XPK Start: Thu Apr 23 19:14:16 UTC 2026 2026-04-23 19:14:20.927931: 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 00:00:1776971660.940597 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776971660.944356 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776971660.955636 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776971660.955665 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776971660.955667 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776971660.955669 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-23 19:14:40.056518: 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) I0423 19:14:40.576222 133524407015232 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-23 19:14:49,618:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0423 19:14:49.618741 133524407015232 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-23 19:14:49,621:jax._src.distributed:157: Connecting to JAX distributed service on mt-03-dropout-450wx-slice-job-0-0.mt-03-dropout-450wx:8482 I0423 19:14:49.621107 133524407015232 distributed.py:157] Connecting to JAX distributed service on mt-03-dropout-450wx-slice-job-0-0.mt-03-dropout-450wx:8482 I0423 19:14:50.980415 133524407015232 max_utils.py:284] Jax distributed system initialized! I0423 19:14:56.698161 133524407015232 max_utils.py:800] System Information: Jax Version: 0.8.1 I0423 19:14:56.698265 133524407015232 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0423 19:14:56.698306 133524407015232 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 I0423 19:14:56.698340 133524407015232 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0423 19:14:57.318194 133524407015232 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0423 19:14:57.318492 133524407015232 checkpointing.py:677] Setting up checkpoint logger... I0423 19:14:57.318548 133524407015232 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0423 19:14:57.318593 133524407015232 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0423 19:14:57.318957 133524407015232 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 0x797018f78ce0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0423 19:15:00.742699 133524407015232 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0423 19:15:00.743044 133524407015232 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 0x795e79b71a30>}, handler_registry=None I0423 19:15:00.743286 133524407015232 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x795e79b71a30>` 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`. I0423 19:15:00.743336 133524407015232 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x795e79c0be90>` 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`. I0423 19:15:00.743372 133524407015232 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 0x795e79b71a30>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x795e79b71a30>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x795e79c0be90>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x795e79c0be90>}). I0423 19:15:00.743885 133524407015232 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0423 19:15:00.743965 133524407015232 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x795e79126840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0423 19:15:01.462922 133524407015232 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_03_dropout/checkpoints I0423 19:15:01.491932 133524407015232 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_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_03_dropout/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x795e786a0320> I0423 19:15:01.492070 133524407015232 checkpointing.py:301] Checkpoint manager created! I0423 19:15:01.808949 133524407015232 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0423 19:15:01.809053 133524407015232 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0423 19:15:02.180792 133524407015232 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0423 19:15:02.180885 133524407015232 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0423 19:15:02.196870 133524407015232 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0423 19:15:02.196929 133524407015232 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0423 19:15:02.219822 133524407015232 attentions.py:1154] attentions/query Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 19:15:02.219889 133524407015232 attentions.py:1154] attentions/query Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0423 19:15:02.235806 133524407015232 attentions.py:1155] attentions/key Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 19:15:02.235872 133524407015232 attentions.py:1155] attentions/key Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0423 19:15:02.251534 133524407015232 attentions.py:1156] attentions/value Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 19:15:02.251599 133524407015232 attentions.py:1156] attentions/value Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0423 19:15:02.280243 133524407015232 attentions.py:1197] attentions/out Logical: bfloat16[32,128,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0423 19:15:02.280344 133524407015232 attentions.py:1197] attentions/out Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0423 19:15:02.306153 133524407015232 linears.py:525] linears/x Logical: bfloat16[32,128,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0423 19:15:02.306230 133524407015232 linears.py:525] linears/x Physical: bfloat16[32,128,7168]....................................... ('fsdp', None, None). I0423 19:15:02.506841 133524407015232 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0423 19:15:02.506950 133524407015232 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0423 19:15:03.958927 133524407015232 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0423 19:15:03.959065 133524407015232 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) I0423 19:15:03.959121 133524407015232 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) I0423 19:15:03.959182 133524407015232 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') I0423 19:15:03.959236 133524407015232 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) I0423 19:15:03.959281 133524407015232 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) I0423 19:15:03.959336 133524407015232 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) I0423 19:15:03.959388 133524407015232 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') I0423 19:15:03.959428 133524407015232 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) I0423 19:15:03.959466 133524407015232 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) I0423 19:15:03.959514 133524407015232 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0423 19:15:03.959563 133524407015232 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0423 19:15:04.496989 133524407015232 train.py:155] train/xent Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0423 19:15:04.497094 133524407015232 train.py:155] train/xent Physical: float32[32,128]............................................. ('fsdp', None). I0423 19:15:04.512033 133524407015232 train.py:162] train/z_loss Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0423 19:15:04.512094 133524407015232 train.py:162] train/z_loss Physical: float32[32,128]............................................. ('fsdp', None). I0423 19:15:07.857247 133524407015232 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. I0423 19:15:07.858070 133524407015232 metric_logger.py:289] number parameters: 1.104 billion I0423 19:15:11.798731 133524407015232 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0423 19:15:12.077720 133524407015232 checkpointing.py:776] Waited 0.27896738052368164 seconds for step 0 to finish before starting checkpointing. I0423 19:15:12.080266 133524407015232 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0423 19:15:12.082643 133524407015232 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0423 19:15:12.085530 133524407015232 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_03_dropout/checkpoints/0. I0423 19:15:12.640558 133524407015232 signaling_client.py:364] Using JaxDistributedSignalingClient I0423 19:15:12.641625 133524407015232 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0423 19:15:12.641698 133524407015232 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 I0423 19:15:12.883795 133524407015232 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.243315s I0423 19:15:12.883969 133524407015232 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.998 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5145518779754639 s) (per-host) I0423 19:15:12.884022 133524407015232 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.514614s (batch_requests_ready=0.267651s, total_serialization_initiated=0.246894s, others=0.000069s) I0423 19:15:12.884114 133524407015232 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.518677s (all_items=0.000021s, per_item={'items': '0.00002074'}, temp_paths=0.518657) I0423 19:15:12.885100 133400073598720 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0423 19:15:12.885267 133524407015232 async_checkpointer.py:561] Finished blocking save. Time taken: 0.802539s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_03_dropout/checkpoints/0. I0423 19:15:13.022430 133524407015232 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0423 19:15:13.022835 133400040027904 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0423 19:15:13.023000 133524407015232 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776971712.0802486, 'wait_for_prev_duration_secs': 6.175041198730469e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776971712.0826972, 'checkpointer_blocking_duration_secs': 0.8026776313781738, 'get_old_steps_start_time': 1776971712.8853939, 'get_old_steps_duration_secs': 2.7179718017578125e-05, 'checkpoint_manager_blocking_start_time': 1776971712.0782828, 'checkpoint_manager_blocking_duration_secs': 0.9446766376495361} I0423 19:15:13.023114 133524407015232 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0423 19:15:13.023168 133524407015232 max_utils.py:750] Memstats: After params initialized: I0423 19:15:13.023236 133524407015232 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_18(process=5,(2,4,0,0)) I0423 19:15:13.023281 133524407015232 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_19(process=5,(3,4,0,0)) I0423 19:15:13.023322 133524407015232 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_22(process=5,(2,5,0,0)) I0423 19:15:13.023350 133524407015232 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_23(process=5,(3,5,0,0)) I0423 19:15:13.286839 133524407015232 metric_logger.py:185] completed step: 0, seconds: 3.941, TFLOP/s/device: 0.203, Tokens/s/device: 32.483, total_weights: 4096, loss: 10.889 I0423 19:15:13.382131 133524407015232 metric_logger.py:185] completed step: 1, seconds: 1.487, TFLOP/s/device: 0.539, Tokens/s/device: 86.088, total_weights: 4096, loss: 10.902 I0423 19:15:13.796861 133524407015232 metric_logger.py:185] completed step: 2, seconds: 0.025, TFLOP/s/device: 31.792, Tokens/s/device: 5081.180, total_weights: 4096, loss: 9.956 I0423 19:15:13.867580 133524407015232 metric_logger.py:185] completed step: 3, seconds: 0.415, TFLOP/s/device: 1.929, Tokens/s/device: 308.373, total_weights: 4096, loss: 9.139 I0423 19:15:14.009535 133524407015232 metric_logger.py:185] completed step: 4, seconds: 0.075, TFLOP/s/device: 10.639, Tokens/s/device: 1700.319, total_weights: 4096, loss: 8.458 I0423 19:15:14.014991 133524407015232 metric_logger.py:185] completed step: 5, seconds: 0.071, TFLOP/s/device: 11.341, Tokens/s/device: 1812.543, total_weights: 4096, loss: 7.924 I0423 19:15:23.282577 133524407015232 metric_logger.py:185] completed step: 6, seconds: 0.143, TFLOP/s/device: 5.618, Tokens/s/device: 897.937, total_weights: 4096, loss: 7.553 I0423 19:15:23.353427 133524407015232 metric_logger.py:185] completed step: 7, seconds: 9.198, TFLOP/s/device: 0.087, Tokens/s/device: 13.916, total_weights: 4096, loss: 7.267 I0423 19:15:23.424310 133524407015232 metric_logger.py:185] completed step: 8, seconds: 0.075, TFLOP/s/device: 10.713, Tokens/s/device: 1712.191, total_weights: 4096, loss: 7.111 I0423 19:15:23.494803 133524407015232 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0423 19:15:23.495429 133524407015232 checkpointing.py:776] Waited 0.000637054443359375 seconds for step 9 to finish before starting checkpointing. I0423 19:15:23.497541 133524407015232 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0423 19:15:24.892833 2589 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0423 19:15:26.830095 133400056813312 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_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_03_dropout/checkpoints/0/items/array_metadatas/process_5 I0423 19:15:54.713384 133400073598720 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.304 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.34392738342285 s) (per-host) I0423 19:15:54.713501 133400073598720 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.828289s. I0423 19:16:02.235497 133400073598720 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 49.350270s. I0423 19:16:02.235820 133400040027904 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0423 19:16:02.235939 133400040027904 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0423 19:16:02.235990 133400040027904 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0423 19:16:02.237428 133400040027904 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0423 19:16:02.237598 133524407015232 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. W0423 19:16:02.237754 133524407015232 checkpoint_manager.py:1458] Waiting for previous save to complete took 38.740216 seconds. If this number is high, consider checkpointing less frequently. I0423 19:16:02.240612 133524407015232 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0423 19:16:02.244125 133524407015232 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_03_dropout/checkpoints/9. I0423 19:16:02.768459 133524407015232 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0423 19:16:02.768555 133524407015232 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 I0423 19:16:02.810008 133524407015232 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.042850s I0423 19:16:02.810175 133524407015232 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.298 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2911686897277832 s) (per-host) I0423 19:16:02.810228 133524407015232 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.291230s (batch_requests_ready=0.246495s, total_serialization_initiated=0.044666s, others=0.000069s) I0423 19:16:02.810323 133524407015232 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.295305s (all_items=0.000016s, per_item={'items': '0.00001597'}, temp_paths=0.295290) I0423 19:16:02.811202 133400056813312 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0423 19:16:02.811362 133524407015232 async_checkpointer.py:561] Finished blocking save. Time taken: 0.570660s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_03_dropout/checkpoints/9. I0423 19:16:02.820469 133524407015232 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0423 19:16:02.820782 133400040027904 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0423 19:16:02.820941 133524407015232 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776971723.4975133, 'wait_for_prev_duration_secs': 38.74021577835083, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776971762.24067, 'checkpointer_blocking_duration_secs': 0.5708000659942627, 'get_old_steps_start_time': 1776971762.81149, 'get_old_steps_duration_secs': 2.4557113647460938e-05, 'checkpoint_manager_blocking_start_time': 1776971723.4956703, 'checkpoint_manager_blocking_duration_secs': 39.32523465156555} I0423 19:16:02.821047 133524407015232 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0423 19:16:02.821092 133524407015232 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0423 19:16:07.580439 133372080703232 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_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_03_dropout/checkpoints/9/items/array_metadatas/process_5 I0423 19:16:44.324875 133400056813312 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.784 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.80582785606384 s) (per-host) I0423 19:16:44.325000 133400056813312 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.513684s. I0423 19:16:52.031131 133400056813312 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 49.219802s. I0423 19:16:52.031426 133400040027904 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0423 19:16:52.031546 133400040027904 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0423 19:16:52.031595 133400040027904 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0423 19:16:52.033363 133400040027904 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0423 19:16:52.033534 133524407015232 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. I0423 19:16:52.033690 133524407015232 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0423 19:16:52.034383 133524407015232 metric_logger.py:185] completed step: 9, seconds: 0.071, TFLOP/s/device: 11.319, Tokens/s/device: 1809.111, 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: Thu Apr 23 19:17:03 UTC 2026 EXIT_CODE=0