XPK Start: Thu Apr 23 19:10:52 UTC 2026 2026-04-23 19:10:56.247754: 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 00:00:1776971456.260717 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776971456.264403 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776971456.275765 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776971456.275782 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776971456.275785 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776971456.275787 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-23 19:11:15.444539: 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 19:11:15.954723 132092006360896 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-23 19:11:24,995:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0423 19:11:24.995628 132092006360896 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-23 19:11:24,997:jax._src.distributed:157: Connecting to JAX distributed service on mt-02-synthetic-jk1dr-slice-job-0-0.mt-02-synthetic-jk1dr:8482 I0423 19:11:24.997983 132092006360896 distributed.py:157] Connecting to JAX distributed service on mt-02-synthetic-jk1dr-slice-job-0-0.mt-02-synthetic-jk1dr:8482 I0423 19:11:26.311114 132092006360896 max_utils.py:284] Jax distributed system initialized! I0423 19:11:33.286532 132092006360896 max_utils.py:800] System Information: Jax Version: 0.8.1 I0423 19:11:33.286638 132092006360896 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0423 19:11:33.286703 132092006360896 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 19:11:33.286744 132092006360896 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0423 19:11:33.903307 132092006360896 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0423 19:11:33.903593 132092006360896 checkpointing.py:677] Setting up checkpoint logger... I0423 19:11:33.903661 132092006360896 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0423 19:11:33.903706 132092006360896 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0423 19:11:33.904099 132092006360896 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 0x78228834a750>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0423 19:11:36.879548 132092006360896 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0423 19:11:36.879752 132092006360896 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 0x7810178bddf0>}, handler_registry=None I0423 19:11:36.879985 132092006360896 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7810178bddf0>` 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 19:11:36.880033 132092006360896 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x780e7fa9f260>` 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 19:11:36.880068 132092006360896 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 0x7810178bddf0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7810178bddf0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x780e7fa9f260>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x780e7fa9f260>}). I0423 19:11:36.880564 132092006360896 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0423 19:11:36.880642 132092006360896 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x780e1477aac0> timeout: 600 secs and primary_host=0 for async checkpoint writes I0423 19:11:38.043964 132092006360896 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_02_synthetic/checkpoints I0423 19:11:38.501194 132092006360896 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_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x780e7fb338f0> I0423 19:11:38.501369 132092006360896 checkpointing.py:301] Checkpoint manager created! I0423 19:11:39.407625 132092006360896 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0423 19:11:39.407739 132092006360896 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0423 19:11:39.776115 132092006360896 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0423 19:11:39.776215 132092006360896 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0423 19:11:39.792055 132092006360896 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0423 19:11:39.792115 132092006360896 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0423 19:11:39.814985 132092006360896 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 19:11:39.815054 132092006360896 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 19:11:39.830830 132092006360896 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 19:11:39.830893 132092006360896 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 19:11:39.846605 132092006360896 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 19:11:39.846678 132092006360896 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 19:11:39.871240 132092006360896 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0423 19:11:39.871311 132092006360896 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 19:11:39.890693 132092006360896 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0423 19:11:39.890759 132092006360896 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0423 19:11:40.078815 132092006360896 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0423 19:11:40.078929 132092006360896 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0423 19:11:41.515305 132092006360896 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0423 19:11:41.515435 132092006360896 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 19:11:41.515490 132092006360896 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 19:11:41.515548 132092006360896 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 19:11:41.515600 132092006360896 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 19:11:41.515637 132092006360896 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 19:11:41.515702 132092006360896 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 19:11:41.515756 132092006360896 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 19:11:41.515796 132092006360896 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 19:11:41.515831 132092006360896 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 19:11:41.515877 132092006360896 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0423 19:11:41.515923 132092006360896 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0423 19:11:42.021299 132092006360896 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0423 19:11:42.021395 132092006360896 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0423 19:11:42.036798 132092006360896 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0423 19:11:42.036856 132092006360896 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0423 19:11:52.647777 132092006360896 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. I0423 19:11:52.648609 132092006360896 metric_logger.py:289] number parameters: 1.104 billion I0423 19:12:03.748912 132092006360896 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0423 19:12:04.159741 132092006360896 checkpointing.py:776] Waited 0.4108099937438965 seconds for step 0 to finish before starting checkpointing. I0423 19:12:04.162361 132092006360896 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0423 19:12:04.164014 132092006360896 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0423 19:12:04.166893 132092006360896 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_02_synthetic/checkpoints/0. I0423 19:12:04.761881 132092006360896 signaling_client.py:364] Using JaxDistributedSignalingClient I0423 19:12:04.763093 132092006360896 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0423 19:12:04.763149 132092006360896 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 19:12:05.021027 132092006360896 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.259223s I0423 19:12:05.021209 132092006360896 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.992 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5155646800994873 s) (per-host) I0423 19:12:05.021262 132092006360896 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.515629s (batch_requests_ready=0.252700s, total_serialization_initiated=0.262851s, others=0.000078s) I0423 19:12:05.021355 132092006360896 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.519869s (all_items=0.000020s, per_item={'items': '0.00001955'}, temp_paths=0.519850) I0423 19:12:05.022366 131968775747328 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0423 19:12:05.022541 132092006360896 async_checkpointer.py:561] Finished blocking save. Time taken: 0.858456s. 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_02_synthetic/checkpoints/0. I0423 19:12:05.470975 132092006360896 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0423 19:12:05.471339 131964455601920 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0423 19:12:05.471504 132092006360896 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_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776971524.1623433, 'wait_for_prev_duration_secs': 6.437301635742188e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776971524.1640515, 'checkpointer_blocking_duration_secs': 0.8586087226867676, 'get_old_steps_start_time': 1776971525.0226831, 'get_old_steps_duration_secs': 2.6941299438476562e-05, 'checkpoint_manager_blocking_start_time': 1776971524.1603816, 'checkpoint_manager_blocking_duration_secs': 1.3110814094543457} I0423 19:12:05.471660 132092006360896 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0423 19:12:05.471714 132092006360896 max_utils.py:750] Memstats: After params initialized: I0423 19:12:05.471772 132092006360896 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0)) I0423 19:12:05.471805 132092006360896 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0)) I0423 19:12:05.471850 132092006360896 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0)) I0423 19:12:05.471874 132092006360896 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0)) I0423 19:12:05.736533 132092006360896 metric_logger.py:185] completed step: 0, seconds: 11.100, TFLOP/s/device: 1.224, Tokens/s/device: 184.502, total_weights: 65536, loss: 10.874 I0423 19:12:05.901716 132092006360896 metric_logger.py:185] completed step: 1, seconds: 1.986, TFLOP/s/device: 6.840, Tokens/s/device: 1031.060, total_weights: 65536, loss: 10.874 I0423 19:12:06.306313 132092006360896 metric_logger.py:185] completed step: 2, seconds: 0.018, TFLOP/s/device: 763.837, Tokens/s/device: 115133.798, total_weights: 65536, loss: 10.263 I0423 19:12:06.453765 132092006360896 metric_logger.py:185] completed step: 3, seconds: 0.405, TFLOP/s/device: 33.519, Tokens/s/device: 5052.312, total_weights: 65536, loss: 9.734 I0423 19:12:06.749192 132092006360896 metric_logger.py:185] completed step: 4, seconds: 0.152, TFLOP/s/device: 89.099, Tokens/s/device: 13430.037, total_weights: 65536, loss: 9.278 I0423 19:12:06.754905 132092006360896 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.159, Tokens/s/device: 13891.244, total_weights: 65536, loss: 8.892 I0423 19:12:26.063513 132092006360896 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.931, Tokens/s/device: 6923.152, total_weights: 65536, loss: 8.592 I0423 19:12:26.210988 132092006360896 metric_logger.py:185] completed step: 7, seconds: 19.162, TFLOP/s/device: 0.709, Tokens/s/device: 106.877, total_weights: 65536, loss: 8.384 I0423 19:12:26.358450 132092006360896 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.494, Tokens/s/device: 13489.570, total_weights: 65536, loss: 8.255 I0423 19:12:26.505598 132092006360896 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0423 19:12:26.506280 132092006360896 checkpointing.py:776] Waited 0.0007021427154541016 seconds for step 9 to finish before starting checkpointing. I0423 19:12:26.508300 132092006360896 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0423 19:12:26.950114 2676 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0423 19:12:28.897154 131969262262016 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_02_synthetic/checkpoints/0/items/array_metadatas/process_5 I0423 19:13:01.493613 131968775747328 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 27.718 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 56.98792886734009 s) (per-host) I0423 19:13:01.493744 131968775747328 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 56.471265s. I0423 19:13:08.042690 131968775747328 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 63.020178s. I0423 19:13:08.042912 131964455601920 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0423 19:13:08.042984 131964455601920 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0423 19:13:08.043073 131964455601920 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0423 19:13:08.044685 131964455601920 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0423 19:13:08.044856 132092006360896 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 19:13:08.045008 132092006360896 checkpoint_manager.py:1458] Waiting for previous save to complete took 41.536711 seconds. If this number is high, consider checkpointing less frequently. I0423 19:13:08.047077 132092006360896 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0423 19:13:08.050671 132092006360896 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_02_synthetic/checkpoints/9. I0423 19:13:08.612715 132092006360896 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0423 19:13:08.612814 132092006360896 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 19:13:08.657323 132092006360896 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.045757s I0423 19:13:08.657488 132092006360896 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.135 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3003983497619629 s) (per-host) I0423 19:13:08.657542 132092006360896 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.300462s (batch_requests_ready=0.252805s, total_serialization_initiated=0.047585s, others=0.000071s) I0423 19:13:08.657633 132092006360896 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.304546s (all_items=0.000015s, per_item={'items': '0.00001478'}, temp_paths=0.304531) I0423 19:13:08.658490 131965017650944 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0423 19:13:08.658659 132092006360896 async_checkpointer.py:561] Finished blocking save. Time taken: 0.611503s. 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_02_synthetic/checkpoints/9. I0423 19:13:08.672337 132092006360896 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0423 19:13:08.672618 131964455601920 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0423 19:13:08.672796 132092006360896 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_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776971546.5082715, 'wait_for_prev_duration_secs': 41.53671050071716, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776971588.0471177, 'checkpointer_blocking_duration_secs': 0.6116480827331543, 'get_old_steps_start_time': 1776971588.6587837, 'get_old_steps_duration_secs': 2.47955322265625e-05, 'checkpoint_manager_blocking_start_time': 1776971546.5065253, 'checkpoint_manager_blocking_duration_secs': 42.16623663902283} I0423 19:13:08.672909 132092006360896 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0423 19:13:08.672955 132092006360896 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0423 19:13:13.389342 131965496649472 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_02_synthetic/checkpoints/9/items/array_metadatas/process_5 I0423 19:13:51.013481 131965017650944 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.031 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.65634489059448 s) (per-host) I0423 19:13:51.013588 131965017650944 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.354984s. I0423 19:14:00.180484 131965017650944 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 51.521863s. I0423 19:14:00.180770 131964455601920 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0423 19:14:00.180888 131964455601920 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0423 19:14:00.180937 131964455601920 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0423 19:14:00.182463 131964455601920 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0423 19:14:00.182640 132092006360896 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 19:14:00.182795 132092006360896 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0423 19:14:00.183417 132092006360896 metric_logger.py:185] completed step: 9, seconds: 0.148, TFLOP/s/device: 92.081, Tokens/s/device: 13879.476, 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: Thu Apr 23 19:14:13 UTC 2026 EXIT_CODE=0