XPK Start: Thu Apr 23 16:29:50 UTC 2026 2026-04-23 16:29:54.703866: 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 00:00:1776961794.716852 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776961794.720570 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776961794.732053 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776961794.732071 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776961794.732073 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776961794.732075 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-23 16:30:13.900002: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303) I0423 16:30:14.427294 133667885578048 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-23 16:30:23,468:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0423 16:30:23.468520 133667885578048 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-23 16:30:23,470:jax._src.distributed:157: Connecting to JAX distributed service on mt-10-shardy-false-mkzab-slice-job-0-0.mt-10-shardy-false-mkzab:8482 I0423 16:30:23.470915 133667885578048 distributed.py:157] Connecting to JAX distributed service on mt-10-shardy-false-mkzab-slice-job-0-0.mt-10-shardy-false-mkzab:8482 I0423 16:30:24.445597 133667885578048 max_utils.py:284] Jax distributed system initialized! I0423 16:30:31.344038 133667885578048 max_utils.py:800] System Information: Jax Version: 0.8.1 I0423 16:30:31.344146 133667885578048 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0423 16:30:31.344186 133667885578048 max_utils.py:802] System Information: Jax Backend: PJRT C API TFRT TPU v6 lite Built on Nov 12 2025 14:16:36 (1762985796) cl/831091709 I0423 16:30:31.344220 133667885578048 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0423 16:30:31.964125 133667885578048 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0423 16:30:31.964771 133667885578048 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0423 16:30:31.964958 133667885578048 checkpointing.py:688] Setting up checkpoint logger... I0423 16:30:31.965012 133667885578048 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0423 16:30:31.965055 133667885578048 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0423 16:30:31.965400 133667885578048 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 0x79916f779640>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0423 16:30:34.844081 133667885578048 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0423 16:30:34.844429 133667885578048 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 0x797eb08a5ac0>}, handler_registry=None I0423 16:30:34.844673 133667885578048 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x797eb08a5ac0>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`. I0423 16:30:34.844724 133667885578048 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x797eb08aa510>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`. I0423 16:30:34.844761 133667885578048 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 0x797eb08a5ac0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x797eb08a5ac0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x797eb08aa510>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x797eb08aa510>}). I0423 16:30:34.845264 133667885578048 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0423 16:30:34.845345 133667885578048 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x797eb0414180> timeout: 600 secs and primary_host=0 for async checkpoint writes I0423 16:30:35.590531 133667885578048 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_10_shardy_false/checkpoints I0423 16:30:35.887882 133667885578048 checkpoint_manager.py:929] [process=5][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_10_shardy_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x799037aa00b0> I0423 16:30:35.888063 133667885578048 checkpointing.py:302] Checkpoint manager created! I0423 16:30:36.791230 133667885578048 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0423 16:30:36.791338 133667885578048 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0423 16:30:37.191214 133667885578048 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0423 16:30:37.191312 133667885578048 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0423 16:30:37.208185 133667885578048 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0423 16:30:37.208260 133667885578048 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0423 16:30:37.232039 133667885578048 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 16:30:37.232114 133667885578048 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 16:30:37.249107 133667885578048 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 16:30:37.249172 133667885578048 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 16:30:37.266057 133667885578048 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 16:30:37.266120 133667885578048 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 16:30:37.291811 133667885578048 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0423 16:30:37.291891 133667885578048 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 16:30:37.312640 133667885578048 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0423 16:30:37.312724 133667885578048 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0423 16:30:37.501635 133667885578048 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0423 16:30:37.501760 133667885578048 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0423 16:30:38.913057 133667885578048 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0423 16:30:38.913198 133667885578048 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0423 16:30:38.913255 133667885578048 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0423 16:30:38.913316 133667885578048 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: PartitionSpec('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0423 16:30:38.913370 133667885578048 maxtext_utils.py:1740] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0423 16:30:38.913409 133667885578048 maxtext_utils.py:1740] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0423 16:30:38.913463 133667885578048 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/key/kernel Shape: float32[2048,16,16,128] Logical: PartitionSpec('embed', 'layers', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None, None) I0423 16:30:38.913518 133667885578048 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/out/kernel Shape: float32[16,16,128,2048] Logical: PartitionSpec('heads', 'layers', 'kv', 'embed') Physical: (None, None, None, 'fsdp') I0423 16:30:38.913559 133667885578048 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/query/kernel Shape: float32[2048,16,16,128] Logical: PartitionSpec('embed', 'layers', 'q_heads', 'kv') Physical: ('fsdp', None, None, None) I0423 16:30:38.913596 133667885578048 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/value/kernel Shape: float32[2048,16,16,128] Logical: PartitionSpec('embed', 'layers', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None, None) I0423 16:30:38.913645 133667885578048 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0423 16:30:38.913712 133667885578048 maxtext_utils.py:1740] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0423 16:30:39.413355 133667885578048 train.py:158] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0423 16:30:39.413448 133667885578048 train.py:158] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0423 16:30:39.428367 133667885578048 train.py:165] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0423 16:30:39.428426 133667885578048 train.py:165] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0423 16:30:50.417178 133667885578048 max_utils.py:791] Total memory size: 1.5 GB, Output size: 0.4 GB, Temp size: 1.1 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0423 16:30:50.418038 133667885578048 metric_logger.py:289] number parameters: 1.104 billion I0423 16:31:01.324472 133667885578048 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0423 16:31:01.477664 133667885578048 checkpointing.py:798] Waited 0.15316414833068848 seconds for step 0 to finish before starting checkpointing. I0423 16:31:01.480062 133667885578048 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0423 16:31:01.481779 133667885578048 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0423 16:31:01.484573 133667885578048 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_10_shardy_false/checkpoints/0. I0423 16:31:02.329760 133667885578048 signaling_client.py:364] Using JaxDistributedSignalingClient I0423 16:31:02.330710 133667885578048 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0423 16:31:02.330767 133667885578048 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0423 16:31:02.570217 133667885578048 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.240558s I0423 16:31:02.570383 133667885578048 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.109 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.49613404273986816 s) (per-host) I0423 16:31:02.570433 133667885578048 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.496194s (batch_requests_ready=0.251962s, total_serialization_initiated=0.244165s, others=0.000067s) I0423 16:31:02.570524 133667885578048 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.500222s (all_items=0.000019s, per_item={'items': '0.00001907'}, temp_paths=0.500203) I0423 16:31:02.571454 133545120999168 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0423 16:31:02.571627 133667885578048 async_checkpointer.py:561] Finished blocking save. Time taken: 1.089781s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_10_shardy_false/checkpoints/0. I0423 16:31:02.615789 133667885578048 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0423 16:31:02.616121 133540322731776 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0423 16:31:02.616286 133667885578048 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_10_shardy_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776961861.4800446, 'wait_for_prev_duration_secs': 6.365776062011719e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776961861.4818158, 'checkpointer_blocking_duration_secs': 1.0899288654327393, 'get_old_steps_start_time': 1776961862.571767, 'get_old_steps_duration_secs': 2.4080276489257812e-05, 'checkpoint_manager_blocking_start_time': 1776961861.4782493, 'checkpoint_manager_blocking_duration_secs': 1.1379952430725098} I0423 16:31:02.616395 133667885578048 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0423 16:31:02.616446 133667885578048 max_utils.py:750] Memstats: After params initialized: I0423 16:31:02.616501 133667885578048 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0)) I0423 16:31:02.616537 133667885578048 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0)) I0423 16:31:02.616565 133667885578048 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0)) I0423 16:31:02.616591 133667885578048 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0)) I0423 16:31:02.881969 133667885578048 metric_logger.py:185] completed step: 0, seconds: 10.906, TFLOP/s/device: 1.246, Tokens/s/device: 187.781, total_weights: 65536, loss: 10.874 I0423 16:31:03.057761 133667885578048 metric_logger.py:185] completed step: 1, seconds: 1.556, TFLOP/s/device: 8.731, Tokens/s/device: 1315.981, total_weights: 65536, loss: 10.874 I0423 16:31:03.491436 133667885578048 metric_logger.py:185] completed step: 2, seconds: 0.029, TFLOP/s/device: 475.108, Tokens/s/device: 71613.400, total_weights: 65536, loss: 10.263 I0423 16:31:03.639000 133667885578048 metric_logger.py:185] completed step: 3, seconds: 0.434, TFLOP/s/device: 31.314, Tokens/s/device: 4720.003, total_weights: 65536, loss: 9.734 I0423 16:31:03.934535 133667885578048 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.891, Tokens/s/device: 13398.582, total_weights: 65536, loss: 9.278 I0423 16:31:03.940336 133667885578048 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.155, Tokens/s/device: 13890.679, total_weights: 65536, loss: 8.892 I0423 16:31:23.329616 133667885578048 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.890, Tokens/s/device: 6917.026, total_weights: 65536, loss: 8.592 I0423 16:31:23.476983 133667885578048 metric_logger.py:185] completed step: 7, seconds: 19.243, TFLOP/s/device: 0.706, Tokens/s/device: 106.430, total_weights: 65536, loss: 8.384 I0423 16:31:23.624510 133667885578048 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.435, Tokens/s/device: 13480.602, total_weights: 65536, loss: 8.255 I0423 16:31:23.771640 133667885578048 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0423 16:31:23.772294 133667885578048 checkpointing.py:798] Waited 0.0006761550903320312 seconds for step 9 to finish before starting checkpointing. I0423 16:31:23.774175 133667885578048 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0423 16:31:24.753836 2657 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0423 16:31:26.628053 133545129391872 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_10_shardy_false/checkpoints/0/items/array_metadatas/process_5 I0423 16:31:58.004201 133545120999168 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 28.242 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 55.92990827560425 s) (per-host) I0423 16:31:58.004326 133545120999168 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 55.432757s. I0423 16:32:05.992757 133545120999168 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 63.421172s. I0423 16:32:05.993059 133540322731776 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0423 16:32:05.993188 133540322731776 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0423 16:32:05.993253 133540322731776 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0423 16:32:05.994820 133540322731776 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0423 16:32:05.995008 133667885578048 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0423 16:32:05.995157 133667885578048 checkpoint_manager.py:1458] Waiting for previous save to complete took 42.220986 seconds. If this number is high, consider checkpointing less frequently. I0423 16:32:05.996921 133667885578048 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0423 16:32:06.000375 133667885578048 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_10_shardy_false/checkpoints/9. I0423 16:32:06.970036 133667885578048 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0423 16:32:06.970131 133667885578048 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0423 16:32:07.016309 133667885578048 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.047452s I0423 16:32:07.016471 133667885578048 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.034 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3064565658569336 s) (per-host) I0423 16:32:07.016525 133667885578048 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.306519s (batch_requests_ready=0.257162s, total_serialization_initiated=0.049288s, others=0.000070s) I0423 16:32:07.016628 133667885578048 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.310717s (all_items=0.000015s, per_item={'items': '0.00001478'}, temp_paths=0.310702) I0423 16:32:07.017560 133541355390720 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0423 16:32:07.017742 133667885578048 async_checkpointer.py:561] Finished blocking save. Time taken: 1.020750s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_10_shardy_false/checkpoints/9. I0423 16:32:07.023114 133667885578048 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0423 16:32:07.023400 133540322731776 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0423 16:32:07.023559 133667885578048 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_10_shardy_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776961883.7741473, 'wait_for_prev_duration_secs': 42.220985889434814, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776961925.9969602, 'checkpointer_blocking_duration_secs': 1.0208933353424072, 'get_old_steps_start_time': 1776961927.0178719, 'get_old_steps_duration_secs': 2.2172927856445312e-05, 'checkpoint_manager_blocking_start_time': 1776961883.7725196, 'checkpoint_manager_blocking_duration_secs': 43.25100660324097} I0423 16:32:07.023670 133667885578048 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0423 16:32:07.023718 133667885578048 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0423 16:32:12.333595 133545129391872 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260423_155242/linen_xpk_feat_nnx_set_defaults_true_20260423_155242_10_shardy_false/checkpoints/9/items/array_metadatas/process_5 I0423 16:32:48.902318 133541355390720 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.438 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.19227051734924 s) (per-host) I0423 16:32:48.902417 133541355390720 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.884733s. I0423 16:32:56.350438 133541355390720 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 49.332738s. I0423 16:32:56.350734 133540322731776 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0423 16:32:56.350848 133540322731776 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0423 16:32:56.350899 133540322731776 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0423 16:32:56.352438 133540322731776 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0423 16:32:56.352611 133667885578048 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0423 16:32:56.352762 133667885578048 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0423 16:32:56.353446 133667885578048 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.196, Tokens/s/device: 13896.711, total_weights: 65536, loss: 8.179 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Thu Apr 23 16:33:06 UTC 2026 EXIT_CODE=0