XPK Start: Wed Apr 22 16:22:16 UTC 2026 2026-04-22 16:22:20.665225: 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 00:00:1776874940.677376 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776874940.681108 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776874940.692325 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776874940.692342 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776874940.692344 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776874940.692346 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-22 16:22:39.866744: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303) I0422 16:22:40.394397 135514368800576 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-22 16:22:49,435:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0422 16:22:49.435023 135514368800576 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-22 16:22:49,437:jax._src.distributed:157: Connecting to JAX distributed service on mt-09-pdb-lt-1-tqrcm-slice-job-0-0.mt-09-pdb-lt-1-tqrcm:8482 I0422 16:22:49.437376 135514368800576 distributed.py:157] Connecting to JAX distributed service on mt-09-pdb-lt-1-tqrcm-slice-job-0-0.mt-09-pdb-lt-1-tqrcm:8482 I0422 16:22:50.527146 135514368800576 max_utils.py:284] Jax distributed system initialized! I0422 16:22:57.426636 135514368800576 max_utils.py:800] System Information: Jax Version: 0.8.1 I0422 16:22:57.426759 135514368800576 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0422 16:22:57.426799 135514368800576 max_utils.py:802] System Information: Jax Backend: PJRT C API TFRT TPU v6 lite Built on Nov 12 2025 14:16:36 (1762985796) cl/831091709 I0422 16:22:57.426833 135514368800576 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0422 16:22:58.076742 135514368800576 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 8, 1, 1, 1, 1, 4, 1, 1, 1, 1) I0422 16:22:58.077358 135514368800576 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 8, 1, 1, 1, 1, 4, 1, 1, 1, 1) I0422 16:22:58.077546 135514368800576 checkpointing.py:688] Setting up checkpoint logger... I0422 16:22:58.077599 135514368800576 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0422 16:22:58.077645 135514368800576 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0422 16:22:58.078014 135514368800576 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 0x7b3f59f7d460>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0422 16:23:01.023222 135514368800576 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0422 16:23:01.023467 135514368800576 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 0x7b2c13959dc0>}, handler_registry=None I0422 16:23:01.023717 135514368800576 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b2c13959dc0>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`. I0422 16:23:01.023767 135514368800576 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b2c13d2b680>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`. I0422 16:23:01.023804 135514368800576 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 0x7b2c13959dc0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b2c13959dc0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b2c13d2b680>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b2c13d2b680>}). I0422 16:23:01.024314 135514368800576 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0422 16:23:01.024391 135514368800576 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7b2c133d4180> timeout: 600 secs and primary_host=0 for async checkpoint writes I0422 16:23:02.205456 135514368800576 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_09_pdb_lt_1/checkpoints I0422 16:23:02.660211 135514368800576 checkpoint_manager.py:929] [process=5][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_09_pdb_lt_1/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7b3e20b7cbf0> I0422 16:23:02.660388 135514368800576 checkpointing.py:302] Checkpoint manager created! I0422 16:23:03.557135 135514368800576 nnx_wrappers.py:455] Unknown Logical: bfloat16[8,2048,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0422 16:23:03.557240 135514368800576 nnx_wrappers.py:455] Unknown Physical: bfloat16[8,2048,2048]....................................... ('fsdp', None, 'tensor'). I0422 16:23:03.928444 135514368800576 attentions.py:1084] attentions/inputs_q Logical: bfloat16[8,2048,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 16:23:03.928540 135514368800576 attentions.py:1084] attentions/inputs_q Physical: bfloat16[8,2048,2048]....................................... ('fsdp', None, 'tensor'). I0422 16:23:03.944377 135514368800576 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[8,2048,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 16:23:03.944440 135514368800576 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[8,2048,2048]....................................... ('fsdp', None, 'tensor'). I0422 16:23:03.967398 135514368800576 attentions.py:1150] attentions/query Logical: bfloat16[8,2048,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 16:23:03.967467 135514368800576 attentions.py:1150] attentions/query Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0422 16:23:03.983424 135514368800576 attentions.py:1151] attentions/key Logical: bfloat16[8,2048,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 16:23:03.983489 135514368800576 attentions.py:1151] attentions/key Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0422 16:23:03.999422 135514368800576 attentions.py:1152] attentions/value Logical: bfloat16[8,2048,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 16:23:03.999484 135514368800576 attentions.py:1152] attentions/value Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0422 16:23:04.024179 135514368800576 attentions.py:1193] attentions/out Logical: bfloat16[8,2048,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0422 16:23:04.024257 135514368800576 attentions.py:1193] attentions/out Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0422 16:23:04.043996 135514368800576 linears.py:541] linears/x Logical: bfloat16[8,2048,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0422 16:23:04.044061 135514368800576 linears.py:541] linears/x Physical: bfloat16[8,2048,7168]....................................... ('fsdp', None, 'tensor'). I0422 16:23:04.234361 135514368800576 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0422 16:23:04.234469 135514368800576 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 8 tensor: 4 I0422 16:23:05.659588 135514368800576 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: ('tensor',) I0422 16:23:05.659746 135514368800576 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, 'tensor') I0422 16:23:05.659805 135514368800576 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, 'tensor') I0422 16:23:05.659865 135514368800576 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: PartitionSpec('mlp', 'layers', 'embed') Physical: ('tensor', None, 'fsdp') I0422 16:23:05.659918 135514368800576 maxtext_utils.py:1740] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: ('tensor', None) I0422 16:23:05.659960 135514368800576 maxtext_utils.py:1740] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: ('tensor', None) I0422 16:23:05.660014 135514368800576 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, 'tensor', None) I0422 16:23:05.660066 135514368800576 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: ('tensor', None, None, 'fsdp') I0422 16:23:05.660109 135514368800576 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, 'tensor', None) I0422 16:23:05.660147 135514368800576 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, 'tensor', None) I0422 16:23:05.660201 135514368800576 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', 'tensor') I0422 16:23:05.660250 135514368800576 maxtext_utils.py:1740] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: ('tensor', 'fsdp') I0422 16:23:06.158844 135514368800576 train.py:158] train/xent Logical: float32[8,2048]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0422 16:23:06.158934 135514368800576 train.py:158] train/xent Physical: float32[8,2048]............................................. ('fsdp', None). I0422 16:23:06.173758 135514368800576 train.py:165] train/z_loss Logical: float32[8,2048]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0422 16:23:06.173817 135514368800576 train.py:165] train/z_loss Physical: float32[8,2048]............................................. ('fsdp', None). I0422 16:23:16.274143 135514368800576 max_utils.py:791] Total memory size: 0.9 GB, Output size: 0.4 GB, Temp size: 0.5 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0422 16:23:16.275007 135514368800576 metric_logger.py:289] number parameters: 1.104 billion I0422 16:23:17.798767 135514368800576 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0422 16:23:26.209871 135514368800576 checkpointing.py:798] Waited 8.411085605621338 seconds for step 0 to finish before starting checkpointing. I0422 16:23:26.212299 135514368800576 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 16:23:26.213931 135514368800576 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0422 16:23:26.217133 135514368800576 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_09_pdb_lt_1/checkpoints/0. I0422 16:23:27.670447 135514368800576 signaling_client.py:364] Using JaxDistributedSignalingClient I0422 16:23:27.671670 135514368800576 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0422 16:23:27.671731 135514368800576 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0422 16:23:27.897944 135514368800576 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.227580s I0422 16:23:27.898120 135514368800576 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.652 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.9335174560546875 s) (per-host) I0422 16:23:27.898175 135514368800576 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.933582s (batch_requests_ready=0.702370s, total_serialization_initiated=0.231140s, others=0.000073s) I0422 16:23:27.898270 135514368800576 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.938300s (all_items=0.000019s, per_item={'items': '0.00001907'}, temp_paths=0.938281) I0422 16:23:27.899239 135391881434880 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0422 16:23:27.899413 135514368800576 async_checkpointer.py:561] Finished blocking save. Time taken: 1.685410s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_09_pdb_lt_1/checkpoints/0. I0422 16:23:27.901358 135514368800576 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0422 16:23:27.901616 135388702172928 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0422 16:23:27.901787 135514368800576 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_09_pdb_lt_1/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776875006.2122812, 'wait_for_prev_duration_secs': 6.222724914550781e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776875006.2139692, 'checkpointer_blocking_duration_secs': 1.6855530738830566, 'get_old_steps_start_time': 1776875007.8995438, 'get_old_steps_duration_secs': 2.4557113647460938e-05, 'checkpoint_manager_blocking_start_time': 1776875006.2103102, 'checkpoint_manager_blocking_duration_secs': 1.6914386749267578} I0422 16:23:27.901891 135514368800576 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0422 16:23:27.901947 135514368800576 max_utils.py:750] Memstats: After params initialized: I0422 16:23:27.901998 135514368800576 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0)) I0422 16:23:27.902032 135514368800576 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0)) I0422 16:23:27.902060 135514368800576 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0)) I0422 16:23:27.902087 135514368800576 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0)) I0422 16:23:28.170248 135514368800576 metric_logger.py:185] completed step: 0, seconds: 1.524, TFLOP/s/device: 2.229, Tokens/s/device: 336.038, total_weights: 16384, loss: 10.862 I0422 16:23:28.249341 135514368800576 metric_logger.py:185] completed step: 1, seconds: 10.370, TFLOP/s/device: 0.328, Tokens/s/device: 49.372, total_weights: 16384, loss: 10.862 I0422 16:23:28.650942 135514368800576 metric_logger.py:185] completed step: 2, seconds: 0.021, TFLOP/s/device: 164.589, Tokens/s/device: 24808.605, total_weights: 16384, loss: 9.763 I0422 16:23:28.709483 135514368800576 metric_logger.py:185] completed step: 3, seconds: 0.402, TFLOP/s/device: 8.450, Tokens/s/device: 1273.686, total_weights: 16384, loss: 8.823 I0422 16:23:28.826938 135514368800576 metric_logger.py:185] completed step: 4, seconds: 0.064, TFLOP/s/device: 53.303, Tokens/s/device: 8034.397, total_weights: 16384, loss: 7.947 I0422 16:23:28.832571 135514368800576 metric_logger.py:185] completed step: 5, seconds: 0.058, TFLOP/s/device: 58.152, Tokens/s/device: 8765.322, total_weights: 16384, loss: 7.212 I0422 16:23:42.155369 135514368800576 metric_logger.py:185] completed step: 6, seconds: 0.118, TFLOP/s/device: 28.775, Tokens/s/device: 4337.219, total_weights: 16384, loss: 6.652 I0422 16:23:42.214104 135514368800576 metric_logger.py:185] completed step: 7, seconds: 13.265, TFLOP/s/device: 0.256, Tokens/s/device: 38.598, total_weights: 16384, loss: 6.269 I0422 16:23:42.272578 135514368800576 metric_logger.py:185] completed step: 8, seconds: 0.063, TFLOP/s/device: 53.925, Tokens/s/device: 8128.145, total_weights: 16384, loss: 6.031 I0422 16:23:42.330839 135514368800576 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0422 16:23:42.331488 135514368800576 checkpointing.py:798] Waited 0.0006613731384277344 seconds for step 9 to finish before starting checkpointing. I0422 16:23:42.333445 135514368800576 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 16:23:43.379297 2375 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0422 16:23:45.737589 135391889827584 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_09_pdb_lt_1/checkpoints/0/items/array_metadatas/process_5 I0422 16:24:16.551465 135391881434880 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 31.855 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 49.58682107925415 s) (per-host) I0422 16:24:16.551582 135391881434880 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 48.652229s. I0422 16:24:26.509474 135391881434880 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 58.610106s. I0422 16:24:26.509776 135388702172928 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 16:24:26.509892 135388702172928 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0422 16:24:26.509941 135388702172928 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0422 16:24:26.511311 135388702172928 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0422 16:24:26.511484 135514368800576 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0422 16:24:26.511611 135514368800576 checkpoint_manager.py:1458] Waiting for previous save to complete took 44.178173 seconds. If this number is high, consider checkpointing less frequently. I0422 16:24:26.513266 135514368800576 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0422 16:24:26.516725 135514368800576 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_09_pdb_lt_1/checkpoints/9. I0422 16:24:27.067740 135514368800576 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0422 16:24:27.067842 135514368800576 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0422 16:24:27.112777 135514368800576 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.046335s I0422 16:24:27.112936 135514368800576 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.191 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2971763610839844 s) (per-host) I0422 16:24:27.112990 135514368800576 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.297239s (batch_requests_ready=0.249017s, total_serialization_initiated=0.048153s, others=0.000070s) I0422 16:24:27.113078 135514368800576 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.301135s (all_items=0.000014s, per_item={'items': '0.00001359'}, temp_paths=0.301121) I0422 16:24:27.113913 135381152409344 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0422 16:24:27.114076 135514368800576 async_checkpointer.py:561] Finished blocking save. Time taken: 0.600741s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_09_pdb_lt_1/checkpoints/9. I0422 16:24:27.130560 135514368800576 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0422 16:24:27.130805 135388702172928 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0422 16:24:27.130936 135514368800576 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_09_pdb_lt_1/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776875022.3334148, 'wait_for_prev_duration_secs': 44.17817282676697, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776875066.5133064, 'checkpointer_blocking_duration_secs': 0.6008756160736084, 'get_old_steps_start_time': 1776875067.1141994, 'get_old_steps_duration_secs': 2.6702880859375e-05, 'checkpoint_manager_blocking_start_time': 1776875022.3317525, 'checkpoint_manager_blocking_duration_secs': 44.799150228500366} I0422 16:24:27.131042 135514368800576 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0422 16:24:27.131088 135514368800576 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 16:24:32.329016 135361502074624 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260422_154647/linen_xpk_feat_nnx_set_defaults_true_20260422_154647_09_pdb_lt_1/checkpoints/9/items/array_metadatas/process_5 I0422 16:25:08.113569 135381152409344 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 38.249 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.297770500183105 s) (per-host) I0422 16:25:08.113688 135381152409344 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 40.999663s. I0422 16:25:15.227260 135381152409344 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 48.113219s. I0422 16:25:15.227566 135388702172928 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 16:25:15.227751 135388702172928 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0422 16:25:15.227824 135388702172928 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0422 16:25:15.228979 135388702172928 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0422 16:25:15.229164 135514368800576 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0422 16:25:15.229314 135514368800576 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 16:25:15.229913 135514368800576 metric_logger.py:185] completed step: 9, seconds: 0.058, TFLOP/s/device: 58.067, Tokens/s/device: 8752.436, total_weights: 16384, loss: 5.892 Per train step: Total TFLOPs: 3.40 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Wed Apr 22 16:25:27 UTC 2026 EXIT_CODE=0