XPK Start: Mon Apr 20 22:38:51 UTC 2026 2026-04-20 22:38:55.616392: 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 00:00:1776724735.629062 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776724735.632733 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776724735.644019 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776724735.644038 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776724735.644041 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776724735.644043 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-20 22:39:14.647941: 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) I0420 22:39:15.161272 134835206932288 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-20 22:39:24,201:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0420 22:39:24.201868 134835206932288 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-20 22:39:24,204:jax._src.distributed:157: Connecting to JAX distributed service on mt-03-dropout-gw133-slice-job-0-0.mt-03-dropout-gw133:8482 I0420 22:39:24.204258 134835206932288 distributed.py:157] Connecting to JAX distributed service on mt-03-dropout-gw133-slice-job-0-0.mt-03-dropout-gw133:8482 I0420 22:39:25.585523 134835206932288 max_utils.py:284] Jax distributed system initialized! I0420 22:39:32.580890 134835206932288 max_utils.py:800] System Information: Jax Version: 0.8.1 I0420 22:39:32.581005 134835206932288 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0420 22:39:32.581057 134835206932288 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 I0420 22:39:32.581109 134835206932288 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0420 22:39:33.196319 134835206932288 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0420 22:39:33.196624 134835206932288 checkpointing.py:677] Setting up checkpoint logger... I0420 22:39:33.196681 134835206932288 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0420 22:39:33.196727 134835206932288 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0420 22:39:33.197077 134835206932288 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 0x7aa13db74f80>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0420 22:39:36.540899 134835206932288 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0420 22:39:36.541244 134835206932288 checkpoint_manager.py:709] [process=1][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a8d4b28af30>}, handler_registry=None I0420 22:39:36.541496 134835206932288 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a8d4b28af30>` 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`. I0420 22:39:36.541545 134835206932288 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a8ccc64fe60>` 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`. I0420 22:39:36.541581 134835206932288 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 0x7a8d4b28af30>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a8d4b28af30>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a8ccc64fe60>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a8ccc64fe60>}). I0420 22:39:36.542088 134835206932288 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0420 22:39:36.542168 134835206932288 async_checkpointer.py:177] [process=1][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7a8ccc59a840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0420 22:39:37.742790 134835206932288 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_03_dropout/checkpoints I0420 22:39:38.587941 134835206932288 checkpoint_manager.py:929] [process=1][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_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_03_dropout/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7a8ccc64c9e0> I0420 22:39:38.588115 134835206932288 checkpointing.py:301] Checkpoint manager created! I0420 22:39:38.901965 134835206932288 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0420 22:39:38.902067 134835206932288 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0420 22:39:39.272655 134835206932288 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0420 22:39:39.272750 134835206932288 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0420 22:39:39.288665 134835206932288 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0420 22:39:39.288729 134835206932288 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0420 22:39:39.311814 134835206932288 attentions.py:1154] attentions/query Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 22:39:39.311883 134835206932288 attentions.py:1154] attentions/query Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0420 22:39:39.328129 134835206932288 attentions.py:1155] attentions/key Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 22:39:39.328194 134835206932288 attentions.py:1155] attentions/key Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0420 22:39:39.344306 134835206932288 attentions.py:1156] attentions/value Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 22:39:39.344366 134835206932288 attentions.py:1156] attentions/value Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0420 22:39:39.368642 134835206932288 attentions.py:1197] attentions/out Logical: bfloat16[32,128,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0420 22:39:39.368713 134835206932288 attentions.py:1197] attentions/out Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0420 22:39:39.394395 134835206932288 linears.py:525] linears/x Logical: bfloat16[32,128,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0420 22:39:39.394472 134835206932288 linears.py:525] linears/x Physical: bfloat16[32,128,7168]....................................... ('fsdp', None, None). I0420 22:39:39.595619 134835206932288 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0420 22:39:39.595727 134835206932288 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0420 22:39:41.057579 134835206932288 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0420 22:39:41.057709 134835206932288 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) I0420 22:39:41.057764 134835206932288 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) I0420 22:39:41.057824 134835206932288 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') I0420 22:39:41.057877 134835206932288 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) I0420 22:39:41.057917 134835206932288 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) I0420 22:39:41.057972 134835206932288 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) I0420 22:39:41.058026 134835206932288 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') I0420 22:39:41.058066 134835206932288 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) I0420 22:39:41.058104 134835206932288 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) I0420 22:39:41.058152 134835206932288 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0420 22:39:41.058206 134835206932288 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0420 22:39:41.590905 134835206932288 train.py:155] train/xent Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0420 22:39:41.591001 134835206932288 train.py:155] train/xent Physical: float32[32,128]............................................. ('fsdp', None). I0420 22:39:41.606138 134835206932288 train.py:162] train/z_loss Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0420 22:39:41.606201 134835206932288 train.py:162] train/z_loss Physical: float32[32,128]............................................. ('fsdp', None). I0420 22:39:44.963470 134835206932288 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. I0420 22:39:44.964301 134835206932288 metric_logger.py:289] number parameters: 1.104 billion I0420 22:39:48.974189 134835206932288 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0420 22:39:49.141809 134835206932288 checkpointing.py:776] Waited 0.1676044464111328 seconds for step 0 to finish before starting checkpointing. I0420 22:39:49.144417 134835206932288 checkpoint_manager.py:2013] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0420 22:39:49.146376 134835206932288 checkpoint_manager.py:1518] [process=1] Saving checkpoint at step 0 I0420 22:39:49.149177 134835206932288 async_checkpointer.py:452] [process=1] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_03_dropout/checkpoints/0. I0420 22:39:50.138871 134835206932288 signaling_client.py:364] Using JaxDistributedSignalingClient I0420 22:39:50.139823 134835206932288 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0420 22:39:50.139884 134835206932288 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 I0420 22:39:50.383198 134835206932288 base_pytree_checkpoint_handler.py:153] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.244405s I0420 22:39:50.383370 134835206932288 base_pytree_checkpoint_handler.py:129] [process=1] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.993 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5153367519378662 s) (per-host) I0420 22:39:50.383438 134835206932288 base_pytree_checkpoint_handler.py:737] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.515412s (batch_requests_ready=0.267374s, total_serialization_initiated=0.247956s, others=0.000083s) I0420 22:39:50.383533 134835206932288 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.519456s (all_items=0.000021s, per_item={'items': '0.00002098'}, temp_paths=0.519435) I0420 22:39:50.384461 134712043497216 async_checkpointer.py:79] [process=1][thread=async_save] Background save thread started. I0420 22:39:50.384618 134835206932288 async_checkpointer.py:561] Finished blocking save. Time taken: 1.238163s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_03_dropout/checkpoints/0. I0420 22:39:50.802615 134835206932288 checkpoint_manager.py:1566] [process=1][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0420 22:39:50.803006 134706666395392 async_checkpointer.py:265] [process=1][thread=save_finalize] Waiting for background save thread=async_save. I0420 22:39:50.803174 134835206932288 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776724789.1443837, 'wait_for_prev_duration_secs': 7.939338684082031e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776724789.1464274, 'checkpointer_blocking_duration_secs': 1.238293170928955, 'get_old_steps_start_time': 1776724790.3847413, 'get_old_steps_duration_secs': 2.5987625122070312e-05, 'checkpoint_manager_blocking_start_time': 1776724789.1423652, 'checkpoint_manager_blocking_duration_secs': 1.6607661247253418} I0420 22:39:50.803285 134835206932288 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0420 22:39:50.803339 134835206932288 max_utils.py:750] Memstats: After params initialized: I0420 22:39:50.803411 134835206932288 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_2(process=1,(2,0,0,0)) I0420 22:39:50.803449 134835206932288 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_3(process=1,(3,0,0,0)) I0420 22:39:50.803480 134835206932288 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_6(process=1,(2,1,0,0)) I0420 22:39:50.803506 134835206932288 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_7(process=1,(3,1,0,0)) I0420 22:39:51.068168 134835206932288 metric_logger.py:185] completed step: 0, seconds: 4.010, TFLOP/s/device: 0.200, Tokens/s/device: 31.922, total_weights: 4096, loss: 10.889 I0420 22:39:51.178319 134835206932288 metric_logger.py:185] completed step: 1, seconds: 2.093, TFLOP/s/device: 0.383, Tokens/s/device: 61.166, total_weights: 4096, loss: 10.902 I0420 22:39:51.571990 134835206932288 metric_logger.py:185] completed step: 2, seconds: 0.040, TFLOP/s/device: 19.876, Tokens/s/device: 3176.652, total_weights: 4096, loss: 9.956 I0420 22:39:51.642884 134835206932288 metric_logger.py:185] completed step: 3, seconds: 0.394, TFLOP/s/device: 2.033, Tokens/s/device: 324.910, total_weights: 4096, loss: 9.139 I0420 22:39:51.784985 134835206932288 metric_logger.py:185] completed step: 4, seconds: 0.075, TFLOP/s/device: 10.617, Tokens/s/device: 1696.915, total_weights: 4096, loss: 8.458 I0420 22:39:51.790949 134835206932288 metric_logger.py:185] completed step: 5, seconds: 0.071, TFLOP/s/device: 11.312, Tokens/s/device: 1807.961, total_weights: 4096, loss: 7.924 I0420 22:40:00.963327 134835206932288 metric_logger.py:185] completed step: 6, seconds: 0.143, TFLOP/s/device: 5.607, Tokens/s/device: 896.133, total_weights: 4096, loss: 7.553 I0420 22:40:01.034518 134835206932288 metric_logger.py:185] completed step: 7, seconds: 9.103, TFLOP/s/device: 0.088, Tokens/s/device: 14.061, total_weights: 4096, loss: 7.267 I0420 22:40:01.105293 134835206932288 metric_logger.py:185] completed step: 8, seconds: 0.075, TFLOP/s/device: 10.714, Tokens/s/device: 1712.283, total_weights: 4096, loss: 7.111 I0420 22:40:01.175720 134835206932288 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0420 22:40:01.176351 134835206932288 checkpointing.py:776] Waited 0.0006430149078369141 seconds for step 9 to finish before starting checkpointing. I0420 22:40:01.178436 134835206932288 checkpoint_manager.py:2024] [process=1][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0420 22:40:02.995860 2651 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0420 22:40:05.271145 134712051889920 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_03_dropout/checkpoints/0/items/array_metadatas/process_1 I0420 22:40:32.573011 134712043497216 base_pytree_checkpoint_handler.py:129] [process=1] /jax/checkpoint/write/gbytes_per_sec: 36.989 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.704935789108276 s) (per-host) I0420 22:40:32.573126 134712043497216 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.188559s. I0420 22:40:40.487047 134712043497216 async_checkpointer.py:144] [process=1][thread=async_save] Background save thread done. Time taken: 50.102464s. I0420 22:40:40.487303 134706666395392 async_checkpointer.py:273] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save. I0420 22:40:40.487431 134706666395392 async_checkpointer.py:283] [process=1][thread=save_finalize] No errors found in background save thread=async_save. I0420 22:40:40.487482 134706666395392 checkpoint_manager.py:2133] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0420 22:40:40.489644 134706666395392 checkpoint_manager.py:2142] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0420 22:40:40.489799 134835206932288 checkpoint_manager.py:2036] [process=1][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0420 22:40:40.489941 134835206932288 checkpoint_manager.py:1458] Waiting for previous save to complete took 39.311523 seconds. If this number is high, consider checkpointing less frequently. I0420 22:40:40.491813 134835206932288 checkpoint_manager.py:1518] [process=1] Saving checkpoint at step 9 I0420 22:40:40.495531 134835206932288 async_checkpointer.py:452] [process=1] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_03_dropout/checkpoints/9. I0420 22:40:41.040321 134835206932288 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0420 22:40:41.040426 134835206932288 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 I0420 22:40:41.083294 134835206932288 base_pytree_checkpoint_handler.py:153] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.044143s I0420 22:40:41.083473 134835206932288 base_pytree_checkpoint_handler.py:129] [process=1] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.214 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2958793640136719 s) (per-host) I0420 22:40:41.083530 134835206932288 base_pytree_checkpoint_handler.py:737] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.295945s (batch_requests_ready=0.249906s, total_serialization_initiated=0.045967s, others=0.000072s) I0420 22:40:41.083620 134835206932288 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.299895s (all_items=0.000018s, per_item={'items': '0.00001764'}, temp_paths=0.299877) I0420 22:40:41.084477 134682846848768 async_checkpointer.py:79] [process=1][thread=async_save] Background save thread started. I0420 22:40:41.084637 134835206932288 async_checkpointer.py:561] Finished blocking save. Time taken: 0.592747s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_03_dropout/checkpoints/9. I0420 22:40:41.517480 134835206932288 checkpoint_manager.py:1566] [process=1][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0420 22:40:41.517844 134706666395392 async_checkpointer.py:265] [process=1][thread=save_finalize] Waiting for background save thread=async_save. I0420 22:40:41.518024 134835206932288 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776724801.1783938, 'wait_for_prev_duration_secs': 39.31152319908142, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776724840.4918566, 'checkpointer_blocking_duration_secs': 0.5928912162780762, 'get_old_steps_start_time': 1776724841.0847676, 'get_old_steps_duration_secs': 2.5987625122070312e-05, 'checkpoint_manager_blocking_start_time': 1776724801.1766036, 'checkpoint_manager_blocking_duration_secs': 40.341383934020996} I0420 22:40:41.518153 134835206932288 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0420 22:40:41.518203 134835206932288 checkpoint_manager.py:2024] [process=1][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0420 22:40:46.311756 134702332897024 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_03_dropout/checkpoints/9/items/array_metadatas/process_1 I0420 22:41:21.801173 134682846848768 base_pytree_checkpoint_handler.py:129] [process=1] /jax/checkpoint/write/gbytes_per_sec: 38.514 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.0135383605957 s) (per-host) I0420 22:41:21.801297 134682846848768 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 40.716708s. I0420 22:41:29.837682 134682846848768 async_checkpointer.py:144] [process=1][thread=async_save] Background save thread done. Time taken: 48.753077s. I0420 22:41:29.837980 134706666395392 async_checkpointer.py:273] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save. I0420 22:41:29.838100 134706666395392 async_checkpointer.py:283] [process=1][thread=save_finalize] No errors found in background save thread=async_save. I0420 22:41:29.838149 134706666395392 checkpoint_manager.py:2133] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0420 22:41:29.839873 134706666395392 checkpoint_manager.py:2142] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0420 22:41:29.840053 134835206932288 checkpoint_manager.py:2036] [process=1][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0420 22:41:29.840197 134835206932288 checkpoint_manager.py:2013] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0420 22:41:29.840911 134835206932288 metric_logger.py:185] completed step: 9, seconds: 0.071, TFLOP/s/device: 11.300, Tokens/s/device: 1805.946, 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: Mon Apr 20 22:41:42 UTC 2026 EXIT_CODE=0