XPK Start: Sat Apr 25 18:10:09 UTC 2026 2026-04-25 18:10:13.353555: 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 00:00:1777140613.366422 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1777140613.370107 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1777140613.381304 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777140613.381326 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777140613.381328 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777140613.381330 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-25 18:10:32.597110: 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) I0425 18:10:33.118964 132949762910016 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-25 18:10:42,159:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0425 18:10:42.159969 132949762910016 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-25 18:10:42,162:jax._src.distributed:157: Connecting to JAX distributed service on mt-03-dropout-50j8j-slice-job-0-0.mt-03-dropout-50j8j:8482 I0425 18:10:42.162331 132949762910016 distributed.py:157] Connecting to JAX distributed service on mt-03-dropout-50j8j-slice-job-0-0.mt-03-dropout-50j8j:8482 I0425 18:10:43.633843 132949762910016 max_utils.py:284] Jax distributed system initialized! I0425 18:10:49.715528 132949762910016 max_utils.py:800] System Information: Jax Version: 0.8.1 I0425 18:10:49.715637 132949762910016 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0425 18:10:49.715695 132949762910016 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 I0425 18:10:49.715733 132949762910016 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0425 18:10:50.332433 132949762910016 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0425 18:10:50.332743 132949762910016 checkpointing.py:677] Setting up checkpoint logger... I0425 18:10:50.332801 132949762910016 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0425 18:10:50.332847 132949762910016 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0425 18:10:50.333197 132949762910016 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 0x78ea38f70c50>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0425 18:10:53.300343 132949762910016 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0425 18:10:53.300534 132949762910016 checkpoint_manager.py:709] [process=5][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78d805a383b0>}, handler_registry=None I0425 18:10:53.300777 132949762910016 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78d805a383b0>` 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`. I0425 18:10:53.300828 132949762910016 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78e9023cc9e0>` 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`. I0425 18:10:53.300863 132949762910016 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 0x78d805a383b0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78d805a383b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78e9023cc9e0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78e9023cc9e0>}). I0425 18:10:53.301368 132949762910016 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0425 18:10:53.301447 132949762910016 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x78d7bfae6840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0425 18:10:55.215352 132949762910016 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_03_dropout/checkpoints I0425 18:10:56.078511 132949762910016 checkpoint_manager.py:929] [process=5][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_03_dropout/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x78d805a3bef0> I0425 18:10:56.078700 132949762910016 checkpointing.py:301] Checkpoint manager created! I0425 18:10:56.393204 132949762910016 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0425 18:10:56.393304 132949762910016 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0425 18:10:56.762628 132949762910016 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0425 18:10:56.762734 132949762910016 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0425 18:10:56.778428 132949762910016 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0425 18:10:56.778493 132949762910016 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0425 18:10:56.801794 132949762910016 attentions.py:1154] attentions/query Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 18:10:56.801869 132949762910016 attentions.py:1154] attentions/query Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0425 18:10:56.817852 132949762910016 attentions.py:1155] attentions/key Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 18:10:56.817911 132949762910016 attentions.py:1155] attentions/key Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0425 18:10:56.833878 132949762910016 attentions.py:1156] attentions/value Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 18:10:56.833937 132949762910016 attentions.py:1156] attentions/value Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0425 18:10:56.857873 132949762910016 attentions.py:1197] attentions/out Logical: bfloat16[32,128,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0425 18:10:56.857945 132949762910016 attentions.py:1197] attentions/out Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0425 18:10:56.883543 132949762910016 linears.py:525] linears/x Logical: bfloat16[32,128,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0425 18:10:56.883613 132949762910016 linears.py:525] linears/x Physical: bfloat16[32,128,7168]....................................... ('fsdp', None, None). I0425 18:10:57.089275 132949762910016 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0425 18:10:57.089390 132949762910016 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0425 18:10:58.539190 132949762910016 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0425 18:10:58.539322 132949762910016 maxtext_utils.py:1620] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0425 18:10:58.539376 132949762910016 maxtext_utils.py:1620] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0425 18:10:58.539433 132949762910016 maxtext_utils.py:1620] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: PartitionSpec('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0425 18:10:58.539485 132949762910016 maxtext_utils.py:1620] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0425 18:10:58.539523 132949762910016 maxtext_utils.py:1620] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0425 18:10:58.539575 132949762910016 maxtext_utils.py:1620] 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) I0425 18:10:58.539627 132949762910016 maxtext_utils.py:1620] 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') I0425 18:10:58.539697 132949762910016 maxtext_utils.py:1620] 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) I0425 18:10:58.539746 132949762910016 maxtext_utils.py:1620] 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) I0425 18:10:58.539804 132949762910016 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0425 18:10:58.539857 132949762910016 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0425 18:10:59.079584 132949762910016 train.py:155] train/xent Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0425 18:10:59.079694 132949762910016 train.py:155] train/xent Physical: float32[32,128]............................................. ('fsdp', None). I0425 18:10:59.094704 132949762910016 train.py:162] train/z_loss Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0425 18:10:59.094764 132949762910016 train.py:162] train/z_loss Physical: float32[32,128]............................................. ('fsdp', None). I0425 18:11:02.440748 132949762910016 max_utils.py:791] Total memory size: 0.8 GB, Output size: 0.4 GB, Temp size: 0.4 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0425 18:11:02.441539 132949762910016 metric_logger.py:289] number parameters: 1.104 billion I0425 18:11:06.358452 132949762910016 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0425 18:11:06.675731 132949762910016 checkpointing.py:776] Waited 0.3172624111175537 seconds for step 0 to finish before starting checkpointing. I0425 18:11:06.679023 132949762910016 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0425 18:11:06.680983 132949762910016 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0425 18:11:06.683759 132949762910016 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_03_dropout/checkpoints/0. I0425 18:11:08.080871 132949762910016 signaling_client.py:364] Using JaxDistributedSignalingClient I0425 18:11:08.081905 132949762910016 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0425 18:11:08.081964 132949762910016 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 I0425 18:11:08.323392 132949762910016 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.242595s I0425 18:11:08.323569 132949762910016 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.661 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.9285073280334473 s) (per-host) I0425 18:11:08.323625 132949762910016 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.928573s (batch_requests_ready=0.682389s, total_serialization_initiated=0.246111s, others=0.000073s) I0425 18:11:08.323729 132949762910016 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.932689s (all_items=0.000021s, per_item={'items': '0.00002050'}, temp_paths=0.932669) I0425 18:11:08.324694 132830478706432 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0425 18:11:08.324859 132949762910016 async_checkpointer.py:561] Finished blocking save. Time taken: 1.643807s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_03_dropout/checkpoints/0. I0425 18:11:08.347341 132949762910016 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0425 18:11:08.347637 132825126782720 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0425 18:11:08.347816 132949762910016 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777140666.6790032, 'wait_for_prev_duration_secs': 6.723403930664062e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777140666.6810195, 'checkpointer_blocking_duration_secs': 1.6439471244812012, 'get_old_steps_start_time': 1777140668.3249853, 'get_old_steps_duration_secs': 2.7418136596679688e-05, 'checkpoint_manager_blocking_start_time': 1777140666.6763191, 'checkpoint_manager_blocking_duration_secs': 1.6714591979980469} I0425 18:11:08.347923 132949762910016 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0425 18:11:08.347974 132949762910016 max_utils.py:750] Memstats: After params initialized: I0425 18:11:08.348028 132949762910016 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_18(process=5,(2,4,0,0)) I0425 18:11:08.348062 132949762910016 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_19(process=5,(3,4,0,0)) I0425 18:11:08.348090 132949762910016 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_22(process=5,(2,5,0,0)) I0425 18:11:08.348115 132949762910016 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_23(process=5,(3,5,0,0)) I0425 18:11:08.610144 132949762910016 metric_logger.py:185] completed step: 0, seconds: 3.917, TFLOP/s/device: 0.204, Tokens/s/device: 32.680, total_weights: 4096, loss: 10.889 I0425 18:11:08.721891 132949762910016 metric_logger.py:185] completed step: 1, seconds: 2.250, TFLOP/s/device: 0.356, Tokens/s/device: 56.880, total_weights: 4096, loss: 10.902 I0425 18:11:09.121207 132949762910016 metric_logger.py:185] completed step: 2, seconds: 0.042, TFLOP/s/device: 19.038, Tokens/s/device: 3042.693, total_weights: 4096, loss: 9.956 I0425 18:11:09.194168 132949762910016 metric_logger.py:185] completed step: 3, seconds: 0.400, TFLOP/s/device: 2.003, Tokens/s/device: 320.206, total_weights: 4096, loss: 9.139 I0425 18:11:09.336169 132949762910016 metric_logger.py:185] completed step: 4, seconds: 0.075, TFLOP/s/device: 10.653, Tokens/s/device: 1702.671, total_weights: 4096, loss: 8.458 I0425 18:11:09.341837 132949762910016 metric_logger.py:185] completed step: 5, seconds: 0.073, TFLOP/s/device: 10.978, Tokens/s/device: 1754.506, total_weights: 4096, loss: 7.924 I0425 18:11:20.272245 132949762910016 metric_logger.py:185] completed step: 6, seconds: 0.142, TFLOP/s/device: 5.627, Tokens/s/device: 899.287, total_weights: 4096, loss: 7.553 I0425 18:11:20.343057 132949762910016 metric_logger.py:185] completed step: 7, seconds: 10.861, TFLOP/s/device: 0.074, Tokens/s/device: 11.785, total_weights: 4096, loss: 7.267 I0425 18:11:20.413735 132949762910016 metric_logger.py:185] completed step: 8, seconds: 0.075, TFLOP/s/device: 10.720, Tokens/s/device: 1713.360, total_weights: 4096, loss: 7.111 I0425 18:11:20.484297 132949762910016 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0425 18:11:20.484955 132949762910016 checkpointing.py:776] Waited 0.0006763935089111328 seconds for step 9 to finish before starting checkpointing. I0425 18:11:20.486816 132949762910016 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 18:11:23.268670 2600 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0425 18:11:25.660118 132830487099136 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_03_dropout/checkpoints/0/items/array_metadatas/process_5 I0425 18:11:51.937633 132830478706432 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 35.463 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 44.54252910614014 s) (per-host) I0425 18:11:51.937770 132830478706432 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.612964s. I0425 18:12:01.882853 132830478706432 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 53.558030s. I0425 18:12:01.883169 132825126782720 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 18:12:01.883295 132825126782720 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0425 18:12:01.883349 132825126782720 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0425 18:12:01.885115 132825126782720 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0425 18:12:01.885299 132949762910016 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0425 18:12:01.885431 132949762910016 checkpoint_manager.py:1458] Waiting for previous save to complete took 41.398619 seconds. If this number is high, consider checkpointing less frequently. I0425 18:12:01.887226 132949762910016 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0425 18:12:01.890588 132949762910016 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_03_dropout/checkpoints/9. I0425 18:12:02.855927 132949762910016 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0425 18:12:02.856031 132949762910016 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 I0425 18:12:02.899855 132949762910016 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.045064s I0425 18:12:02.900014 132949762910016 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.136 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3003678321838379 s) (per-host) I0425 18:12:02.900067 132949762910016 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.300435s (batch_requests_ready=0.253504s, total_serialization_initiated=0.046859s, others=0.000072s) I0425 18:12:02.900152 132949762910016 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.304436s (all_items=0.000014s, per_item={'items': '0.00001431'}, temp_paths=0.304421) I0425 18:12:02.900999 132830487099136 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0425 18:12:02.901157 132949762910016 async_checkpointer.py:561] Finished blocking save. Time taken: 1.013859s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_03_dropout/checkpoints/9. I0425 18:12:03.336056 132949762910016 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0425 18:12:03.336449 132825126782720 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0425 18:12:03.336622 132949762910016 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777140680.4867873, 'wait_for_prev_duration_secs': 41.398618936538696, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777140721.8872662, 'checkpointer_blocking_duration_secs': 1.0139963626861572, 'get_old_steps_start_time': 1777140722.9012833, 'get_old_steps_duration_secs': 2.7418136596679688e-05, 'checkpoint_manager_blocking_start_time': 1777140680.4851894, 'checkpoint_manager_blocking_duration_secs': 42.85139560699463} I0425 18:12:03.336776 132949762910016 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0425 18:12:03.336825 132949762910016 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 18:12:08.995995 132797406557952 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_03_dropout/checkpoints/9/items/array_metadatas/process_5 I0425 18:12:45.015348 132830487099136 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.241 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.41566491127014 s) (per-host) I0425 18:12:45.015473 132830487099136 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.114362s. I0425 18:12:52.781618 132830487099136 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 49.880493s. I0425 18:12:52.781916 132825126782720 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 18:12:52.782052 132825126782720 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0425 18:12:52.782100 132825126782720 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0425 18:12:52.783595 132825126782720 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0425 18:12:52.783783 132949762910016 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0425 18:12:52.783926 132949762910016 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0425 18:12:52.784688 132949762910016 metric_logger.py:185] completed step: 9, seconds: 0.071, TFLOP/s/device: 11.325, Tokens/s/device: 1810.083, total_weights: 4096, loss: 6.981 Per train step: Total TFLOPs: 0.80 split as 99.60% learnable weight flops and 0.40% attention flops XPK End: Sat Apr 25 18:13:04 UTC 2026 EXIT_CODE=0