XPK Start: Sun Apr 19 03:58:07 UTC 2026 2026-04-19 03:58:11.598007: 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 00:00:1776571091.610581 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776571091.614548 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776571091.625831 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776571091.625850 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776571091.625852 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776571091.625854 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-19 03:58:30.784723: 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) I0419 03:58:31.307638 140562962360128 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-19 03:58:40,347:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0419 03:58:40.347937 140562962360128 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-19 03:58:40,350:jax._src.distributed:157: Connecting to JAX distributed service on mt-11-optimizer-offload-bjz1j-slice-job-0-0.mt-11-optimizer-offload-bjz1j:8482 I0419 03:58:40.350518 140562962360128 distributed.py:157] Connecting to JAX distributed service on mt-11-optimizer-offload-bjz1j-slice-job-0-0.mt-11-optimizer-offload-bjz1j:8482 I0419 03:58:42.156283 140562962360128 max_utils.py:284] Jax distributed system initialized! I0419 03:58:48.243649 140562962360128 max_utils.py:800] System Information: Jax Version: 0.8.1 I0419 03:58:48.243753 140562962360128 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0419 03:58:48.243793 140562962360128 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 I0419 03:58:48.243827 140562962360128 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0419 03:58:48.859870 140562962360128 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 03:58:48.860465 140562962360128 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 03:58:48.860660 140562962360128 checkpointing.py:688] Setting up checkpoint logger... I0419 03:58:48.860712 140562962360128 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0419 03:58:48.860756 140562962360128 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0419 03:58:48.861103 140562962360128 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 0x7fd6e3fa0200>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0419 03:58:53.416185 140562962360128 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0419 03:58:53.416415 140562962360128 checkpoint_manager.py:709] [process=3][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fc2dc465a00>}, handler_registry=None I0419 03:58:53.416661 140562962360128 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fc2dc465a00>` 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`. I0419 03:58:53.416713 140562962360128 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fd598103500>` 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`. I0419 03:58:53.416749 140562962360128 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 0x7fc2dc465a00>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fc2dc465a00>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fd598103500>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fd598103500>}). I0419 03:58:53.417255 140562962360128 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0419 03:58:53.417334 140562962360128 async_checkpointer.py:177] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7fc2dc224180> timeout: 600 secs and primary_host=0 for async checkpoint writes I0419 03:58:55.405953 140562962360128 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_11_optimizer_offload_true/checkpoints I0419 03:58:55.408189 140562962360128 checkpoint_manager.py:929] [process=3][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_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_11_optimizer_offload_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7fc2dc3d2090> I0419 03:58:55.408302 140562962360128 checkpointing.py:302] Checkpoint manager created! I0419 03:58:56.302286 140562962360128 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0419 03:58:56.302394 140562962360128 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 03:58:56.670168 140562962360128 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 03:58:56.670262 140562962360128 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 03:58:56.686352 140562962360128 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 03:58:56.686413 140562962360128 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 03:58:56.709475 140562962360128 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 03:58:56.709545 140562962360128 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 03:58:56.725774 140562962360128 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 03:58:56.725838 140562962360128 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 03:58:56.742049 140562962360128 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 03:58:56.742115 140562962360128 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 03:58:56.767039 140562962360128 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0419 03:58:56.767112 140562962360128 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 03:58:56.786629 140562962360128 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0419 03:58:56.786691 140562962360128 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0419 03:58:56.976635 140562962360128 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0419 03:58:56.976746 140562962360128 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0419 03:58:58.856923 140562962360128 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0419 03:58:58.857058 140562962360128 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0419 03:58:58.857113 140562962360128 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0419 03:58:58.857174 140562962360128 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: PartitionSpec('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0419 03:58:58.857229 140562962360128 maxtext_utils.py:1740] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0419 03:58:58.857272 140562962360128 maxtext_utils.py:1740] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0419 03:58:58.857325 140562962360128 maxtext_utils.py:1740] 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) I0419 03:58:58.857378 140562962360128 maxtext_utils.py:1740] 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') I0419 03:58:58.857418 140562962360128 maxtext_utils.py:1740] 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) I0419 03:58:58.857455 140562962360128 maxtext_utils.py:1740] 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) I0419 03:58:58.857502 140562962360128 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0419 03:58:58.857551 140562962360128 maxtext_utils.py:1740] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0419 03:58:59.374141 140562962360128 train.py:158] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 03:58:59.374240 140562962360128 train.py:158] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0419 03:58:59.389297 140562962360128 train.py:165] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 03:58:59.389359 140562962360128 train.py:165] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0419 03:59:10.194561 140562962360128 max_utils.py:791] Total memory size: 1.3 GB, Output size: 0.1 GB, Temp size: 1.1 GB, Argument size: 0.1 GB, Host temp size: 0.0 GB. I0419 03:59:10.195518 140562962360128 metric_logger.py:289] number parameters: 1.104 billion I0419 03:59:21.146865 140562962360128 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0419 03:59:21.321105 140562962360128 checkpointing.py:798] Waited 0.1742229461669922 seconds for step 0 to finish before starting checkpointing. I0419 03:59:21.323704 140562962360128 checkpoint_manager.py:2013] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 03:59:21.325652 140562962360128 checkpoint_manager.py:1518] [process=3] Saving checkpoint at step 0 I0419 03:59:21.328364 140562962360128 async_checkpointer.py:452] [process=3] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_11_optimizer_offload_true/checkpoints/0. I0419 03:59:22.783471 140562962360128 signaling_client.py:364] Using JaxDistributedSignalingClient I0419 03:59:22.784536 140562962360128 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 03:59:22.784594 140562962360128 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 I0419 03:59:23.032565 140562962360128 base_pytree_checkpoint_handler.py:153] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.249171s I0419 03:59:23.032755 140562962360128 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.036 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.508063793182373 s) (per-host) I0419 03:59:23.032809 140562962360128 base_pytree_checkpoint_handler.py:737] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.508127s (batch_requests_ready=0.254086s, total_serialization_initiated=0.253970s, others=0.000072s) I0419 03:59:23.032901 140562962360128 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.512570s (all_items=0.000020s, per_item={'items': '0.00002003'}, temp_paths=0.512550) I0419 03:59:23.033903 140441789912832 async_checkpointer.py:79] [process=3][thread=async_save] Background save thread started. I0419 03:59:23.034034 140562962360128 async_checkpointer.py:561] Finished blocking save. Time taken: 1.708308s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_11_optimizer_offload_true/checkpoints/0. I0419 03:59:23.476139 140562962360128 checkpoint_manager.py:1566] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0419 03:59:23.476515 140430079399680 async_checkpointer.py:265] [process=3][thread=save_finalize] Waiting for background save thread=async_save. I0419 03:59:23.476698 140562962360128 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_11_optimizer_offload_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776571161.3236864, 'wait_for_prev_duration_secs': 6.246566772460938e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776571161.3256912, 'checkpointer_blocking_duration_secs': 1.708543062210083, 'get_old_steps_start_time': 1776571163.0342546, 'get_old_steps_duration_secs': 2.5987625122070312e-05, 'checkpoint_manager_blocking_start_time': 1776571161.3216605, 'checkpoint_manager_blocking_duration_secs': 2.154995918273926} I0419 03:59:23.476815 140562962360128 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0419 03:59:23.476865 140562962360128 max_utils.py:750] Memstats: After params initialized: I0419 03:59:23.476918 140562962360128 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_10(process=3,(2,2,0,0)) I0419 03:59:23.476951 140562962360128 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_11(process=3,(3,2,0,0)) I0419 03:59:23.476977 140562962360128 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_14(process=3,(2,3,0,0)) I0419 03:59:23.477001 140562962360128 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_15(process=3,(3,3,0,0)) I0419 03:59:23.742336 140562962360128 metric_logger.py:185] completed step: 0, seconds: 10.951, TFLOP/s/device: 1.241, Tokens/s/device: 187.011, total_weights: 65536, loss: 10.874 I0419 03:59:23.929026 140562962360128 metric_logger.py:185] completed step: 1, seconds: 2.594, TFLOP/s/device: 5.238, Tokens/s/device: 789.455, total_weights: 65536, loss: 10.874 I0419 03:59:24.308371 140562962360128 metric_logger.py:185] completed step: 2, seconds: 0.023, TFLOP/s/device: 587.958, Tokens/s/device: 88623.480, total_weights: 65536, loss: 10.263 I0419 03:59:24.472076 140562962360128 metric_logger.py:185] completed step: 3, seconds: 0.380, TFLOP/s/device: 35.776, Tokens/s/device: 5392.582, total_weights: 65536, loss: 9.735 I0419 03:59:24.800201 140562962360128 metric_logger.py:185] completed step: 4, seconds: 0.169, TFLOP/s/device: 80.453, Tokens/s/device: 12126.739, total_weights: 65536, loss: 9.278 I0419 03:59:24.805738 140562962360128 metric_logger.py:185] completed step: 5, seconds: 0.164, TFLOP/s/device: 82.974, Tokens/s/device: 12506.794, total_weights: 65536, loss: 8.892 I0419 03:59:44.302261 140562962360128 metric_logger.py:185] completed step: 6, seconds: 0.328, TFLOP/s/device: 41.377, Tokens/s/device: 6236.734, total_weights: 65536, loss: 8.592 I0419 03:59:44.465910 140562962360128 metric_logger.py:185] completed step: 7, seconds: 19.334, TFLOP/s/device: 0.703, Tokens/s/device: 105.930, total_weights: 65536, loss: 8.384 I0419 03:59:44.629556 140562962360128 metric_logger.py:185] completed step: 8, seconds: 0.168, TFLOP/s/device: 80.773, Tokens/s/device: 12175.040, total_weights: 65536, loss: 8.255 I0419 03:59:44.792991 140562962360128 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0419 03:59:44.793645 140562962360128 checkpointing.py:798] Waited 0.0006532669067382812 seconds for step 9 to finish before starting checkpointing. I0419 03:59:44.795500 140562962360128 checkpoint_manager.py:2024] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 03:59:45.807595 2600 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0419 03:59:47.715361 140441798305536 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_11_optimizer_offload_true/checkpoints/0/items/array_metadatas/process_3 I0419 04:00:19.777590 140441789912832 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/gbytes_per_sec: 27.589 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 57.252859115600586 s) (per-host) I0419 04:00:19.777717 140441789912832 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 56.743586s. I0419 04:00:28.451311 140441789912832 async_checkpointer.py:144] [process=3][thread=async_save] Background save thread done. Time taken: 65.417164s. I0419 04:00:28.451550 140430079399680 async_checkpointer.py:273] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 04:00:28.451650 140430079399680 async_checkpointer.py:283] [process=3][thread=save_finalize] No errors found in background save thread=async_save. I0419 04:00:28.451730 140430079399680 checkpoint_manager.py:2133] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0419 04:00:28.453416 140430079399680 checkpoint_manager.py:2142] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0419 04:00:28.453604 140562962360128 checkpoint_manager.py:2036] [process=3][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0419 04:00:28.453776 140562962360128 checkpoint_manager.py:1458] Waiting for previous save to complete took 43.658280 seconds. If this number is high, consider checkpointing less frequently. I0419 04:00:28.455592 140562962360128 checkpoint_manager.py:1518] [process=3] Saving checkpoint at step 9 I0419 04:00:28.459116 140562962360128 async_checkpointer.py:452] [process=3] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_11_optimizer_offload_true/checkpoints/9. I0419 04:00:29.839478 140562962360128 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 04:00:29.839576 140562962360128 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 I0419 04:00:29.879820 140562962360128 base_pytree_checkpoint_handler.py:153] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.041511s I0419 04:00:29.880001 140562962360128 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.230 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.29491257667541504 s) (per-host) I0419 04:00:29.880055 140562962360128 base_pytree_checkpoint_handler.py:737] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.294982s (batch_requests_ready=0.251675s, total_serialization_initiated=0.043229s, others=0.000078s) I0419 04:00:29.880142 140562962360128 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.299048s (all_items=0.000022s, per_item={'items': '0.00002217'}, temp_paths=0.299026) I0419 04:00:29.881024 140434269116160 async_checkpointer.py:79] [process=3][thread=async_save] Background save thread started. I0419 04:00:29.881157 140562962360128 async_checkpointer.py:561] Finished blocking save. Time taken: 1.425480s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_11_optimizer_offload_true/checkpoints/9. I0419 04:00:30.326582 140562962360128 checkpoint_manager.py:1566] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0419 04:00:30.326999 140430079399680 async_checkpointer.py:265] [process=3][thread=save_finalize] Waiting for background save thread=async_save. I0419 04:00:30.327181 140562962360128 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_11_optimizer_offload_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776571184.7954714, 'wait_for_prev_duration_secs': 43.65827965736389, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776571228.455646, 'checkpointer_blocking_duration_secs': 1.4256887435913086, 'get_old_steps_start_time': 1776571229.8813522, 'get_old_steps_duration_secs': 2.6702880859375e-05, 'checkpoint_manager_blocking_start_time': 1776571184.7938652, 'checkpoint_manager_blocking_duration_secs': 45.533278703689575} I0419 04:00:30.327290 140562962360128 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0419 04:00:30.327339 140562962360128 checkpoint_manager.py:2024] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 04:00:36.908937 140441798305536 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_11_optimizer_offload_true/checkpoints/9/items/array_metadatas/process_3 I0419 04:01:13.497954 140434269116160 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/gbytes_per_sec: 35.970 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.91283178329468 s) (per-host) I0419 04:01:13.498086 140434269116160 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.616846s. I0419 04:01:22.300075 140434269116160 async_checkpointer.py:144] [process=3][thread=async_save] Background save thread done. Time taken: 52.418821s. I0419 04:01:22.300390 140430079399680 async_checkpointer.py:273] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 04:01:22.300508 140430079399680 async_checkpointer.py:283] [process=3][thread=save_finalize] No errors found in background save thread=async_save. I0419 04:01:22.300555 140430079399680 checkpoint_manager.py:2133] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0419 04:01:22.302197 140430079399680 checkpoint_manager.py:2142] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0419 04:01:22.302370 140562962360128 checkpoint_manager.py:2036] [process=3][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0419 04:01:22.302539 140562962360128 checkpoint_manager.py:2013] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 04:01:22.303306 140562962360128 metric_logger.py:185] completed step: 9, seconds: 0.164, TFLOP/s/device: 83.000, Tokens/s/device: 12510.690, 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: Sun Apr 19 04:01:33 UTC 2026 EXIT_CODE=0