XPK Start: Sun Apr 19 19:28:07 UTC 2026 PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. 2026-04-19 19:28:31.153277: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303) I0419 19:28:31.689905 134495027705664 max_utils.py:273] Attempting to initialize the jax distributed system... I0419 19:28:40.729349 134495027705664 distributed.py:149] Starting JAX distributed service on [::]:8482 I0419 19:28:40.731693 134495027705664 distributed.py:172] Connecting to JAX distributed service on mt-15-ocdbt-false-rs-rk6tn-slice-job-0-0.mt-15-ocdbt-false-rs-rk6tn:8482 I0419 19:28:42.567483 134495027705664 max_utils.py:284] Jax distributed system initialized! I0419 19:28:47.594939 134495027705664 max_utils.py:800] System Information: Jax Version: 0.9.2 I0419 19:28:47.595040 134495027705664 max_utils.py:801] System Information: Jaxlib Version: 0.9.2 I0419 19:28:47.595080 134495027705664 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 I0419 19:28:47.595114 134495027705664 train_utils.py:348] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0419 19:28:48.283844 134495027705664 maxtext_utils.py:1551] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 19:28:48.284117 134495027705664 checkpointing.py:677] Setting up checkpoint logger... I0419 19:28:48.284168 134495027705664 checkpointing.py:233] Creating checkpoint manager with ocdbt=False and zarr3=True I0419 19:28:48.284212 134495027705664 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None I0419 19:28:48.284570 134495027705664 base_pytree_checkpoint_handler.py:441] Created BasePyTreeCheckpointHandler: use_ocdbt=False, use_zarr3=True, pytree_metadata_options=PyTreeMetadataOptions(support_rich_types=False), array_metadata_store=<orbax.checkpoint._src.metadata.array_metadata_store.Store object at 0x7a51d8f12510>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0419 19:28:50.016133 134495027705664 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0419 19:28:50.016383 134495027705664 checkpoint_manager.py:708] [process=6][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a3e74162c30>}, handler_registry=None I0419 19:28:50.016675 134495027705664 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a3e74162c30>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`. I0419 19:28:50.016737 134495027705664 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a3e54641340>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`. I0419 19:28:50.016787 134495027705664 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 0x7a3e74162c30>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a3e74162c30>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a3e54641340>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a3e54641340>}). I0419 19:28:50.017184 134495027705664 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34 I0419 19:28:50.017275 134495027705664 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7a3e543d3880> timeout: 1200 secs and primary_host=0 for async checkpoint writes I0419 19:28:51.508567 134376465606400 checkpoint.py:228] Read Metadata={'item_handlers': {'items': 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler'}, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776626796744937854, 'commit_timestamp_nsecs': 1776626810744181693, 'custom_metadata': {}} from gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/9/_CHECKPOINT_METADATA I0419 19:28:51.758234 134376465606400 checkpoint_manager.py:1871] Missing metrics for step 9 I0419 19:28:52.393266 134495027705664 checkpoint_manager.py:1812] Found 3 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints I0419 19:28:53.481575 134495027705664 checkpoint_manager.py:929] [process=6][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=5), 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_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7a3e5479f080> I0419 19:28:53.481743 134495027705664 checkpointing.py:301] Checkpoint manager created! I0419 19:28:54.405629 134495027705664 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0419 19:28:54.405743 134495027705664 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 19:28:54.786887 134495027705664 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 19:28:54.786973 134495027705664 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 19:28:54.803045 134495027705664 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 19:28:54.803103 134495027705664 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 19:28:54.827314 134495027705664 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 19:28:54.827387 134495027705664 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 19:28:54.843580 134495027705664 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 19:28:54.843637 134495027705664 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 19:28:54.859764 134495027705664 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 19:28:54.859825 134495027705664 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 19:28:54.884225 134495027705664 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0419 19:28:54.884295 134495027705664 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 19:28:54.905179 134495027705664 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0419 19:28:54.905243 134495027705664 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0419 19:28:55.121515 134495027705664 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0419 19:28:55.121622 134495027705664 checkpointing.py:581] restoring from this run's directory step 9 I0419 19:28:55.792503 134495027705664 event_tracking.py:70] [process=6] [sync] Started load checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/9. I0419 19:28:56.589746 134495027705664 checkpointer.py:307] Restoring checkpoint from gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/9. I0419 19:28:56.590173 134495027705664 event_tracking.py:125] [process=6] [sync] Finished blocking load in 0.80 seconds. Continuing load @ gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/9. I0419 19:28:59.510375 1440 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0419 19:29:09.290030 134495027705664 jax_array_handlers.py:843] [process=6] /jax/orbax/read/worker/io/requested throughput: 142.974 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 11.052903413772583 s) (per-host) I0419 19:29:09.296737 134495027705664 base_pytree_checkpoint_handler.py:130] [process=6] /jax/checkpoint/read/gbytes_per_sec: 138.076 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 11.461746215820312 s) (per-host) I0419 19:29:09.298686 134495027705664 event_tracking.py:138] [process=6] [sync] Finished load in 13.51 seconds @ gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/9 I0419 19:29:09.298790 134495027705664 standard_logger.py:34] {'step': 9, 'event_type': 'restore', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints', 'checkpointer_start_time': 1776626935.7923887, 'checkpointer_duration_secs': 13.506367444992065, 'checkpoint_manager_start_time': 1776626935.1218598, 'checkpoint_manager_duration_secs': 14.176898717880249} fsdp: 32 I0419 19:29:09.711222 134495027705664 maxtext_utils.py:1654] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 19:29:09.711347 134495027705664 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) I0419 19:29:09.711482 134495027705664 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) I0419 19:29:09.711585 134495027705664 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') I0419 19:29:09.711651 134495027705664 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) I0419 19:29:09.711694 134495027705664 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) I0419 19:29:09.711750 134495027705664 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) I0419 19:29:09.711807 134495027705664 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') I0419 19:29:09.711848 134495027705664 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) I0419 19:29:09.711884 134495027705664 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) I0419 19:29:09.711931 134495027705664 maxtext_utils.py:1654] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: P('embed_vocab', 'vocab') Physical: ('fsdp', None) I0419 19:29:09.711986 134495027705664 maxtext_utils.py:1654] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: P('vocab', 'embed_vocab') Physical: (None, 'fsdp') I0419 19:29:10.193549 134495027705664 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 19:29:10.193643 134495027705664 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0419 19:29:10.208887 134495027705664 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 19:29:10.208950 134495027705664 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0419 19:29:20.657840 134495027705664 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. I0419 19:29:20.658544 134495027705664 metric_logger.py:301] number parameters: 1.104 billion I0419 19:29:31.826677 134495027705664 checkpointing.py:772] Waiting for step 10 to finish before checkpoint... I0419 19:29:32.426343 134495027705664 checkpointing.py:776] Waited 0.5996451377868652 seconds for step 10 to finish before starting checkpointing. I0419 19:29:32.428747 134495027705664 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 19:29:32.430610 134495027705664 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 10 I0419 19:29:32.432313 134495027705664 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/10. I0419 19:29:32.766225 134495027705664 signaling_client.py:364] Using JaxDistributedSignalingClient I0419 19:29:32.767208 134495027705664 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0419 19:29:32.767266 134495027705664 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0419 19:29:33.046061 134495027705664 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.279912s I0419 19:29:33.046245 134495027705664 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.436 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28375768661499023 s) (per-host) I0419 19:29:33.046311 134495027705664 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.283833s (batch_requests_ready=0.002128s, total_serialization_initiated=0.281618s, others=0.000086s) I0419 19:29:33.046475 134495027705664 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.287761s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.287744) I0419 19:29:33.047288 134495027705664 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.62 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/10. I0419 19:29:33.047585 134373714142976 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 19:49:33.047555 I0419 19:29:33.059876 134495027705664 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=10] Starting CheckpointManager Save Finalize thread=save_finalize I0419 19:29:33.060189 134372122597120 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0419 19:29:33.060341 134495027705664 standard_logger.py:34] {'step': 10, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776626972.4287293, 'wait_for_prev_duration_secs': 6.008148193359375e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776626972.4306495, 'checkpointer_blocking_duration_secs': 0.6170816421508789, 'get_old_steps_start_time': 1776626973.0477543, 'get_old_steps_duration_secs': 2.9802322387695312e-05, 'checkpoint_manager_blocking_start_time': 1776626972.4267511, 'checkpoint_manager_blocking_duration_secs': 0.633554220199585} I0419 19:29:33.060455 134495027705664 checkpointing.py:408] Started an asynchronous checkpoint save for step 10 I0419 19:29:33.060507 134495027705664 max_utils.py:750] Memstats: After params initialized: I0419 19:29:33.060555 134495027705664 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0)) I0419 19:29:33.060586 134495027705664 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0)) I0419 19:29:33.060612 134495027705664 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0)) I0419 19:29:33.060635 134495027705664 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0)) I0419 19:29:33.378716 134495027705664 metric_logger.py:196] completed step: 10, seconds: 11.168, TFLOP/s/device: 1.217, Tokens/s/device: 183.381, total_weights: 65536, loss: 6.833, lm_loss: 6.833, perplexity: 927.869 I0419 19:29:33.541305 134495027705664 metric_logger.py:196] completed step: 11, seconds: 1.551, TFLOP/s/device: 8.763, Tokens/s/device: 1320.859, total_weights: 65536, loss: 6.527, lm_loss: 6.527, perplexity: 683.659 I0419 19:29:33.688813 134495027705664 metric_logger.py:196] completed step: 12, seconds: 0.015, TFLOP/s/device: 882.338, Tokens/s/device: 132995.649, total_weights: 65536, loss: 6.265, lm_loss: 6.265, perplexity: 526.068 I0419 19:29:33.836343 134495027705664 metric_logger.py:196] completed step: 13, seconds: 0.153, TFLOP/s/device: 88.621, Tokens/s/device: 13357.858, total_weights: 65536, loss: 6.045, lm_loss: 6.045, perplexity: 422.001 I0419 19:29:33.840819 134495027705664 checkpointing.py:772] Waiting for step 15 to finish before checkpoint... I0419 19:29:34.130555 134495027705664 checkpointing.py:776] Waited 0.289700984954834 seconds for step 15 to finish before starting checkpointing. I0419 19:29:34.133642 134495027705664 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=10][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 19:29:38.953623 134372130989824 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/10/items/array_metadatas/process_6 I0419 19:29:43.939683 134373714142976 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 141.324 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 11.177162647247314 s) (per-host) I0419 19:29:43.939816 134373714142976 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 10.892119s. I0419 19:29:51.817016 134373714142976 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 18.769303s. I0419 19:29:51.817304 134372122597120 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 19:29:51.817451 134372122597120 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0419 19:29:51.817500 134372122597120 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=10] CheckpointManager Save Finalize is syncing with other hosts... I0419 19:29:51.819076 134372122597120 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=10] CheckpointManager Save Finalize is done on all hosts. I0419 19:29:51.819267 134495027705664 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=10][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=10. W0419 19:29:51.819399 134495027705664 checkpoint_manager.py:1452] Waiting for previous save to complete took 17.685758 seconds. If this number is high, consider checkpointing less frequently. I0419 19:29:51.821401 134495027705664 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 15 I0419 19:29:51.823915 134495027705664 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/15. I0419 19:29:52.111893 134495027705664 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0419 19:29:52.111994 134495027705664 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0419 19:29:52.153113 134495027705664 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.042382s I0419 19:29:52.153285 134495027705664 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 33.697 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04577779769897461 s) (per-host) I0419 19:29:52.153337 134495027705664 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.045840s (batch_requests_ready=0.001733s, total_serialization_initiated=0.044040s, others=0.000067s) I0419 19:29:52.153441 134495027705664 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.050765s (all_items=0.000013s, per_item={'items': '0.00001264'}, temp_paths=0.050752) I0419 19:29:52.154171 134495027705664 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.33 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/15. I0419 19:29:52.154533 134372122597120 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 19:49:52.154495 I0419 19:29:52.566959 134495027705664 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=15] Starting CheckpointManager Save Finalize thread=save_finalize I0419 19:29:52.567394 134372097419008 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0419 19:29:52.567599 134495027705664 standard_logger.py:34] {'step': 15, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776626974.1336124, 'wait_for_prev_duration_secs': 17.685757637023926, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776626991.821451, 'checkpointer_blocking_duration_secs': 0.3331878185272217, 'get_old_steps_start_time': 1776626992.1546648, 'get_old_steps_duration_secs': 3.0040740966796875e-05, 'checkpoint_manager_blocking_start_time': 1776626974.1311886, 'checkpoint_manager_blocking_duration_secs': 18.43637180328369} I0419 19:29:52.567723 134495027705664 checkpointing.py:408] Started an asynchronous checkpoint save for step 15 I0419 19:29:52.568656 134495027705664 metric_logger.py:196] completed step: 14, seconds: 0.147, TFLOP/s/device: 92.135, Tokens/s/device: 13887.570, total_weights: 65536, loss: 5.864, lm_loss: 5.864, perplexity: 352.193 I0419 19:29:52.574960 134495027705664 metric_logger.py:196] completed step: 15, seconds: 0.147, TFLOP/s/device: 92.214, Tokens/s/device: 13899.541, total_weights: 65536, loss: 5.719, lm_loss: 5.719, perplexity: 304.661 I0419 19:29:58.377666 134372130989824 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/15/items/array_metadatas/process_6 I0419 19:30:03.182774 134372122597120 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 142.625 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 11.075209140777588 s) (per-host) I0419 19:30:03.182893 134372122597120 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 11.027726s. I0419 19:30:09.140654 134372122597120 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 16.985471s. I0419 19:30:09.140961 134372097419008 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 19:30:09.141083 134372097419008 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0419 19:30:09.141132 134372097419008 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=15] CheckpointManager Save Finalize is syncing with other hosts... I0419 19:30:09.142677 134372097419008 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=15] CheckpointManager Save Finalize is done on all hosts. I0419 19:30:14.236057 134495027705664 metric_logger.py:196] completed step: 16, seconds: 18.733, TFLOP/s/device: 0.725, Tokens/s/device: 109.328, total_weights: 65536, loss: 5.605, lm_loss: 5.605, perplexity: 271.872 I0419 19:30:14.383495 134495027705664 metric_logger.py:196] completed step: 17, seconds: 21.515, TFLOP/s/device: 0.632, Tokens/s/device: 95.190, total_weights: 65536, loss: 5.517, lm_loss: 5.517, perplexity: 248.975 I0419 19:30:14.530914 134495027705664 metric_logger.py:196] completed step: 18, seconds: 0.152, TFLOP/s/device: 89.144, Tokens/s/device: 13436.733, total_weights: 65536, loss: 5.448, lm_loss: 5.448, perplexity: 232.333 I0419 19:30:14.677895 134495027705664 checkpointing.py:772] Waiting for step 19 to finish before checkpoint... I0419 19:30:14.678621 134495027705664 checkpointing.py:776] Waited 0.0007464885711669922 seconds for step 19 to finish before starting checkpointing. I0419 19:30:14.680581 134495027705664 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 19:30:14.682069 134495027705664 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 19 I0419 19:30:14.683529 134495027705664 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/19. I0419 19:30:14.978277 134495027705664 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0419 19:30:14.978369 134495027705664 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0419 19:30:15.018652 134495027705664 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.041265s I0419 19:30:15.018813 134495027705664 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 34.494 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04472064971923828 s) (per-host) I0419 19:30:15.018862 134495027705664 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.044780s (batch_requests_ready=0.001664s, total_serialization_initiated=0.043052s, others=0.000064s) I0419 19:30:15.018950 134495027705664 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.048949s (all_items=0.000010s, per_item={'items': '0.00000954'}, temp_paths=0.048940) I0419 19:30:15.019649 134495027705664 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.34 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/19. I0419 19:30:15.019927 134372097419008 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 19:50:15.019895 I0419 19:30:15.021836 134495027705664 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=19] Starting CheckpointManager Save Finalize thread=save_finalize I0419 19:30:15.022059 134367271692032 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0419 19:30:15.022159 134495027705664 standard_logger.py:34] {'step': 19, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776627014.680549, 'wait_for_prev_duration_secs': 7.224082946777344e-05, 'time_between_consecutive_saves_sec': 5.537832975387573, 'checkpointer_blocking_start_time': 1776627014.682107, 'checkpointer_blocking_duration_secs': 0.3379549980163574, 'get_old_steps_start_time': 1776627015.0200799, 'get_old_steps_duration_secs': 2.5510787963867188e-05, 'checkpoint_manager_blocking_start_time': 1776627014.6788468, 'checkpoint_manager_blocking_duration_secs': 0.34328174591064453} I0419 19:30:15.022310 134495027705664 checkpointing.py:408] Started an asynchronous checkpoint save for step 19 I0419 19:30:15.022357 134495027705664 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=19][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 19:30:20.905473 134372130989824 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260419_180002/linen_xpk_main_20260419_180002_15_ocdbt_false/checkpoints/19/items/array_metadatas/process_6 I0419 19:30:25.625308 134372097419008 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 148.302 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 10.651185989379883 s) (per-host) I0419 19:30:25.625432 134372097419008 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 10.605399s. I0419 19:30:31.519547 134372097419008 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 16.499499s. I0419 19:30:31.519772 134367271692032 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 19:30:31.519828 134367271692032 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0419 19:30:31.519867 134367271692032 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=19] CheckpointManager Save Finalize is syncing with other hosts... I0419 19:30:31.522150 134367271692032 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=19] CheckpointManager Save Finalize is done on all hosts. I0419 19:30:31.522449 134495027705664 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=19][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=19. I0419 19:30:31.522596 134495027705664 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 19:30:31.523579 134495027705664 metric_logger.py:196] completed step: 19, seconds: 0.147, TFLOP/s/device: 92.280, Tokens/s/device: 13909.453, total_weights: 65536, loss: 5.391, lm_loss: 5.391, perplexity: 219.349 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Sun Apr 19 19:30:42 UTC 2026 EXIT_CODE=0