XPK Start: Fri Apr 24 18:02:46 UTC 2026 2026-04-24 18:02:50.917118: 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 00:00:1777053770.929737 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1777053770.933437 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1777053770.944915 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777053770.944933 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777053770.944935 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777053770.944937 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-24 18:03:09.993283: 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) I0424 18:03:10.508050 137564429707072 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-24 18:03:19,548:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0424 18:03:19.548901 137564429707072 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-24 18:03:19,551:jax._src.distributed:157: Connecting to JAX distributed service on mt-02-synthetic-m27fz-slice-job-0-0.mt-02-synthetic-m27fz:8482 I0424 18:03:19.551122 137564429707072 distributed.py:157] Connecting to JAX distributed service on mt-02-synthetic-m27fz-slice-job-0-0.mt-02-synthetic-m27fz:8482 I0424 18:03:21.009058 137564429707072 max_utils.py:284] Jax distributed system initialized! I0424 18:03:28.029323 137564429707072 max_utils.py:800] System Information: Jax Version: 0.8.1 I0424 18:03:28.029430 137564429707072 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0424 18:03:28.029471 137564429707072 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 I0424 18:03:28.029504 137564429707072 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0424 18:03:28.645028 137564429707072 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0424 18:03:28.645316 137564429707072 checkpointing.py:677] Setting up checkpoint logger... I0424 18:03:28.645372 137564429707072 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0424 18:03:28.645416 137564429707072 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0424 18:03:28.645777 137564429707072 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 0x7d1cbc970f50>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0424 18:03:32.040541 137564429707072 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0424 18:03:32.040836 137564429707072 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 0x7d0a44f88620>}, handler_registry=None I0424 18:03:32.041069 137564429707072 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d0a44f88620>` 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`. I0424 18:03:32.041118 137564429707072 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d0a4536ff80>` 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`. I0424 18:03:32.041153 137564429707072 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 0x7d0a44f88620>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d0a44f88620>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d0a4536ff80>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d0a4536ff80>}). I0424 18:03:32.041664 137564429707072 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0424 18:03:32.041744 137564429707072 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d0a44af6840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0424 18:03:32.741545 137564429707072 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_02_synthetic/checkpoints I0424 18:03:32.770793 137564429707072 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_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d0a44f8bf20> I0424 18:03:32.770926 137564429707072 checkpointing.py:301] Checkpoint manager created! I0424 18:03:33.658969 137564429707072 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0424 18:03:33.659077 137564429707072 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0424 18:03:34.030263 137564429707072 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0424 18:03:34.030357 137564429707072 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0424 18:03:34.046160 137564429707072 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0424 18:03:34.046224 137564429707072 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0424 18:03:34.069740 137564429707072 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 18:03:34.069808 137564429707072 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:03:34.085790 137564429707072 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 18:03:34.085856 137564429707072 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:03:34.101758 137564429707072 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 18:03:34.101819 137564429707072 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:03:34.125729 137564429707072 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0424 18:03:34.125796 137564429707072 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:03:34.145226 137564429707072 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0424 18:03:34.145292 137564429707072 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0424 18:03:34.334896 137564429707072 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0424 18:03:34.335019 137564429707072 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0424 18:03:35.762571 137564429707072 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0424 18:03:35.762727 137564429707072 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) I0424 18:03:35.762790 137564429707072 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) I0424 18:03:35.762852 137564429707072 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') I0424 18:03:35.762907 137564429707072 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) I0424 18:03:35.762949 137564429707072 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) I0424 18:03:35.763008 137564429707072 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) I0424 18:03:35.763068 137564429707072 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') I0424 18:03:35.763112 137564429707072 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) I0424 18:03:35.763150 137564429707072 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) I0424 18:03:35.763200 137564429707072 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0424 18:03:35.763250 137564429707072 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0424 18:03:36.266826 137564429707072 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0424 18:03:36.266928 137564429707072 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0424 18:03:36.282027 137564429707072 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0424 18:03:36.282086 137564429707072 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0424 18:03:46.955469 137564429707072 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. I0424 18:03:46.956345 137564429707072 metric_logger.py:289] number parameters: 1.104 billion I0424 18:03:58.158516 137564429707072 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0424 18:03:58.327549 137564429707072 checkpointing.py:776] Waited 0.16901350021362305 seconds for step 0 to finish before starting checkpointing. I0424 18:03:58.330008 137564429707072 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0424 18:03:58.331585 137564429707072 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0424 18:03:58.334287 137564429707072 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_02_synthetic/checkpoints/0. I0424 18:03:58.887542 137564429707072 signaling_client.py:364] Using JaxDistributedSignalingClient I0424 18:03:58.888506 137564429707072 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0424 18:03:58.888565 137564429707072 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 I0424 18:03:59.144165 137564429707072 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.256698s I0424 18:03:59.144336 137564429707072 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.970 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5193288326263428 s) (per-host) I0424 18:03:59.144389 137564429707072 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.519390s (batch_requests_ready=0.259109s, total_serialization_initiated=0.260213s, others=0.000068s) I0424 18:03:59.144479 137564429707072 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.523518s (all_items=0.000021s, per_item={'items': '0.00002050'}, temp_paths=0.523498) I0424 18:03:59.145481 137438434227968 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0424 18:03:59.145671 137564429707072 async_checkpointer.py:561] Finished blocking save. Time taken: 0.814003s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_02_synthetic/checkpoints/0. I0424 18:03:59.610382 137564429707072 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0424 18:03:59.610782 137437368002304 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0424 18:03:59.610951 137564429707072 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777053838.3299913, 'wait_for_prev_duration_secs': 6.103515625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777053838.3316228, 'checkpointer_blocking_duration_secs': 0.814154863357544, 'get_old_steps_start_time': 1777053839.1457987, 'get_old_steps_duration_secs': 2.5987625122070312e-05, 'checkpoint_manager_blocking_start_time': 1777053838.3281543, 'checkpoint_manager_blocking_duration_secs': 1.2827527523040771} I0424 18:03:59.611069 137564429707072 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0424 18:03:59.611122 137564429707072 max_utils.py:750] Memstats: After params initialized: I0424 18:03:59.611180 137564429707072 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0)) I0424 18:03:59.611215 137564429707072 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0)) I0424 18:03:59.611243 137564429707072 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0)) I0424 18:03:59.611271 137564429707072 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0)) I0424 18:03:59.879502 137564429707072 metric_logger.py:185] completed step: 0, seconds: 11.202, TFLOP/s/device: 1.213, Tokens/s/device: 182.823, total_weights: 65536, loss: 10.874 I0424 18:04:00.041625 137564429707072 metric_logger.py:185] completed step: 1, seconds: 1.720, TFLOP/s/device: 7.901, Tokens/s/device: 1190.948, total_weights: 65536, loss: 10.874 I0424 18:04:00.507280 137564429707072 metric_logger.py:185] completed step: 2, seconds: 0.015, TFLOP/s/device: 905.990, Tokens/s/device: 136560.645, total_weights: 65536, loss: 10.263 I0424 18:04:00.654725 137564429707072 metric_logger.py:185] completed step: 3, seconds: 0.466, TFLOP/s/device: 29.159, Tokens/s/device: 4395.189, total_weights: 65536, loss: 9.734 I0424 18:04:00.950300 137564429707072 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.996, Tokens/s/device: 13414.466, total_weights: 65536, loss: 9.278 I0424 18:04:00.955897 137564429707072 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.190, Tokens/s/device: 13895.862, total_weights: 65536, loss: 8.892 I0424 18:04:19.696316 137564429707072 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.911, Tokens/s/device: 6920.275, total_weights: 65536, loss: 8.592 I0424 18:04:19.843860 137564429707072 metric_logger.py:185] completed step: 7, seconds: 18.594, TFLOP/s/device: 0.731, Tokens/s/device: 110.145, total_weights: 65536, loss: 8.384 I0424 18:04:19.991149 137564429707072 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.300, Tokens/s/device: 13460.312, total_weights: 65536, loss: 8.255 I0424 18:04:20.138254 137564429707072 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0424 18:04:20.138913 137564429707072 checkpointing.py:776] Waited 0.0006792545318603516 seconds for step 9 to finish before starting checkpointing. I0424 18:04:20.140838 137564429707072 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0424 18:04:20.780530 2663 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0424 18:04:22.738312 137438483707648 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_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_02_synthetic/checkpoints/0/items/array_metadatas/process_5 I0424 18:04:54.365569 137438434227968 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 28.338 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 55.74052357673645 s) (per-host) I0424 18:04:54.365686 137438434227968 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 55.220090s. I0424 18:05:02.515451 137438434227968 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 63.369840s. I0424 18:05:02.515769 137437368002304 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0424 18:05:02.515889 137437368002304 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0424 18:05:02.515951 137437368002304 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0424 18:05:02.517605 137437368002304 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0424 18:05:02.517818 137564429707072 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. W0424 18:05:02.517947 137564429707072 checkpoint_manager.py:1458] Waiting for previous save to complete took 42.377113 seconds. If this number is high, consider checkpointing less frequently. I0424 18:05:02.519759 137564429707072 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0424 18:05:02.523185 137564429707072 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_02_synthetic/checkpoints/9. I0424 18:05:03.074393 137564429707072 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0424 18:05:03.074492 137564429707072 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 I0424 18:05:03.122352 137564429707072 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.049130s I0424 18:05:03.122501 137564429707072 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 4.895 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3151376247406006 s) (per-host) I0424 18:05:03.122553 137564429707072 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.315198s (batch_requests_ready=0.264151s, total_serialization_initiated=0.050980s, others=0.000067s) I0424 18:05:03.122643 137564429707072 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.319248s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.319233) I0424 18:05:03.123492 137437919282944 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0424 18:05:03.123662 137564429707072 async_checkpointer.py:561] Finished blocking save. Time taken: 0.603817s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_02_synthetic/checkpoints/9. I0424 18:05:03.131313 137564429707072 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0424 18:05:03.131593 137437368002304 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0424 18:05:03.131766 137564429707072 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777053860.1408098, 'wait_for_prev_duration_secs': 42.37711310386658, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777053902.5197985, 'checkpointer_blocking_duration_secs': 0.6039714813232422, 'get_old_steps_start_time': 1777053903.1237907, 'get_old_steps_duration_secs': 2.5033950805664062e-05, 'checkpoint_manager_blocking_start_time': 1777053860.139149, 'checkpoint_manager_blocking_duration_secs': 42.99258327484131} I0424 18:05:03.131871 137564429707072 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0424 18:05:03.131916 137564429707072 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0424 18:05:08.001570 137438483707648 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_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_02_synthetic/checkpoints/9/items/array_metadatas/process_5 I0424 18:05:45.359300 137437919282944 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.122 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.55189251899719 s) (per-host) I0424 18:05:45.359418 137437919282944 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.235817s. I0424 18:05:51.323639 137437919282944 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 48.200023s. I0424 18:05:51.323940 137437368002304 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0424 18:05:51.324053 137437368002304 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0424 18:05:51.324101 137437368002304 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0424 18:05:51.325783 137437368002304 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0424 18:05:51.325939 137564429707072 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. I0424 18:05:51.326083 137564429707072 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0424 18:05:51.326727 137564429707072 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.176, Tokens/s/device: 13893.788, 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: Fri Apr 24 18:06:02 UTC 2026 EXIT_CODE=0