoint_manager_blocking_start_time': 1776591766.365908, 'checkpoint_manager_blocking_duration_secs': 1.3087592124938965} I0419 09:42:47.674811 138224638232384 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0419 09:42:47.674862 138224638232384 max_utils.py:750] Memstats: After params initialized: I0419 09:42:47.674909 138224638232384 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_16(process=4,(0,4,0,0)) I0419 09:42:47.674952 138224638232384 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_17(process=4,(1,4,0,0)) I0419 09:42:47.674981 138224638232384 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_20(process=4,(0,5,0,0)) I0419 09:42:47.675004 138224638232384 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_21(process=4,(1,5,0,0)) I0419 09:42:47.937427 138224638232384 metric_logger.py:185] completed step: 0, seconds: 10.949, TFLOP/s/device: 1.241, Tokens/s/device: 187.045, total_weights: 65536, loss: 10.874 I0419 09:42:48.109154 138224638232384 metric_logger.py:185] completed step: 1, seconds: 1.725, TFLOP/s/device: 7.879, Tokens/s/device: 1187.536, total_weights: 65536, loss: 10.874 I0419 09:42:48.498380 138224638232384 metric_logger.py:185] completed step: 2, seconds: 0.025, TFLOP/s/device: 549.641, Tokens/s/device: 82847.896, total_weights: 65536, loss: 10.263 I0419 09:42:48.646008 138224638232384 metric_logger.py:185] completed step: 3, seconds: 0.390, TFLOP/s/device: 34.870, Tokens/s/device: 5256.012, total_weights: 65536, loss: 9.734 I0419 09:42:48.941292 138224638232384 metric_logger.py:185] completed step: 4, seconds: 0.152, TFLOP/s/device: 89.293, Tokens/s/device: 13459.162, total_weights: 65536, loss: 9.278 I0419 09:42:48.946474 138224638232384 metric_logger.py:185] completed step: 5, seconds: 0.148, TFLOP/s/device: 92.064, Tokens/s/device: 13876.937, total_weights: 65536, loss: 8.892 I0419 09:43:08.360779 138224638232384 metric_logger.py:185] completed step: 6, seconds: 0.295, TFLOP/s/device: 45.987, Tokens/s/device: 6931.635, total_weights: 65536, loss: 8.592 I0419 09:43:08.508611 138224638232384 metric_logger.py:185] completed step: 7, seconds: 19.268, TFLOP/s/device: 0.705, Tokens/s/device: 106.292, total_weights: 65536, loss: 8.384 I0419 09:43:08.655920 138224638232384 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.536, Tokens/s/device: 13495.881, total_weights: 65536, loss: 8.255 I0419 09:43:08.803242 138224638232384 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0419 09:43:08.803820 138224638232384 checkpointing.py:776] Waited 0.0006024837493896484 seconds for step 9 to finish before starting checkpointing. I0419 09:43:08.805778 138224638232384 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 09:43:09.922622 2586 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0419 09:43:11.830308 138105445136128 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_13_scan_layers_true/checkpoints/0/items/array_metadatas/process_4 I0419 09:43:43.129245 138105436743424 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 27.995 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 56.423741579055786 s) (per-host) I0419 09:43:43.129364 138105436743424 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 55.457394s. I0419 09:43:51.213607 138105436743424 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 63.541620s. I0419 09:43:51.213899 138099532039936 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 09:43:51.214035 138099532039936 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save. I0419 09:43:51.214087 138099532039936 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0419 09:43:51.215533 138099532039936 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0419 09:43:51.215716 138224638232384 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 09:43:51.215868 138224638232384 checkpoint_manager.py:1458] Waiting for previous save to complete took 42.410100 seconds. If this number is high, consider checkpointing less frequently. I0419 09:43:51.217659 138224638232384 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 9 I0419 09:43:51.221256 138224638232384 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_13_scan_layers_true/checkpoints/9. I0419 09:43:52.604770 138224638232384 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 09:43:52.604872 138224638232384 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 09:43:52.652600 138224638232384 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.049616s I0419 09:43:52.652762 138224638232384 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.018 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.30738186836242676 s) (per-host) I0419 09:43:52.652815 138224638232384 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.307443s (batch_requests_ready=0.255841s, total_serialization_initiated=0.051533s, others=0.000068s) I0419 09:43:52.652901 138224638232384 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.311346s (all_items=0.000016s, per_item={'items': '0.00001597'}, temp_paths=0.311330) I0419 09:43:52.653811 138100637353728 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started. I0419 09:43:52.653987 138224638232384 async_checkpointer.py:561] Finished blocking save. Time taken: 1.436255s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260419_085454/linen_xpk_test_pipeline_scan_nnx_20260419_085454_13_scan_layers_true/checkpoints/9. I0419 09:43:52.657007 138224638232384 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0419 09:43:52.657262 138099532039936 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save. I0419 09:43:52.657421 138224638232384 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_13_scan_layers_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776591788.8057432, 'wait_for_prev_duration_secs': 42.41009974479675, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776591831.2176998, 'checkpointer_blocking_duration_secs': 1.436394214630127, 'get_old_steps_start_time': 1776591832.6541162, 'get_old_steps_duration_secs': 2.574920654296875e-05, 'checkpoint_manager_blocking_start_time': 1776591788.8040574, 'checkpoint_manager_blocking_duration_secs': 43.85332894325256} I0419 09:43:52.657525 138224638232384 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0419 09:43:52.657572 138224638232384 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 09:43:57.838642 138105445136128 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_13_scan_layers_true/checkpoints/9/items/array_metadatas/process_4 I0419 09:44:34.050695 138100637353728 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 37.875 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.70527267456055 s) (per-host) I0419 09:44:34.050813 138100637353728 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.396891s. I0419 09:44:43.052098 138100637353728 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 50.398160s. I0419 09:44:43.052385 138099532039936 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 09:44:43.052502 138099532039936 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save. I0419 09:44:43.052552 138099532039936 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0419 09:44:43.054019 138099532039936 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0419 09:44:43.054207 138224638232384 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 09:44:43.054352 138224638232384 checkpoint_manager.py:2013] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 09:44:43.055049 138224638232384 metric_logger.py:185] completed step: 9, seconds: 0.148, TFLOP/s/device: 91.968, Tokens/s/device: 13862.472, total_weights: 65536, loss: 8.179 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Sun Apr 19 09:44:55 UTC 2026 EXIT_CODE=0