XPK Start: Fri Apr 24 18:15:58 UTC 2026 2026-04-24 18:16:02.125725: 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 00:00:1777054562.138511 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1777054562.142189 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1777054562.153446 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777054562.153463 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777054562.153466 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777054562.153468 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-24 18:16:21.255822: 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 18:16:21.765497 139555569948480 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-24 18:16:30,805:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0424 18:16:30.805367 139555569948480 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-24 18:16:30,807:jax._src.distributed:157: Connecting to JAX distributed service on mt-07-eval-qvzxb-slice-job-0-0.mt-07-eval-qvzxb:8482 I0424 18:16:30.807683 139555569948480 distributed.py:157] Connecting to JAX distributed service on mt-07-eval-qvzxb-slice-job-0-0.mt-07-eval-qvzxb:8482 I0424 18:16:32.435072 139555569948480 max_utils.py:284] Jax distributed system initialized! I0424 18:16:39.437020 139555569948480 max_utils.py:800] System Information: Jax Version: 0.8.1 I0424 18:16:39.437126 139555569948480 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0424 18:16:39.437176 139555569948480 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 18:16:39.437215 139555569948480 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0424 18:16:40.061741 139555569948480 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0424 18:16:40.062028 139555569948480 checkpointing.py:677] Setting up checkpoint logger... I0424 18:16:40.062088 139555569948480 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0424 18:16:40.062134 139555569948480 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0424 18:16:40.062474 139555569948480 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 0x7eec53d78da0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0424 18:16:43.024073 139555569948480 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0424 18:16:43.024308 139555569948480 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 0x7eeb0ae4a120>}, handler_registry=None I0424 18:16:43.024540 139555569948480 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7eeb0ae4a120>` 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 18:16:43.024591 139555569948480 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ed941cb3320>` 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 18:16:43.024627 139555569948480 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 0x7eeb0ae4a120>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7eeb0ae4a120>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ed941cb3320>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ed941cb3320>}). I0424 18:16:43.025134 139555569948480 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0424 18:16:43.025214 139555569948480 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7ed7cc2c2ac0> timeout: 600 secs and primary_host=0 for async checkpoint writes I0424 18:16:43.739687 139555569948480 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_07_eval/checkpoints I0424 18:16:43.741931 139555569948480 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_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7ed941caf770> I0424 18:16:43.742049 139555569948480 checkpointing.py:301] Checkpoint manager created! I0424 18:16:44.634886 139555569948480 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0424 18:16:44.634999 139555569948480 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0424 18:16:45.002033 139555569948480 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0424 18:16:45.002132 139555569948480 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0424 18:16:45.017857 139555569948480 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0424 18:16:45.017913 139555569948480 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0424 18:16:45.040791 139555569948480 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 18:16:45.040861 139555569948480 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:16:45.056823 139555569948480 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 18:16:45.056891 139555569948480 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:16:45.072615 139555569948480 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 18:16:45.072687 139555569948480 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:16:45.097387 139555569948480 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0424 18:16:45.097453 139555569948480 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:16:45.116692 139555569948480 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0424 18:16:45.116756 139555569948480 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0424 18:16:45.305455 139555569948480 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0424 18:16:45.305565 139555569948480 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0424 18:16:46.734848 139555569948480 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0424 18:16:46.734989 139555569948480 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 18:16:46.735052 139555569948480 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 18:16:46.735112 139555569948480 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 18:16:46.735164 139555569948480 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 18:16:46.735203 139555569948480 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 18:16:46.735255 139555569948480 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 18:16:46.735308 139555569948480 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 18:16:46.735351 139555569948480 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 18:16:46.735389 139555569948480 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 18:16:46.735437 139555569948480 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0424 18:16:46.735486 139555569948480 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0424 18:16:47.243954 139555569948480 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0424 18:16:47.244056 139555569948480 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0424 18:16:47.258930 139555569948480 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0424 18:16:47.258989 139555569948480 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0424 18:16:58.047472 139555569948480 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 18:16:58.048316 139555569948480 metric_logger.py:289] number parameters: 1.104 billion I0424 18:17:09.039492 139555569948480 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0424 18:17:09.217480 139555569948480 checkpointing.py:776] Waited 0.17797112464904785 seconds for step 0 to finish before starting checkpointing. I0424 18:17:09.219971 139555569948480 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0424 18:17:09.221740 139555569948480 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0424 18:17:09.224472 139555569948480 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_07_eval/checkpoints/0. I0424 18:17:09.752990 139555569948480 signaling_client.py:364] Using JaxDistributedSignalingClient I0424 18:17:09.754198 139555569948480 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0424 18:17:09.754255 139555569948480 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 18:17:10.002388 139555569948480 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.249483s I0424 18:17:10.002560 139555569948480 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.078 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5011658668518066 s) (per-host) I0424 18:17:10.002614 139555569948480 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.501230s (batch_requests_ready=0.248118s, total_serialization_initiated=0.253040s, others=0.000071s) I0424 18:17:10.002722 139555569948480 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.505305s (all_items=0.000019s, per_item={'items': '0.00001931'}, temp_paths=0.505286) I0424 18:17:10.003723 139430181381888 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0424 18:17:10.003897 139555569948480 async_checkpointer.py:561] Finished blocking save. Time taken: 0.782088s. 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_07_eval/checkpoints/0. I0424 18:17:10.218486 139555569948480 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0424 18:17:10.218871 139427523327744 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0424 18:17:10.219017 139555569948480 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_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777054629.219953, 'wait_for_prev_duration_secs': 6.365776062011719e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777054629.2217774, 'checkpointer_blocking_duration_secs': 0.7822301387786865, 'get_old_steps_start_time': 1777054630.0040278, 'get_old_steps_duration_secs': 2.6941299438476562e-05, 'checkpoint_manager_blocking_start_time': 1777054629.218086, 'checkpoint_manager_blocking_duration_secs': 1.000887155532837} I0424 18:17:10.219171 139555569948480 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0424 18:17:10.219224 139555569948480 max_utils.py:750] Memstats: After params initialized: I0424 18:17:10.219282 139555569948480 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0)) I0424 18:17:10.219313 139555569948480 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0)) I0424 18:17:10.219343 139555569948480 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0)) I0424 18:17:10.219367 139555569948480 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0)) I0424 18:17:10.483401 139555569948480 metric_logger.py:185] completed step: 0, seconds: 10.991, TFLOP/s/device: 1.236, Tokens/s/device: 186.333, total_weights: 65536, loss: 10.874 I0424 18:17:10.654316 139555569948480 metric_logger.py:185] completed step: 1, seconds: 1.443, TFLOP/s/device: 9.418, Tokens/s/device: 1419.633, total_weights: 65536, loss: 10.874 I0424 18:17:11.077024 139555569948480 metric_logger.py:185] completed step: 2, seconds: 0.024, TFLOP/s/device: 573.297, Tokens/s/device: 86413.502, total_weights: 65536, loss: 10.263 I0424 18:17:11.224519 139555569948480 metric_logger.py:185] completed step: 3, seconds: 0.423, TFLOP/s/device: 32.109, Tokens/s/device: 4839.788, total_weights: 65536, loss: 9.734 I0424 18:17:11.519856 139555569948480 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 89.072, Tokens/s/device: 13425.899, total_weights: 65536, loss: 9.278 I0424 18:17:11.525057 139555569948480 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.164, Tokens/s/device: 13891.904, total_weights: 65536, loss: 8.892 I0424 18:17:30.038063 139555569948480 metric_logger.py:185] completed step: 6, seconds: 0.295, TFLOP/s/device: 45.999, Tokens/s/device: 6933.465, total_weights: 65536, loss: 8.592 I0424 18:17:30.185575 139555569948480 metric_logger.py:185] completed step: 7, seconds: 18.366, TFLOP/s/device: 0.740, Tokens/s/device: 111.509, total_weights: 65536, loss: 8.384 I0424 18:17:30.333052 139555569948480 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.304, Tokens/s/device: 13460.932, total_weights: 65536, loss: 8.255 I0424 18:17:30.480103 139555569948480 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0424 18:17:30.480765 139555569948480 checkpointing.py:776] Waited 0.0006818771362304688 seconds for step 9 to finish before starting checkpointing. I0424 18:17:30.482785 139555569948480 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0424 18:17:31.944200 2589 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0424 18:17:33.904476 139430189774592 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_07_eval/checkpoints/0/items/array_metadatas/process_5 I0424 18:18:06.079579 139430181381888 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 27.919 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 56.578147888183594 s) (per-host) I0424 18:18:06.079696 139430181381888 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 56.075856s. I0424 18:18:13.868349 139430181381888 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 63.864495s. I0424 18:18:13.868633 139427523327744 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0424 18:18:13.868713 139427523327744 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0424 18:18:13.868760 139427523327744 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0424 18:18:13.870575 139427523327744 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0424 18:18:13.870756 139555569948480 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 18:18:13.870897 139555569948480 checkpoint_manager.py:1458] Waiting for previous save to complete took 43.388116 seconds. If this number is high, consider checkpointing less frequently. I0424 18:18:13.872760 139555569948480 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0424 18:18:13.876145 139555569948480 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_07_eval/checkpoints/9. I0424 18:18:14.406840 139555569948480 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0424 18:18:14.406939 139555569948480 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 18:18:14.452360 139555569948480 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.046764s I0424 18:18:14.452532 139555569948480 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.154 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2992885112762451 s) (per-host) I0424 18:18:14.452585 139555569948480 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.299352s (batch_requests_ready=0.250669s, total_serialization_initiated=0.048611s, others=0.000071s) I0424 18:18:14.452692 139555569948480 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.303416s (all_items=0.000016s, per_item={'items': '0.00001597'}, temp_paths=0.303400) I0424 18:18:14.453559 139428547589888 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0424 18:18:14.453697 139555569948480 async_checkpointer.py:561] Finished blocking save. Time taken: 0.580866s. 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_07_eval/checkpoints/9. I0424 18:18:14.469551 139555569948480 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0424 18:18:14.469883 139427523327744 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0424 18:18:14.470072 139555569948480 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_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777054650.4827566, 'wait_for_prev_duration_secs': 43.38811635971069, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777054693.8727984, 'checkpointer_blocking_duration_secs': 0.5810673236846924, 'get_old_steps_start_time': 1777054694.4538841, 'get_old_steps_duration_secs': 2.8133392333984375e-05, 'checkpoint_manager_blocking_start_time': 1777054650.4809868, 'checkpoint_manager_blocking_duration_secs': 43.98905038833618} I0424 18:18:14.470178 139555569948480 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0424 18:18:14.470223 139555569948480 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0424 18:18:19.412173 139430189774592 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_07_eval/checkpoints/9/items/array_metadatas/process_5 I0424 18:18:56.361890 139428547589888 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.424 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.208603382110596 s) (per-host) I0424 18:18:56.362008 139428547589888 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.908233s. I0424 18:19:02.919355 139428547589888 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 48.465567s. I0424 18:19:02.919622 139427523327744 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0424 18:19:02.919803 139427523327744 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0424 18:19:02.919852 139427523327744 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0424 18:19:02.921417 139427523327744 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0424 18:19:02.921576 139555569948480 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 18:19:02.921725 139555569948480 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0424 18:19:02.922369 139555569948480 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.189, Tokens/s/device: 13895.768, 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 18:19:13 UTC 2026 EXIT_CODE=0