XPK Start: Thu Apr 23 15:53:19 UTC 2026
2026-04-23 15:53:23.680374: 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 00:00:1776959603.692731 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1776959603.696420 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
W0000 00:00:1776959603.707784 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776959603.707802 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776959603.707805 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776959603.707807 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
2026-04-23 15:53:42.817424: 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)
I0423 15:53:43.341212 133935404320576 max_utils.py:273] Attempting to initialize the jax distributed system...
INFO:2026-04-23 15:53:52,381:jax._src.distributed:140: Starting JAX distributed service on [::]:8482
I0423 15:53:52.381943 133935404320576 distributed.py:140] Starting JAX distributed service on [::]:8482
INFO:2026-04-23 15:53:52,384:jax._src.distributed:157: Connecting to JAX distributed service on mt-01-base-ax27p-slice-job-0-0.mt-01-base-ax27p:8482
I0423 15:53:52.384293 133935404320576 distributed.py:157] Connecting to JAX distributed service on mt-01-base-ax27p-slice-job-0-0.mt-01-base-ax27p:8482
I0423 15:53:53.352993 133935404320576 max_utils.py:284] Jax distributed system initialized!
I0423 15:54:00.471065 133935404320576 max_utils.py:800] System Information: Jax Version: 0.8.1
I0423 15:54:00.471171 133935404320576 max_utils.py:801] System Information: Jaxlib Version: 0.8.1
I0423 15:54:00.471215 133935404320576 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
I0423 15:54:00.471249 133935404320576 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0423 15:54:01.096784 133935404320576 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0423 15:54:01.097380 133935404320576 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0423 15:54:01.097565 133935404320576 checkpointing.py:688] Setting up checkpoint logger...
I0423 15:54:01.097616 133935404320576 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0423 15:54:01.097671 133935404320576 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None
I0423 15:54:01.098033 133935404320576 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 0x79cfc2575100>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0423 15:54:04.747052 133935404320576 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0423 15:54:04.747294 133935404320576 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 0x79bb584cd640>}, handler_registry=None
I0423 15:54:04.747527 133935404320576 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79bb584cd640>` 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`.
I0423 15:54:04.747577 133935404320576 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79bb584d2b40>` 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`.
I0423 15:54:04.747612 133935404320576 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 0x79bb584cd640>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79bb584cd640>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79bb584d2b40>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79bb584d2b40>}).
I0423 15:54:04.748203 133935404320576 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33
I0423 15:54:04.748285 133935404320576 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x79bb1830c180> timeout: 600 secs and primary_host=0 for async checkpoint writes
I0423 15:54:05.901348 133935404320576 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_01_base/checkpoints
I0423 15:54:05.943995 133935404320576 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_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_01_base/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x79bb584cb140>
I0423 15:54:05.944143 133935404320576 checkpointing.py:302] Checkpoint manager created!
I0423 15:54:06.846931 133935404320576 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0423 15:54:06.847035 133935404320576 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 15:54:07.217673 133935404320576 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0423 15:54:07.217768 133935404320576 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 15:54:07.233826 133935404320576 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0423 15:54:07.233883 133935404320576 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 15:54:07.256928 133935404320576 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 15:54:07.256995 133935404320576 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 15:54:07.273150 133935404320576 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 15:54:07.273220 133935404320576 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 15:54:07.289271 133935404320576 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 15:54:07.289335 133935404320576 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 15:54:07.314494 133935404320576 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0423 15:54:07.314562 133935404320576 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 15:54:07.334449 133935404320576 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0423 15:54:07.334515 133935404320576 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0423 15:54:07.524959 133935404320576 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0423 15:54:07.525068 133935404320576 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0423 15:54:08.960444 133935404320576 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: PartitionSpec('norm',)
Physical: (None,)
I0423 15:54:08.960574 133935404320576 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)
I0423 15:54:08.960630 133935404320576 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)
I0423 15:54:08.960847 133935404320576 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')
I0423 15:54:08.960928 133935404320576 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)
I0423 15:54:08.960976 133935404320576 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)
I0423 15:54:08.961044 133935404320576 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)
I0423 15:54:08.961103 133935404320576 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')
I0423 15:54:08.961145 133935404320576 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)
I0423 15:54:08.961184 133935404320576 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)
I0423 15:54:08.961235 133935404320576 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: PartitionSpec('embed', 'vocab')
Physical: ('fsdp', None)
I0423 15:54:08.961286 133935404320576 maxtext_utils.py:1740] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: PartitionSpec('vocab', 'embed')
Physical: (None, 'fsdp')
I0423 15:54:09.461430 133935404320576 train.py:158] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0423 15:54:09.461524 133935404320576 train.py:158] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0423 15:54:09.476511 133935404320576 train.py:165] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0423 15:54:09.476571 133935404320576 train.py:165] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0423 15:54:20.146362 133935404320576 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.
I0423 15:54:20.147216 133935404320576 metric_logger.py:289] number parameters: 1.104 billion
I0423 15:54:31.327280 133935404320576 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0423 15:54:31.485748 133935404320576 checkpointing.py:798] Waited 0.15845656394958496 seconds for step 0 to finish before starting checkpointing.
I0423 15:54:31.488251 133935404320576 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 15:54:31.489975 133935404320576 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0
I0423 15:54:31.492723 133935404320576 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_01_base/checkpoints/0.
I0423 15:54:32.090621 133935404320576 signaling_client.py:364] Using JaxDistributedSignalingClient
I0423 15:54:32.091556 133935404320576 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array.
I0423 15:54:32.091614 133935404320576 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
I0423 15:54:32.338380 133935404320576 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.247837s
I0423 15:54:32.338554 133935404320576 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.004 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5134434700012207 s) (per-host)
I0423 15:54:32.338608 133935404320576 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.513506s (batch_requests_ready=0.262010s, total_serialization_initiated=0.251425s, others=0.000071s)
I0423 15:54:32.338716 133935404320576 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.517914s (all_items=0.000021s, per_item={'items': '0.00002122'}, temp_paths=0.517893)
I0423 15:54:32.339634 133809744668416 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started.
I0423 15:54:32.339821 133935404320576 async_checkpointer.py:561] Finished blocking save. Time taken: 0.849779s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_01_base/checkpoints/0.
I0423 15:54:32.367893 133935404320576 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 15:54:32.368199 133808162600704 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0423 15:54:32.368332 133935404320576 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_01_base/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776959671.488232, 'wait_for_prev_duration_secs': 6.985664367675781e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776959671.4900134, 'checkpointer_blocking_duration_secs': 0.8499140739440918, 'get_old_steps_start_time': 1776959672.3399475, 'get_old_steps_duration_secs': 2.6226043701171875e-05, 'checkpoint_manager_blocking_start_time': 1776959671.486331, 'checkpoint_manager_blocking_duration_secs': 0.8819611072540283}
I0423 15:54:32.368435 133935404320576 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0423 15:54:32.368486 133935404320576 max_utils.py:750]
Memstats: After params initialized:
I0423 15:54:32.368537 133935404320576 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0))
I0423 15:54:32.368570 133935404320576 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0))
I0423 15:54:32.368598 133935404320576 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0))
I0423 15:54:32.368623 133935404320576 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0))
I0423 15:54:32.634752 133935404320576 metric_logger.py:185] completed step: 0, seconds: 11.180, TFLOP/s/device: 1.215, Tokens/s/device: 183.185, total_weights: 65536, loss: 10.874
I0423 15:54:32.805606 133935404320576 metric_logger.py:185] completed step: 1, seconds: 1.306, TFLOP/s/device: 10.402, Tokens/s/device: 1567.925, total_weights: 65536, loss: 10.874
I0423 15:54:33.194661 133935404320576 metric_logger.py:185] completed step: 2, seconds: 0.023, TFLOP/s/device: 579.186, Tokens/s/device: 87301.249, total_weights: 65536, loss: 10.263
I0423 15:54:33.342245 133935404320576 metric_logger.py:185] completed step: 3, seconds: 0.389, TFLOP/s/device: 34.889, Tokens/s/device: 5258.820, total_weights: 65536, loss: 9.734
I0423 15:54:33.637638 133935404320576 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.901, Tokens/s/device: 13400.072, total_weights: 65536, loss: 9.278
I0423 15:54:33.643277 133935404320576 metric_logger.py:185] completed step: 5, seconds: 0.148, TFLOP/s/device: 92.068, Tokens/s/device: 13877.501, total_weights: 65536, loss: 8.892
I0423 15:54:51.977465 133935404320576 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.945, Tokens/s/device: 6925.376, total_weights: 65536, loss: 8.592
I0423 15:54:52.125185 133935404320576 metric_logger.py:185] completed step: 7, seconds: 18.187, TFLOP/s/device: 0.747, Tokens/s/device: 112.606, total_weights: 65536, loss: 8.384
I0423 15:54:52.272642 133935404320576 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.261, Tokens/s/device: 13454.388, total_weights: 65536, loss: 8.255
I0423 15:54:52.419756 133935404320576 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0423 15:54:52.420409 133935404320576 checkpointing.py:798] Waited 0.0006728172302246094 seconds for step 9 to finish before starting checkpointing.
I0423 15:54:52.422459 133935404320576 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 15:54:53.634357 2633 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0423 15:54:55.564891 133809753061120 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_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_01_base/checkpoints/0/items/array_metadatas/process_5
I0423 15:55:27.881232 133809744668416 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 28.179 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 56.056079149246216 s) (per-host)
I0423 15:55:27.881354 133809744668416 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 55.541588s.
I0423 15:55:35.311566 133809744668416 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 62.971784s.
I0423 15:55:35.311867 133808162600704 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 15:55:35.312000 133808162600704 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0423 15:55:35.312052 133808162600704 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0423 15:55:35.313711 133808162600704 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0423 15:55:35.313909 133935404320576 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.
W0423 15:55:35.314042 133935404320576 checkpoint_manager.py:1458] Waiting for previous save to complete took 42.891585 seconds. If this number is high, consider checkpointing less frequently.
I0423 15:55:35.316023 133935404320576 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9
I0423 15:55:35.319504 133935404320576 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_01_base/checkpoints/9.
I0423 15:55:35.859715 133935404320576 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array.
I0423 15:55:35.859816 133935404320576 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
I0423 15:55:35.905445 133935404320576 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.046984s
I0423 15:55:35.905616 133935404320576 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.088 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3031895160675049 s) (per-host)
I0423 15:55:35.905685 133935404320576 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.303267s (batch_requests_ready=0.254343s, total_serialization_initiated=0.048839s, others=0.000086s)
I0423 15:55:35.905781 133935404320576 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.307509s (all_items=0.000014s, per_item={'items': '0.00001359'}, temp_paths=0.307496)
I0423 15:55:35.906682 133809178470144 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started.
I0423 15:55:35.906846 133935404320576 async_checkpointer.py:561] Finished blocking save. Time taken: 0.590754s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_01_base/checkpoints/9.
I0423 15:55:35.929013 133935404320576 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 15:55:35.929271 133808162600704 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0423 15:55:35.929403 133935404320576 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_01_base/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776959692.4224298, 'wait_for_prev_duration_secs': 42.89158511161804, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776959735.3160613, 'checkpointer_blocking_duration_secs': 0.5908920764923096, 'get_old_steps_start_time': 1776959735.9069726, 'get_old_steps_duration_secs': 2.8371810913085938e-05, 'checkpoint_manager_blocking_start_time': 1776959692.4206362, 'checkpoint_manager_blocking_duration_secs': 43.5087308883667}
I0423 15:55:35.929517 133935404320576 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0423 15:55:35.929562 133935404320576 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 15:55:40.732278 133809753061120 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_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_01_base/checkpoints/9/items/array_metadatas/process_5
I0423 15:56:17.495352 133809178470144 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.706 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.892884254455566 s) (per-host)
I0423 15:56:17.495477 133809178470144 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.588680s.
I0423 15:56:25.612553 133809178470144 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 49.705740s.
I0423 15:56:25.612793 133808162600704 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 15:56:25.612858 133808162600704 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0423 15:56:25.612914 133808162600704 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0423 15:56:25.614512 133808162600704 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0423 15:56:25.614696 133935404320576 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.
I0423 15:56:25.614839 133935404320576 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 15:56:25.615474 133935404320576 metric_logger.py:185] completed step: 9, seconds: 0.148, TFLOP/s/device: 91.940, Tokens/s/device: 13858.251, 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: Thu Apr 23 15:56:35 UTC 2026
EXIT_CODE=0