XPK Start: Fri Apr 24 18:06:06 UTC 2026 2026-04-24 18:06:10.766021: 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 00:00:1777053970.778999 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1777053970.782702 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1777053970.794321 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777053970.794339 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777053970.794341 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777053970.794343 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-24 18:06:29.843819: 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 18:06:30.364146 137623935600448 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-24 18:06:39,405:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0424 18:06:39.405967 137623935600448 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-24 18:06:39,408:jax._src.distributed:157: Connecting to JAX distributed service on mt-03-dropout-0lrpf-slice-job-0-0.mt-03-dropout-0lrpf:8482 I0424 18:06:39.408674 137623935600448 distributed.py:157] Connecting to JAX distributed service on mt-03-dropout-0lrpf-slice-job-0-0.mt-03-dropout-0lrpf:8482 I0424 18:06:40.932346 137623935600448 max_utils.py:284] Jax distributed system initialized! I0424 18:06:47.859667 137623935600448 max_utils.py:800] System Information: Jax Version: 0.8.1 I0424 18:06:47.859772 137623935600448 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0424 18:06:47.859814 137623935600448 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 18:06:47.859850 137623935600448 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0424 18:06:48.474937 137623935600448 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0424 18:06:48.475222 137623935600448 checkpointing.py:677] Setting up checkpoint logger... I0424 18:06:48.475278 137623935600448 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0424 18:06:48.475328 137623935600448 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0424 18:06:48.475674 137623935600448 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 0x7d2a8d774fe0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0424 18:06:51.440294 137623935600448 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0424 18:06:51.440637 137623935600448 checkpoint_manager.py:709] [process=4][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d17f85dd4f0>}, handler_registry=None I0424 18:06:51.440884 137623935600448 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d17f85dd4f0>` 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 18:06:51.440933 137623935600448 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d1727d37bc0>` 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 18:06:51.440968 137623935600448 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 0x7d17f85dd4f0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d17f85dd4f0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d1727d37bc0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d1727d37bc0>}). I0424 18:06:51.441479 137623935600448 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0424 18:06:51.441556 137623935600448 async_checkpointer.py:177] [process=4][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d16cc27a840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0424 18:06:52.221223 137623935600448 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_03_dropout/checkpoints I0424 18:06:52.483632 137623935600448 checkpoint_manager.py:929] [process=4][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_03_dropout/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d1727d375f0> I0424 18:06:52.483811 137623935600448 checkpointing.py:301] Checkpoint manager created! I0424 18:06:52.799941 137623935600448 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0424 18:06:52.800046 137623935600448 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0424 18:06:53.171848 137623935600448 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0424 18:06:53.171953 137623935600448 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0424 18:06:53.188121 137623935600448 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0424 18:06:53.188180 137623935600448 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0424 18:06:53.211380 137623935600448 attentions.py:1154] attentions/query Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 18:06:53.211452 137623935600448 attentions.py:1154] attentions/query Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0424 18:06:53.227704 137623935600448 attentions.py:1155] attentions/key Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 18:06:53.227770 137623935600448 attentions.py:1155] attentions/key Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0424 18:06:53.243910 137623935600448 attentions.py:1156] attentions/value Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 18:06:53.243969 137623935600448 attentions.py:1156] attentions/value Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0424 18:06:53.268373 137623935600448 attentions.py:1197] attentions/out Logical: bfloat16[32,128,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0424 18:06:53.268445 137623935600448 attentions.py:1197] attentions/out Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0424 18:06:53.294591 137623935600448 linears.py:525] linears/x Logical: bfloat16[32,128,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0424 18:06:53.294661 137623935600448 linears.py:525] linears/x Physical: bfloat16[32,128,7168]....................................... ('fsdp', None, None). I0424 18:06:53.496774 137623935600448 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0424 18:06:53.496905 137623935600448 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0424 18:06:54.959485 137623935600448 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0424 18:06:54.959616 137623935600448 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 18:06:54.959671 137623935600448 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 18:06:54.959731 137623935600448 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 18:06:54.959782 137623935600448 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 18:06:54.959820 137623935600448 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 18:06:54.959892 137623935600448 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 18:06:54.959974 137623935600448 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 18:06:54.960023 137623935600448 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 18:06:54.960061 137623935600448 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 18:06:54.960111 137623935600448 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0424 18:06:54.960161 137623935600448 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0424 18:06:55.496178 137623935600448 train.py:155] train/xent Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0424 18:06:55.496279 137623935600448 train.py:155] train/xent Physical: float32[32,128]............................................. ('fsdp', None). I0424 18:06:55.511435 137623935600448 train.py:162] train/z_loss Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0424 18:06:55.511494 137623935600448 train.py:162] train/z_loss Physical: float32[32,128]............................................. ('fsdp', None). I0424 18:06:58.799956 137623935600448 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. I0424 18:06:58.800750 137623935600448 metric_logger.py:289] number parameters: 1.104 billion I0424 18:07:02.704153 137623935600448 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0424 18:07:03.035147 137623935600448 checkpointing.py:776] Waited 0.3309767246246338 seconds for step 0 to finish before starting checkpointing. I0424 18:07:03.038089 137623935600448 checkpoint_manager.py:2013] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0424 18:07:03.039840 137623935600448 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 0 I0424 18:07:03.042645 137623935600448 async_checkpointer.py:452] [process=4] 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_03_dropout/checkpoints/0. I0424 18:07:03.591902 137623935600448 signaling_client.py:364] Using JaxDistributedSignalingClient I0424 18:07:03.592860 137623935600448 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0424 18:07:03.592930 137623935600448 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 18:07:03.840773 137623935600448 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.248963s I0424 18:07:03.840961 137623935600448 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.035 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5083003044128418 s) (per-host) I0424 18:07:03.841013 137623935600448 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.508363s (batch_requests_ready=0.255815s, total_serialization_initiated=0.252479s, others=0.000070s) I0424 18:07:03.841103 137623935600448 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.512373s (all_items=0.000019s, per_item={'items': '0.00001931'}, temp_paths=0.512354) I0424 18:07:03.842017 137498315450112 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started. I0424 18:07:03.842147 137623935600448 async_checkpointer.py:561] Finished blocking save. Time taken: 0.802218s. 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_03_dropout/checkpoints/0. I0424 18:07:03.859796 137623935600448 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0424 18:07:03.860110 137497807943424 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save. I0424 18:07:03.860270 137623935600448 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_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777054023.0380719, 'wait_for_prev_duration_secs': 6.389617919921875e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777054023.0398948, 'checkpointer_blocking_duration_secs': 0.8024392127990723, 'get_old_steps_start_time': 1777054023.842354, 'get_old_steps_duration_secs': 2.6941299438476562e-05, 'checkpoint_manager_blocking_start_time': 1777054023.0356894, 'checkpoint_manager_blocking_duration_secs': 0.8245415687561035} I0424 18:07:03.860381 137623935600448 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0424 18:07:03.860432 137623935600448 max_utils.py:750] Memstats: After params initialized: I0424 18:07:03.860481 137623935600448 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_16(process=4,(0,4,0,0)) I0424 18:07:03.860514 137623935600448 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_17(process=4,(1,4,0,0)) I0424 18:07:03.860543 137623935600448 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_20(process=4,(0,5,0,0)) I0424 18:07:03.860569 137623935600448 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_21(process=4,(1,5,0,0)) I0424 18:07:04.125298 137623935600448 metric_logger.py:185] completed step: 0, seconds: 3.903, TFLOP/s/device: 0.205, Tokens/s/device: 32.793, total_weights: 4096, loss: 10.889 I0424 18:07:04.230740 137623935600448 metric_logger.py:185] completed step: 1, seconds: 1.420, TFLOP/s/device: 0.564, Tokens/s/device: 90.152, total_weights: 4096, loss: 10.902 I0424 18:07:04.632691 137623935600448 metric_logger.py:185] completed step: 2, seconds: 0.035, TFLOP/s/device: 22.571, Tokens/s/device: 3607.361, total_weights: 4096, loss: 9.956 I0424 18:07:04.703472 137623935600448 metric_logger.py:185] completed step: 3, seconds: 0.402, TFLOP/s/device: 1.991, Tokens/s/device: 318.255, total_weights: 4096, loss: 9.139 I0424 18:07:04.845438 137623935600448 metric_logger.py:185] completed step: 4, seconds: 0.076, TFLOP/s/device: 10.576, Tokens/s/device: 1690.305, total_weights: 4096, loss: 8.458 I0424 18:07:04.851242 137623935600448 metric_logger.py:185] completed step: 5, seconds: 0.071, TFLOP/s/device: 11.343, Tokens/s/device: 1812.826, total_weights: 4096, loss: 7.924 I0424 18:07:13.670272 137623935600448 metric_logger.py:185] completed step: 6, seconds: 0.143, TFLOP/s/device: 5.620, Tokens/s/device: 898.201, total_weights: 4096, loss: 7.553 I0424 18:07:13.741969 137623935600448 metric_logger.py:185] completed step: 7, seconds: 8.749, TFLOP/s/device: 0.092, Tokens/s/device: 14.629, total_weights: 4096, loss: 7.267 I0424 18:07:13.813511 137623935600448 metric_logger.py:185] completed step: 8, seconds: 0.075, TFLOP/s/device: 10.667, Tokens/s/device: 1704.825, total_weights: 4096, loss: 7.111 I0424 18:07:13.884840 137623935600448 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0424 18:07:13.885462 137623935600448 checkpointing.py:776] Waited 0.0006418228149414062 seconds for step 9 to finish before starting checkpointing. I0424 18:07:13.887424 137623935600448 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0424 18:07:15.687933 2607 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0424 18:07:17.624917 137498302850816 array_metadata_store.py:203] [process=4][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_03_dropout/checkpoints/0/items/array_metadatas/process_4 I0424 18:07:45.361223 137498315450112 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 37.584 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.02852439880371 s) (per-host) I0424 18:07:45.361343 137498315450112 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.519120s. I0424 18:07:52.885525 137498315450112 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 49.043286s. I0424 18:07:52.885829 137497807943424 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save. I0424 18:07:52.885996 137497807943424 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save. I0424 18:07:52.886052 137497807943424 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0424 18:07:52.887573 137497807943424 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0424 18:07:52.887754 137623935600448 checkpoint_manager.py:2036] [process=4][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0424 18:07:52.887892 137623935600448 checkpoint_manager.py:1458] Waiting for previous save to complete took 39.000463 seconds. If this number is high, consider checkpointing less frequently. I0424 18:07:52.889813 137623935600448 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 9 I0424 18:07:52.893255 137623935600448 async_checkpointer.py:452] [process=4] 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_03_dropout/checkpoints/9. I0424 18:07:53.418355 137623935600448 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0424 18:07:53.418453 137623935600448 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 18:07:53.462131 137623935600448 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.045094s I0424 18:07:53.462304 137623935600448 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.348 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28842639923095703 s) (per-host) I0424 18:07:53.462356 137623935600448 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.288488s (batch_requests_ready=0.241496s, total_serialization_initiated=0.046923s, others=0.000069s) I0424 18:07:53.462449 137623935600448 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.292729s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=0.292713) I0424 18:07:53.463360 137499418556160 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started. I0424 18:07:53.463486 137623935600448 async_checkpointer.py:561] Finished blocking save. Time taken: 0.573596s. 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_03_dropout/checkpoints/9. I0424 18:07:53.480170 137623935600448 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0424 18:07:53.480468 137497807943424 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save. I0424 18:07:53.480638 137623935600448 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_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777054033.8873956, 'wait_for_prev_duration_secs': 39.00046253204346, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777054072.8898551, 'checkpointer_blocking_duration_secs': 0.573807954788208, 'get_old_steps_start_time': 1777054073.4636827, 'get_old_steps_duration_secs': 2.6226043701171875e-05, 'checkpoint_manager_blocking_start_time': 1777054033.885679, 'checkpoint_manager_blocking_duration_secs': 39.594923973083496} I0424 18:07:53.480747 137623935600448 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0424 18:07:53.480792 137623935600448 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0424 18:07:58.429704 137471698327296 array_metadata_store.py:203] [process=4][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_03_dropout/checkpoints/9/items/array_metadatas/process_4 I0424 18:08:34.829092 137499418556160 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 37.921 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.65518832206726 s) (per-host) I0424 18:08:34.829196 137499418556160 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.365628s. I0424 18:08:41.606778 137499418556160 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 48.143196s. I0424 18:08:41.607106 137497807943424 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save. I0424 18:08:41.607219 137497807943424 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save. I0424 18:08:41.607270 137497807943424 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0424 18:08:41.608657 137497807943424 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0424 18:08:41.608827 137623935600448 checkpoint_manager.py:2036] [process=4][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0424 18:08:41.608984 137623935600448 checkpoint_manager.py:2013] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0424 18:08:41.609625 137623935600448 metric_logger.py:185] completed step: 9, seconds: 0.072, TFLOP/s/device: 11.188, Tokens/s/device: 1788.084, 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: Fri Apr 24 18:08:52 UTC 2026 EXIT_CODE=0