XPK Start: Sat Apr 18 00:46:03 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-18 00:46:27.476897: 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)
I0418 00:46:28.011082 133912997402432 max_utils.py:273] Attempting to initialize the jax distributed system...
I0418 00:46:37.051456 133912997402432 distributed.py:149] Starting JAX distributed service on [::]:8482
I0418 00:46:37.053887 133912997402432 distributed.py:172] Connecting to JAX distributed service on mt-10-shardy-true-mj668-slice-job-0-0.mt-10-shardy-true-mj668:8482
I0418 00:46:38.158460 133912997402432 max_utils.py:284] Jax distributed system initialized!
I0418 00:46:44.326398 133912997402432 max_utils.py:800] System Information: Jax Version: 0.9.2
I0418 00:46:44.326499 133912997402432 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0418 00:46:44.326540 133912997402432 max_utils.py:802] System Information: Jax Backend: PJRT C API
TFRT TPU v6 lite
Built on Mar 4 2026 11:32:08 (1772652728) cl/878335365
I0418 00:46:44.326574 133912997402432 train_utils.py:348] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0418 00:46:45.005068 133912997402432 maxtext_utils.py:1551] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0418 00:46:45.005335 133912997402432 checkpointing.py:677] Setting up checkpoint logger...
I0418 00:46:45.005390 133912997402432 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0418 00:46:45.005435 133912997402432 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0418 00:46:45.005778 133912997402432 base_pytree_checkpoint_handler.py:441] 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 0x79ca69b1cf20>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0418 00:46:48.695703 133912997402432 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0418 00:46:48.695963 133912997402432 checkpoint_manager.py:708] [process=0][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79b64069f9b0>}, handler_registry=None
I0418 00:46:48.696328 133912997402432 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79b64069f9b0>` 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`.
I0418 00:46:48.696398 133912997402432 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79b64054daf0>` 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`.
I0418 00:46:48.696447 133912997402432 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 0x79b64069f9b0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79b64069f9b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79b64054daf0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79b64054daf0>}).
I0418 00:46:48.696859 133912997402432 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0418 00:46:48.696960 133912997402432 async_checkpointer.py:192] [process=0][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x79b6402d7c40> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0418 00:46:50.252298 133912997402432 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints
I0418 00:46:51.125889 133912997402432 checkpoint_manager.py:929] [process=0][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_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x79b640643800>
I0418 00:46:51.126100 133912997402432 checkpointing.py:301] Checkpoint manager created!
I0418 00:46:52.026471 133912997402432 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0418 00:46:52.026587 133912997402432 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 00:46:52.404534 133912997402432 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 00:46:52.404628 133912997402432 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 00:46:52.420723 133912997402432 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 00:46:52.420783 133912997402432 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 00:46:52.444134 133912997402432 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 00:46:52.444202 133912997402432 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 00:46:52.460264 133912997402432 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 00:46:52.460322 133912997402432 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 00:46:52.476386 133912997402432 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 00:46:52.476445 133912997402432 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 00:46:52.501816 133912997402432 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0418 00:46:52.501901 133912997402432 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 00:46:52.522681 133912997402432 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0418 00:46:52.522748 133912997402432 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0418 00:46:52.748319 133912997402432 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0418 00:46:52.748425 133912997402432 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0418 00:46:54.173227 133912997402432 maxtext_utils.py:1654] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0418 00:46:54.173348 133912997402432 maxtext_utils.py:1654] params/params/decoder/layers/mlp/wi_0/kernel
Shape: float32[2048,16,7168]
Logical: P('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0418 00:46:54.173402 133912997402432 maxtext_utils.py:1654] params/params/decoder/layers/mlp/wi_1/kernel
Shape: float32[2048,16,7168]
Logical: P('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0418 00:46:54.173461 133912997402432 maxtext_utils.py:1654] params/params/decoder/layers/mlp/wo/kernel
Shape: float32[7168,16,2048]
Logical: P('mlp', 'layers', 'embed')
Physical: (None, None, 'fsdp')
I0418 00:46:54.173513 133912997402432 maxtext_utils.py:1654] params/params/decoder/layers/post_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: P('norm', 'layers')
Physical: (None, None)
I0418 00:46:54.173551 133912997402432 maxtext_utils.py:1654] params/params/decoder/layers/pre_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: P('norm', 'layers')
Physical: (None, None)
I0418 00:46:54.173605 133912997402432 maxtext_utils.py:1654] params/params/decoder/layers/self_attention/key/kernel
Shape: float32[2048,16,16,128]
Logical: P('embed', 'layers', 'kv_heads', 'kv_head_dim')
Physical: ('fsdp', None, None, None)
I0418 00:46:54.173660 133912997402432 maxtext_utils.py:1654] params/params/decoder/layers/self_attention/out/kernel
Shape: float32[16,16,128,2048]
Logical: P('heads', 'layers', 'kv', 'embed')
Physical: (None, None, None, 'fsdp')
I0418 00:46:54.173700 133912997402432 maxtext_utils.py:1654] params/params/decoder/layers/self_attention/query/kernel
Shape: float32[2048,16,16,128]
Logical: P('embed', 'layers', 'q_heads', 'kv')
Physical: ('fsdp', None, None, None)
I0418 00:46:54.173738 133912997402432 maxtext_utils.py:1654] params/params/decoder/layers/self_attention/value/kernel
Shape: float32[2048,16,16,128]
Logical: P('embed', 'layers', 'kv_heads', 'kv_head_dim')
Physical: ('fsdp', None, None, None)
I0418 00:46:54.173787 133912997402432 maxtext_utils.py:1654] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0418 00:46:54.173854 133912997402432 maxtext_utils.py:1654] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0418 00:46:54.665049 133912997402432 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0418 00:46:54.665151 133912997402432 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0418 00:46:54.680428 133912997402432 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0418 00:46:54.680489 133912997402432 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0418 00:47:05.428782 133912997402432 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.
I0418 00:47:05.436560 133912997402432 metric_logger.py:301] number parameters: 1.104 billion
I0418 00:47:16.868179 133912997402432 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0418 00:47:17.024369 133912997402432 checkpointing.py:776] Waited 0.156172513961792 seconds for step 0 to finish before starting checkpointing.
I0418 00:47:17.026897 133912997402432 checkpoint_manager.py:2009] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 00:47:17.028505 133912997402432 checkpoint_manager.py:1512] [process=0] Saving checkpoint at step 0
I0418 00:47:17.030234 133912997402432 event_tracking.py:70] [process=0] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0.
I0418 00:47:17.344182 133912997402432 signaling_client.py:364] Using JaxDistributedSignalingClient
I0418 00:47:17.358042 133912997402432 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 00:47:17.358114 133912997402432 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
I0418 00:47:17.627509 133784271845120 atomicity.py:140] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0
I0418 00:47:17.629489 133912997402432 base_pytree_checkpoint_handler.py:154] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.272246s
I0418 00:47:17.630362 133912997402432 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/blocking_gbytes_per_sec: 5.535 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.27867555618286133 s) (per-host)
I0418 00:47:17.630428 133912997402432 base_pytree_checkpoint_handler.py:768] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.278754s (batch_requests_ready=0.002210s, total_serialization_initiated=0.275763s, others=0.000782s)
I0418 00:47:17.630504 133912997402432 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.282687s (all_items=0.000016s, per_item={'items': '0.00001621'}, temp_paths=0.282670)
I0418 00:47:17.631273 133912997402432 event_tracking.py:125] [process=0] [async] Finished blocking save in 0.60 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0.
I0418 00:47:17.631545 133784348010240 async_checkpointer.py:76] [process=0][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-18 01:07:17.631521
I0418 00:47:17.646126 133912997402432 checkpoint_manager.py:1560] [process=0][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 00:47:17.646426 133782212429568 async_checkpointer.py:280] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0418 00:47:17.646581 133912997402432 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776473237.0268788, 'wait_for_prev_duration_secs': 6.175041198730469e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776473237.0285425, 'checkpointer_blocking_duration_secs': 0.6031005382537842, 'get_old_steps_start_time': 1776473237.6316695, 'get_old_steps_duration_secs': 2.9087066650390625e-05, 'checkpoint_manager_blocking_start_time': 1776473237.0248754, 'checkpoint_manager_blocking_duration_secs': 0.6216671466827393}
I0418 00:47:17.646685 133912997402432 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0418 00:47:17.646733 133912997402432 max_utils.py:750]
Memstats: After params initialized:
I0418 00:47:17.646781 133912997402432 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_0(process=0,(0,0,0,0))
I0418 00:47:17.646813 133912997402432 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_1(process=0,(1,0,0,0))
I0418 00:47:17.646856 133912997402432 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_4(process=0,(0,1,0,0))
I0418 00:47:17.646885 133912997402432 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_5(process=0,(1,1,0,0))
I0418 00:47:18.002093 133912997402432 metric_logger.py:196] completed step: 0, seconds: 11.127, TFLOP/s/device: 1.221, Tokens/s/device: 184.050, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0418 00:47:18.005065 133912997402432 metric_logger.py:281] To see full metrics 'tensorboard --logdir=gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/tensorboard/'
I0418 00:47:18.411373 133912997402432 metric_logger.py:196] completed step: 1, seconds: 1.132, TFLOP/s/device: 12.003, Tokens/s/device: 1809.149, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0418 00:47:18.557485 133912997402432 metric_logger.py:196] completed step: 2, seconds: 0.410, TFLOP/s/device: 33.139, Tokens/s/device: 4995.049, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0418 00:47:18.704592 133912997402432 metric_logger.py:196] completed step: 3, seconds: 0.009, TFLOP/s/device: 1463.184, Tokens/s/device: 220547.060, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0418 00:47:19.498151 133784255059712 atomicity.py:140] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0/items
I0418 00:47:20.054120 133781174548224 checkpoint.py:188] Wrote Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776473239230716536, 'commit_timestamp_nsecs': None, 'custom_metadata': {}}, json={"item_handlers": null, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1776473239230716536, "commit_timestamp_nsecs": null, "custom_metadata": {}} to gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0/_CHECKPOINT_METADATA
I0418 00:47:21.472116 2825 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0418 00:47:23.231019 133783273953024 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0/items/array_metadatas/process_0
I0418 00:47:42.809422 133912997402432 metric_logger.py:196] completed step: 4, seconds: 0.147, TFLOP/s/device: 92.338, Tokens/s/device: 13918.244, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0418 00:47:42.818345 133912997402432 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.521, Tokens/s/device: 13945.824, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0418 00:47:42.964588 133912997402432 metric_logger.py:196] completed step: 6, seconds: 24.104, TFLOP/s/device: 0.564, Tokens/s/device: 84.966, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0418 00:47:43.111880 133912997402432 metric_logger.py:196] completed step: 7, seconds: 0.007, TFLOP/s/device: 2088.721, Tokens/s/device: 314834.743, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0418 00:47:43.259277 133912997402432 metric_logger.py:196] completed step: 8, seconds: 0.147, TFLOP/s/device: 92.221, Tokens/s/device: 13900.484, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0418 00:47:43.405997 133912997402432 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0418 00:47:43.406695 133912997402432 checkpointing.py:776] Waited 0.0007178783416748047 seconds for step 9 to finish before starting checkpointing.
I0418 00:47:43.408780 133912997402432 checkpoint_manager.py:2020] [process=0][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 00:47:58.099158 133782258579200 base_pytree_checkpoint_handler.py:1282] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 36.635029s (commit=35.426209s, array_metadata_write=1.208820s)
I0418 00:47:58.100290 133784348010240 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/gbytes_per_sec: 38.764 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 40.74856400489807 s) (per-host)
I0418 00:47:58.100407 133784348010240 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 40.468787s.
I0418 00:47:59.675937 133784348010240 checkpoint.py:228] Read Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776473239230716536, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} from gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0/_CHECKPOINT_METADATA
I0418 00:48:00.440431 133784348010240 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000249s.
I0418 00:48:01.042503 133784348010240 ocdbt_utils.py:49] Param validation support for Zarr3 will be added later (b/362328389).
I0418 00:48:01.268749 133781174548224 checkpoint.py:247] Updated Metadata={'item_handlers': {'items': 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler'}, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776473239230716536, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} to gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0/_CHECKPOINT_METADATA
I0418 00:48:01.963259 133784348010240 base_pytree_checkpoint_handler.py:1406] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 2.161022s. use_zarr3=True, enable_post_merge_validation=True, directory=gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0/items
I0418 00:48:01.964919 133784348010240 atomicity.py:666] Finalizing gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0/items
I0418 00:48:02.477028 133784348010240 atomicity.py:666] Finalizing gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0
I0418 00:48:03.904472 133784348010240 atomicity.py:847] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0`.
I0418 00:48:03.905412 133784348010240 event_tracking.py:138] [process=0] [async] Finished save (blocking + background) in 46.88 seconds @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/0
I0418 00:48:03.906800 133784348010240 async_checkpointer.py:160] [process=0][thread=async_save] Background save thread done. Time taken: 46.275176s.
I0418 00:48:03.907011 133782212429568 async_checkpointer.py:288] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 00:48:03.907131 133782212429568 async_checkpointer.py:298] [process=0][thread=save_finalize] No errors found in background save thread=async_save.
I0418 00:48:03.907197 133782212429568 checkpoint_manager.py:2137] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0418 00:48:03.910171 133782212429568 checkpoint_manager.py:2146] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0418 00:48:03.910356 133912997402432 checkpoint_manager.py:2032] [process=0][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0418 00:48:03.910521 133912997402432 checkpoint_manager.py:1452] Waiting for previous save to complete took 20.501735 seconds. If this number is high, consider checkpointing less frequently.
I0418 00:48:03.912424 133912997402432 checkpoint_manager.py:1512] [process=0] Saving checkpoint at step 9
I0418 00:48:03.914521 133912997402432 event_tracking.py:70] [process=0] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/9.
I0418 00:48:04.625830 133912997402432 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 00:48:04.625936 133912997402432 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
I0418 00:48:04.651157 133912997402432 base_pytree_checkpoint_handler.py:154] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.026071s
I0418 00:48:04.651824 133912997402432 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/blocking_gbytes_per_sec: 50.189 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.030735492706298828 s) (per-host)
I0418 00:48:04.651910 133912997402432 base_pytree_checkpoint_handler.py:768] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.030835s (batch_requests_ready=0.001809s, total_serialization_initiated=0.028388s, others=0.000638s)
I0418 00:48:04.651979 133912997402432 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.034710s (all_items=0.000016s, per_item={'items': '0.00001597'}, temp_paths=0.034694)
I0418 00:48:04.652608 133912997402432 event_tracking.py:125] [process=0] [async] Finished blocking save in 0.74 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/9.
I0418 00:48:04.652939 133782212429568 async_checkpointer.py:76] [process=0][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-18 01:08:04.652902
I0418 00:48:04.662325 133912997402432 checkpoint_manager.py:1560] [process=0][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 00:48:04.662591 133782747305728 async_checkpointer.py:280] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0418 00:48:04.662725 133912997402432 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776473263.4087472, 'wait_for_prev_duration_secs': 20.501734972000122, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776473283.912471, 'checkpointer_blocking_duration_secs': 0.7406136989593506, 'get_old_steps_start_time': 1776473284.6531096, 'get_old_steps_duration_secs': 2.9325485229492188e-05, 'checkpoint_manager_blocking_start_time': 1776473263.406945, 'checkpoint_manager_blocking_duration_secs': 21.25574517250061}
I0418 00:48:04.662828 133912997402432 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0418 00:48:04.662884 133912997402432 checkpoint_manager.py:2020] [process=0][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 00:48:05.720541 133784348010240 atomicity.py:140] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/9
I0418 00:48:07.893657 133784221488896 atomicity.py:140] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/9/items
I0418 00:48:11.175286 133782764091136 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/9/items/array_metadatas/process_0
I0418 00:48:51.404447 133782755698432 base_pytree_checkpoint_handler.py:1282] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 41.980658s (commit=40.815427s, array_metadata_write=1.165231s)
I0418 00:48:51.405465 133782212429568 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/gbytes_per_sec: 33.763 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 46.78437042236328 s) (per-host)
I0418 00:48:51.405529 133782212429568 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 46.752475s.
I0418 00:48:53.458345 133782212429568 ocdbt_utils.py:49] Param validation support for Zarr3 will be added later (b/362328389).
I0418 00:48:53.709131 133782212429568 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000238s.
I0418 00:48:54.536350 133782212429568 base_pytree_checkpoint_handler.py:1406] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 2.316279s. use_zarr3=True, enable_post_merge_validation=True, directory=gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/9/items
I0418 00:48:54.537871 133782212429568 atomicity.py:666] Finalizing gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/9/items
I0418 00:48:55.070326 133782212429568 atomicity.py:666] Finalizing gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/9
I0418 00:48:57.340773 133782212429568 atomicity.py:847] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/9`.
I0418 00:48:57.341659 133782212429568 event_tracking.py:138] [process=0] [async] Finished save (blocking + background) in 53.43 seconds @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_10_shardy_true/checkpoints/9
I0418 00:48:57.343081 133782212429568 async_checkpointer.py:160] [process=0][thread=async_save] Background save thread done. Time taken: 52.690023s.
I0418 00:48:57.343210 133782747305728 async_checkpointer.py:288] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 00:48:57.343270 133782747305728 async_checkpointer.py:298] [process=0][thread=save_finalize] No errors found in background save thread=async_save.
I0418 00:48:57.343327 133782747305728 checkpoint_manager.py:2137] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0418 00:48:57.345544 133782747305728 checkpoint_manager.py:2146] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0418 00:48:57.345672 133912997402432 checkpoint_manager.py:2032] [process=0][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0418 00:48:57.345764 133912997402432 checkpoint_manager.py:2009] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 00:48:57.346807 133912997402432 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.289, Tokens/s/device: 13910.775, total_weights: 65536, loss: 8.179, lm_loss: 8.179, perplexity: 3564.635
Per train step:
Total TFLOPs: 13.59
split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Sat Apr 18 00:49:09 UTC 2026
EXIT_CODE=0