XPK Start: Wed Apr 22 16:36:44 UTC 2026 2026-04-22 16:36:48.105484: 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 00:00:1776875808.118193 9 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776875808.122007 9 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776875808.133221 9 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776875808.133242 9 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776875808.133244 9 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776875808.133246 9 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-22 16:37:07.377049: 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 16:37:07.900795 133153302591296 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-22 16:37:16,940:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0422 16:37:16.940494 133153302591296 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-22 16:37:16,942:jax._src.distributed:157: Connecting to JAX distributed service on mt-13-scan-layers-true-oez2w-slice-job-0-0.mt-13-scan-layers-true-oez2w:8482 I0422 16:37:16.942805 133153302591296 distributed.py:157] Connecting to JAX distributed service on mt-13-scan-layers-true-oez2w-slice-job-0-0.mt-13-scan-layers-true-oez2w:8482 I0422 16:37:17.983933 133153302591296 max_utils.py:284] Jax distributed system initialized! I0422 16:37:24.826986 133153302591296 max_utils.py:800] System Information: Jax Version: 0.8.1 I0422 16:37:24.827091 133153302591296 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0422 16:37:24.827132 133153302591296 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 16:37:24.827166 133153302591296 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0422 16:37:25.451327 133153302591296 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0422 16:37:25.451924 133153302591296 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0422 16:37:25.452109 133153302591296 checkpointing.py:688] Setting up checkpoint logger... I0422 16:37:25.452159 133153302591296 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0422 16:37:25.452208 133153302591296 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0422 16:37:25.452561 133153302591296 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 0x7919abf790d0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0422 16:37:28.459136 133153302591296 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0422 16:37:28.459486 133153302591296 checkpoint_manager.py:709] [process=5][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7906c9921580>}, handler_registry=None I0422 16:37:28.459733 133153302591296 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7906c9921580>` 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 16:37:28.459784 133153302591296 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7906c9925820>` 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 16:37:28.459821 133153302591296 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 0x7906c9921580>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7906c9921580>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7906c9925820>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7906c9925820>}). I0422 16:37:28.460331 133153302591296 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0422 16:37:28.460410 133153302591296 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x79065fad4180> timeout: 600 secs and primary_host=0 for async checkpoint writes I0422 16:37:29.573603 133153302591296 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_13_scan_layers_true/checkpoints I0422 16:37:30.026121 133153302591296 checkpoint_manager.py:929] [process=5][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_13_scan_layers_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7906ca4b9670> I0422 16:37:30.026302 133153302591296 checkpointing.py:302] Checkpoint manager created! I0422 16:37:30.912751 133153302591296 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0422 16:37:30.912856 133153302591296 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 16:37:31.281661 133153302591296 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 16:37:31.281754 133153302591296 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 16:37:31.297569 133153302591296 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 16:37:31.297632 133153302591296 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 16:37:31.320273 133153302591296 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 16:37:31.320343 133153302591296 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 16:37:31.336197 133153302591296 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 16:37:31.336258 133153302591296 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 16:37:31.352063 133153302591296 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 16:37:31.352122 133153302591296 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 16:37:31.376866 133153302591296 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0422 16:37:31.376935 133153302591296 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 16:37:31.396455 133153302591296 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0422 16:37:31.396520 133153302591296 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0422 16:37:31.588248 133153302591296 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0422 16:37:31.588356 133153302591296 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0422 16:37:33.007831 133153302591296 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0422 16:37:33.007966 133153302591296 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) I0422 16:37:33.008022 133153302591296 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) I0422 16:37:33.008080 133153302591296 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') I0422 16:37:33.008132 133153302591296 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) I0422 16:37:33.008171 133153302591296 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) I0422 16:37:33.008227 133153302591296 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) I0422 16:37:33.008280 133153302591296 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') I0422 16:37:33.008321 133153302591296 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) I0422 16:37:33.008358 133153302591296 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) I0422 16:37:33.008406 133153302591296 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0422 16:37:33.008453 133153302591296 maxtext_utils.py:1740] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0422 16:37:33.505378 133153302591296 train.py:158] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0422 16:37:33.505472 133153302591296 train.py:158] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0422 16:37:33.520453 133153302591296 train.py:165] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0422 16:37:33.520511 133153302591296 train.py:165] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0422 16:37:44.443853 133153302591296 max_utils.py:791] Total memory size: 1.5 GB, Output size: 0.4 GB, Temp size: 1.1 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0422 16:37:44.444685 133153302591296 metric_logger.py:289] number parameters: 1.104 billion I0422 16:37:55.596269 133153302591296 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0422 16:37:55.752827 133153302591296 checkpointing.py:798] Waited 0.15653753280639648 seconds for step 0 to finish before starting checkpointing. I0422 16:37:55.755423 133153302591296 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 16:37:55.757110 133153302591296 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0422 16:37:55.759855 133153302591296 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_13_scan_layers_true/checkpoints/0. I0422 16:37:56.351151 133153302591296 signaling_client.py:364] Using JaxDistributedSignalingClient I0422 16:37:56.352254 133153302591296 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0422 16:37:56.352313 133153302591296 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 16:37:56.596967 133153302591296 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.245903s I0422 16:37:56.597140 133153302591296 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.019 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5109121799468994 s) (per-host) I0422 16:37:56.597192 133153302591296 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.510974s (batch_requests_ready=0.261425s, total_serialization_initiated=0.249481s, others=0.000068s) I0422 16:37:56.597280 133153302591296 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.514990s (all_items=0.000020s, per_item={'items': '0.00002027'}, temp_paths=0.514969) I0422 16:37:56.598211 133028349208320 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0422 16:37:56.598381 133153302591296 async_checkpointer.py:561] Finished blocking save. Time taken: 0.841204s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_13_scan_layers_true/checkpoints/0. I0422 16:37:56.644569 133153302591296 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0422 16:37:56.644878 133027275466496 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0422 16:37:56.645025 133153302591296 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_13_scan_layers_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776875875.755403, 'wait_for_prev_duration_secs': 6.508827209472656e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776875875.757149, 'checkpointer_blocking_duration_secs': 0.8413403034210205, 'get_old_steps_start_time': 1776875876.5985081, 'get_old_steps_duration_secs': 2.6464462280273438e-05, 'checkpoint_manager_blocking_start_time': 1776875875.7533972, 'checkpoint_manager_blocking_duration_secs': 0.8915889263153076} I0422 16:37:56.645129 133153302591296 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0422 16:37:56.645185 133153302591296 max_utils.py:750] Memstats: After params initialized: I0422 16:37:56.645240 133153302591296 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0)) I0422 16:37:56.645275 133153302591296 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0)) I0422 16:37:56.645304 133153302591296 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0)) I0422 16:37:56.645331 133153302591296 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0)) I0422 16:37:56.913626 133153302591296 metric_logger.py:185] completed step: 0, seconds: 11.151, TFLOP/s/device: 1.218, Tokens/s/device: 183.653, total_weights: 65536, loss: 10.874 I0422 16:37:57.082870 133153302591296 metric_logger.py:185] completed step: 1, seconds: 1.316, TFLOP/s/device: 10.324, Tokens/s/device: 1556.154, total_weights: 65536, loss: 10.874 I0422 16:37:57.471356 133153302591296 metric_logger.py:185] completed step: 2, seconds: 0.023, TFLOP/s/device: 600.563, Tokens/s/device: 90523.338, total_weights: 65536, loss: 10.263 I0422 16:37:57.618901 133153302591296 metric_logger.py:185] completed step: 3, seconds: 0.388, TFLOP/s/device: 34.989, Tokens/s/device: 5273.960, total_weights: 65536, loss: 9.734 I0422 16:37:57.914365 133153302591296 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 89.025, Tokens/s/device: 13418.773, total_weights: 65536, loss: 9.278 I0422 16:37:57.920142 133153302591296 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.130, Tokens/s/device: 13886.817, total_weights: 65536, loss: 8.892 I0422 16:38:16.406433 133153302591296 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.903, Tokens/s/device: 6919.059, total_weights: 65536, loss: 8.592 I0422 16:38:16.554017 133153302591296 metric_logger.py:185] completed step: 7, seconds: 18.340, TFLOP/s/device: 0.741, Tokens/s/device: 111.668, total_weights: 65536, loss: 8.384 I0422 16:38:16.701447 133153302591296 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.615, Tokens/s/device: 13507.809, total_weights: 65536, loss: 8.255 I0422 16:38:16.848480 133153302591296 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0422 16:38:16.849109 133153302591296 checkpointing.py:798] Waited 0.0006499290466308594 seconds for step 9 to finish before starting checkpointing. I0422 16:38:16.851018 133153302591296 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 16:38:17.793328 2634 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0422 16:38:19.726527 133028357601024 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_13_scan_layers_true/checkpoints/0/items/array_metadatas/process_5 I0422 16:38:52.072470 133028349208320 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 28.214 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 55.98620009422302 s) (per-host) I0422 16:38:52.072591 133028349208320 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 55.474267s. I0422 16:39:00.020767 133028349208320 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 63.422421s. I0422 16:39:00.021052 133027275466496 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 16:39:00.021183 133027275466496 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0422 16:39:00.021234 133027275466496 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0422 16:39:00.022773 133027275466496 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0422 16:39:00.022946 133153302591296 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0422 16:39:00.023093 133153302591296 checkpoint_manager.py:1458] Waiting for previous save to complete took 43.172082 seconds. If this number is high, consider checkpointing less frequently. I0422 16:39:00.024961 133153302591296 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0422 16:39:00.028403 133153302591296 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_13_scan_layers_true/checkpoints/9. I0422 16:39:01.004842 133153302591296 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0422 16:39:01.004946 133153302591296 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 16:39:01.047829 133153302591296 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.044248s I0422 16:39:01.047994 133153302591296 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.114 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.7295703887939453 s) (per-host) I0422 16:39:01.048044 133153302591296 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.729630s (batch_requests_ready=0.683520s, total_serialization_initiated=0.046043s, others=0.000067s) I0422 16:39:01.048129 133153302591296 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.733696s (all_items=0.000015s, per_item={'items': '0.00001478'}, temp_paths=0.733681) I0422 16:39:01.049022 133027820730112 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0422 16:39:01.049180 133153302591296 async_checkpointer.py:561] Finished blocking save. Time taken: 1.024146s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_13_scan_layers_true/checkpoints/9. I0422 16:39:01.061362 133153302591296 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0422 16:39:01.061591 133027275466496 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0422 16:39:01.061731 133153302591296 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_13_scan_layers_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776875896.8509872, 'wait_for_prev_duration_secs': 43.17208170890808, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776875940.025001, 'checkpointer_blocking_duration_secs': 1.0242867469787598, 'get_old_steps_start_time': 1776875941.0493042, 'get_old_steps_duration_secs': 2.765655517578125e-05, 'checkpoint_manager_blocking_start_time': 1776875896.8493373, 'checkpoint_manager_blocking_duration_secs': 44.212358713150024} I0422 16:39:01.061831 133153302591296 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0422 16:39:01.061874 133153302591296 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 16:39:06.521365 133028357601024 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_13_scan_layers_true/checkpoints/9/items/array_metadatas/process_5 I0422 16:39:43.619796 133027820730112 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 36.479 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.30133080482483 s) (per-host) I0422 16:39:43.619920 133027820730112 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.570788s. I0422 16:39:51.845887 133027820730112 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 50.796738s. I0422 16:39:51.846221 133027275466496 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 16:39:51.846334 133027275466496 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0422 16:39:51.846382 133027275466496 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0422 16:39:51.847936 133027275466496 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0422 16:39:51.848099 133153302591296 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0422 16:39:51.848243 133153302591296 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 16:39:51.848881 133153302591296 metric_logger.py:185] completed step: 9, seconds: 0.148, TFLOP/s/device: 92.087, Tokens/s/device: 13880.417, 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: Wed Apr 22 16:40:04 UTC 2026 EXIT_CODE=0