XPK Start: Wed Apr 22 19:13:37 UTC 2026 2026-04-22 19:13:41.952323: 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 00:00:1776885221.965414 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776885221.969426 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776885221.980797 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776885221.980816 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776885221.980818 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776885221.980820 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-22 19:14:01.211557: 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) I0422 19:14:01.738446 138566671361856 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-22 19:14:10,780:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0422 19:14:10.780930 138566671361856 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-22 19:14:10,783:jax._src.distributed:157: Connecting to JAX distributed service on mt-03-dropout-js8o1-slice-job-0-0.mt-03-dropout-js8o1:8482 I0422 19:14:10.783384 138566671361856 distributed.py:157] Connecting to JAX distributed service on mt-03-dropout-js8o1-slice-job-0-0.mt-03-dropout-js8o1:8482 I0422 19:14:11.986660 138566671361856 max_utils.py:284] Jax distributed system initialized! I0422 19:14:18.830699 138566671361856 max_utils.py:800] System Information: Jax Version: 0.8.1 I0422 19:14:18.830804 138566671361856 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0422 19:14:18.830844 138566671361856 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 I0422 19:14:18.830878 138566671361856 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0422 19:14:19.472785 138566671361856 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0422 19:14:19.473091 138566671361856 checkpointing.py:677] Setting up checkpoint logger... I0422 19:14:19.473165 138566671361856 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0422 19:14:19.473208 138566671361856 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0422 19:14:19.473572 138566671361856 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 0x7e0607d71160>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0422 19:14:22.505803 138566671361856 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0422 19:14:22.505995 138566671361856 checkpoint_manager.py:709] [process=6][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7df39f851be0>}, handler_registry=None I0422 19:14:22.506242 138566671361856 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7df39f851be0>` 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`. I0422 19:14:22.506292 138566671361856 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7df404452960>` 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`. I0422 19:14:22.506328 138566671361856 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 0x7df39f851be0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7df39f851be0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7df404452960>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7df404452960>}). I0422 19:14:22.506838 138566671361856 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0422 19:14:22.506917 138566671361856 async_checkpointer.py:177] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7df39f112840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0422 19:14:24.015702 138566671361856 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_03_dropout/checkpoints I0422 19:14:24.019910 138566671361856 checkpoint_manager.py:929] [process=6][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_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_03_dropout/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7df39f848170> I0422 19:14:24.020028 138566671361856 checkpointing.py:301] Checkpoint manager created! I0422 19:14:24.335307 138566671361856 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0422 19:14:24.335413 138566671361856 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0422 19:14:24.713115 138566671361856 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 19:14:24.713208 138566671361856 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0422 19:14:24.729366 138566671361856 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 19:14:24.729428 138566671361856 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0422 19:14:24.752697 138566671361856 attentions.py:1154] attentions/query Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 19:14:24.752766 138566671361856 attentions.py:1154] attentions/query Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0422 19:14:24.768957 138566671361856 attentions.py:1155] attentions/key Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 19:14:24.769021 138566671361856 attentions.py:1155] attentions/key Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0422 19:14:24.785171 138566671361856 attentions.py:1156] attentions/value Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 19:14:24.785234 138566671361856 attentions.py:1156] attentions/value Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0422 19:14:24.809493 138566671361856 attentions.py:1197] attentions/out Logical: bfloat16[32,128,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0422 19:14:24.809563 138566671361856 attentions.py:1197] attentions/out Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0422 19:14:24.835931 138566671361856 linears.py:525] linears/x Logical: bfloat16[32,128,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0422 19:14:24.836003 138566671361856 linears.py:525] linears/x Physical: bfloat16[32,128,7168]....................................... ('fsdp', None, None). I0422 19:14:25.039409 138566671361856 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0422 19:14:25.039535 138566671361856 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0422 19:14:26.486918 138566671361856 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0422 19:14:26.487048 138566671361856 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) I0422 19:14:26.487119 138566671361856 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) I0422 19:14:26.487192 138566671361856 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') I0422 19:14:26.487248 138566671361856 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) I0422 19:14:26.487287 138566671361856 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) I0422 19:14:26.487344 138566671361856 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) I0422 19:14:26.487396 138566671361856 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') I0422 19:14:26.487436 138566671361856 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) I0422 19:14:26.487473 138566671361856 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) I0422 19:14:26.487520 138566671361856 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0422 19:14:26.487570 138566671361856 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0422 19:14:27.030339 138566671361856 train.py:155] train/xent Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0422 19:14:27.030448 138566671361856 train.py:155] train/xent Physical: float32[32,128]............................................. ('fsdp', None). I0422 19:14:27.045560 138566671361856 train.py:162] train/z_loss Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0422 19:14:27.045627 138566671361856 train.py:162] train/z_loss Physical: float32[32,128]............................................. ('fsdp', None). I0422 19:14:30.446655 138566671361856 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. I0422 19:14:30.447500 138566671361856 metric_logger.py:289] number parameters: 1.104 billion I0422 19:14:34.483047 138566671361856 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0422 19:14:34.555449 138566671361856 checkpointing.py:776] Waited 0.07239174842834473 seconds for step 0 to finish before starting checkpointing. I0422 19:14:34.558026 138566671361856 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 19:14:34.559905 138566671361856 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 0 I0422 19:14:34.562818 138566671361856 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_03_dropout/checkpoints/0. I0422 19:14:35.549591 138566671361856 signaling_client.py:364] Using JaxDistributedSignalingClient I0422 19:14:35.550496 138566671361856 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0422 19:14:35.550556 138566671361856 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 I0422 19:14:35.794896 138566671361856 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.245385s I0422 19:14:35.795072 138566671361856 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.980 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5176749229431152 s) (per-host) I0422 19:14:35.795145 138566671361856 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.517756s (batch_requests_ready=0.268775s, total_serialization_initiated=0.248891s, others=0.000091s) I0422 19:14:35.795242 138566671361856 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.521835s (all_items=0.000020s, per_item={'items': '0.00002027'}, temp_paths=0.521815) I0422 19:14:35.796184 138448488863488 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0422 19:14:35.796363 138566671361856 async_checkpointer.py:561] Finished blocking save. Time taken: 1.236388s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_03_dropout/checkpoints/0. I0422 19:14:36.206819 138566671361856 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0422 19:14:36.207238 138442046412544 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0422 19:14:36.207385 138566671361856 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776885274.5580075, 'wait_for_prev_duration_secs': 6.580352783203125e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776885274.5599446, 'checkpointer_blocking_duration_secs': 1.2365241050720215, 'get_old_steps_start_time': 1776885275.7964885, 'get_old_steps_duration_secs': 2.6226043701171875e-05, 'checkpoint_manager_blocking_start_time': 1776885274.5559864, 'checkpoint_manager_blocking_duration_secs': 1.6513586044311523} I0422 19:14:36.207560 138566671361856 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0422 19:14:36.207614 138566671361856 max_utils.py:750] Memstats: After params initialized: I0422 19:14:36.207670 138566671361856 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_24(process=6,(0,6,0,0)) I0422 19:14:36.207704 138566671361856 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_25(process=6,(1,6,0,0)) I0422 19:14:36.207732 138566671361856 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_28(process=6,(0,7,0,0)) I0422 19:14:36.207758 138566671361856 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_29(process=6,(1,7,0,0)) I0422 19:14:36.470645 138566671361856 metric_logger.py:185] completed step: 0, seconds: 4.035, TFLOP/s/device: 0.198, Tokens/s/device: 31.719, total_weights: 4096, loss: 10.889 I0422 19:14:36.575586 138566671361856 metric_logger.py:185] completed step: 1, seconds: 1.986, TFLOP/s/device: 0.403, Tokens/s/device: 64.443, total_weights: 4096, loss: 10.902 I0422 19:14:36.976497 138566671361856 metric_logger.py:185] completed step: 2, seconds: 0.035, TFLOP/s/device: 22.905, Tokens/s/device: 3660.804, total_weights: 4096, loss: 9.956 I0422 19:14:37.047177 138566671361856 metric_logger.py:185] completed step: 3, seconds: 0.401, TFLOP/s/device: 1.997, Tokens/s/device: 319.127, total_weights: 4096, loss: 9.139 I0422 19:14:37.189008 138566671361856 metric_logger.py:185] completed step: 4, seconds: 0.076, TFLOP/s/device: 10.576, Tokens/s/device: 1690.349, total_weights: 4096, loss: 8.458 I0422 19:14:37.194859 138566671361856 metric_logger.py:185] completed step: 5, seconds: 0.071, TFLOP/s/device: 11.354, Tokens/s/device: 1814.702, total_weights: 4096, loss: 7.924 I0422 19:14:46.471302 138566671361856 metric_logger.py:185] completed step: 6, seconds: 0.142, TFLOP/s/device: 5.623, Tokens/s/device: 898.643, total_weights: 4096, loss: 7.553 I0422 19:14:46.542544 138566671361856 metric_logger.py:185] completed step: 7, seconds: 9.207, TFLOP/s/device: 0.087, Tokens/s/device: 13.902, total_weights: 4096, loss: 7.267 I0422 19:14:46.613265 138566671361856 metric_logger.py:185] completed step: 8, seconds: 0.075, TFLOP/s/device: 10.743, Tokens/s/device: 1716.968, total_weights: 4096, loss: 7.111 I0422 19:14:46.683850 138566671361856 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0422 19:14:46.684472 138566671361856 checkpointing.py:776] Waited 0.0006401538848876953 seconds for step 9 to finish before starting checkpointing. I0422 19:14:46.686387 138566671361856 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 19:14:47.690706 2669 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0422 19:14:50.478479 138448497256192 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_03_dropout/checkpoints/0/items/array_metadatas/process_6 I0422 19:15:17.093009 138448488863488 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 37.775 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.815571308135986 s) (per-host) I0422 19:15:17.093137 138448488863488 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.296832s. I0422 19:15:26.345733 138448488863488 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 50.549412s. I0422 19:15:26.346034 138442046412544 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 19:15:26.346194 138442046412544 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0422 19:15:26.346250 138442046412544 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0422 19:15:26.347844 138442046412544 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0422 19:15:26.348020 138566671361856 checkpoint_manager.py:2036] [process=6][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0422 19:15:26.348162 138566671361856 checkpoint_manager.py:1458] Waiting for previous save to complete took 39.661779 seconds. If this number is high, consider checkpointing less frequently. I0422 19:15:26.350154 138566671361856 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 9 I0422 19:15:26.353513 138566671361856 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_03_dropout/checkpoints/9. I0422 19:15:26.886519 138566671361856 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0422 19:15:26.886621 138566671361856 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 I0422 19:15:26.930952 138566671361856 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.045746s I0422 19:15:26.931133 138566671361856 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.213 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2959158420562744 s) (per-host) I0422 19:15:26.931187 138566671361856 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.295979s (batch_requests_ready=0.248327s, total_serialization_initiated=0.047581s, others=0.000071s) I0422 19:15:26.931276 138566671361856 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.299981s (all_items=0.000017s, per_item={'items': '0.00001669'}, temp_paths=0.299964) I0422 19:15:26.932137 138448497256192 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0422 19:15:26.932272 138566671361856 async_checkpointer.py:561] Finished blocking save. Time taken: 0.582040s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_03_dropout/checkpoints/9. I0422 19:15:26.943895 138566671361856 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0422 19:15:26.944147 138442046412544 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0422 19:15:26.944291 138566671361856 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776885286.6863575, 'wait_for_prev_duration_secs': 39.661779165267944, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776885326.3501954, 'checkpointer_blocking_duration_secs': 0.5822768211364746, 'get_old_steps_start_time': 1776885326.932491, 'get_old_steps_duration_secs': 2.7179718017578125e-05, 'checkpoint_manager_blocking_start_time': 1776885286.6846993, 'checkpoint_manager_blocking_duration_secs': 40.259546995162964} I0422 19:15:26.944396 138566671361856 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0422 19:15:26.944442 138566671361856 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 19:15:31.744973 138414397507328 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_03_dropout/checkpoints/9/items/array_metadatas/process_6 I0422 19:16:05.907306 138448497256192 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 40.222 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 39.27204775810242 s) (per-host) I0422 19:16:05.907427 138448497256192 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 38.975049s. I0422 19:16:16.046842 138448497256192 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 49.114449s. I0422 19:16:16.047080 138442046412544 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 19:16:16.047161 138442046412544 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0422 19:16:16.047224 138442046412544 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0422 19:16:16.049500 138442046412544 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0422 19:16:16.049734 138566671361856 checkpoint_manager.py:2036] [process=6][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0422 19:16:16.049881 138566671361856 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 19:16:16.050523 138566671361856 metric_logger.py:185] completed step: 9, seconds: 0.071, TFLOP/s/device: 11.264, Tokens/s/device: 1800.256, 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: Wed Apr 22 19:16:28 UTC 2026 EXIT_CODE=0