XPK Start: Sun Apr 19 04:02:34 UTC 2026 2026-04-19 04:02:39.024274: 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 00:00:1776571359.037273 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776571359.041627 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776571359.052861 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776571359.052877 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776571359.052880 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776571359.052882 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-19 04:02:58.280929: 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 04:02:58.802663 137485416163136 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-19 04:03:07,843:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0419 04:03:07.843065 137485416163136 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-19 04:03:07,845:jax._src.distributed:157: Connecting to JAX distributed service on mt-13-scan-layers-true-mczik-slice-job-0-0.mt-13-scan-layers-true-mczik:8482 I0419 04:03:07.845495 137485416163136 distributed.py:157] Connecting to JAX distributed service on mt-13-scan-layers-true-mczik-slice-job-0-0.mt-13-scan-layers-true-mczik:8482 I0419 04:03:09.959094 137485416163136 max_utils.py:284] Jax distributed system initialized! I0419 04:03:16.979556 137485416163136 max_utils.py:800] System Information: Jax Version: 0.8.1 I0419 04:03:16.979664 137485416163136 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0419 04:03:16.979705 137485416163136 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 I0419 04:03:16.979739 137485416163136 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0419 04:03:17.599223 137485416163136 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 04:03:17.599840 137485416163136 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 04:03:17.600022 137485416163136 checkpointing.py:688] Setting up checkpoint logger... I0419 04:03:17.600072 137485416163136 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0419 04:03:17.600116 137485416163136 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0419 04:03:17.600490 137485416163136 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 0x7d0a54169550>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0419 04:03:20.924877 137485416163136 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0419 04:03:20.925222 137485416163136 checkpoint_manager.py:709] [process=6][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7cf5c43695e0>}, handler_registry=None I0419 04:03:20.925470 137485416163136 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7cf5c43695e0>` 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 04:03:20.925526 137485416163136 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7cf5c436a360>` 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 04:03:20.925563 137485416163136 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 0x7cf5c43695e0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7cf5c43695e0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7cf5c436a360>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7cf5c436a360>}). I0419 04:03:20.926080 137485416163136 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0419 04:03:20.926158 137485416163136 async_checkpointer.py:177] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7cf5c411c180> timeout: 600 secs and primary_host=0 for async checkpoint writes I0419 04:03:23.773426 137485416163136 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_13_scan_layers_true/checkpoints I0419 04:03:23.775673 137485416163136 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=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_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_13_scan_layers_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7cf5c42d6120> I0419 04:03:23.775788 137485416163136 checkpointing.py:302] Checkpoint manager created! I0419 04:03:24.675890 137485416163136 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0419 04:03:24.675998 137485416163136 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 04:03:25.044307 137485416163136 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 04:03:25.044407 137485416163136 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 04:03:25.060261 137485416163136 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 04:03:25.060326 137485416163136 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 04:03:25.083204 137485416163136 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 04:03:25.083292 137485416163136 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:03:25.099311 137485416163136 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 04:03:25.099377 137485416163136 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:03:25.115111 137485416163136 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 04:03:25.115177 137485416163136 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:03:25.139881 137485416163136 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0419 04:03:25.139961 137485416163136 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 04:03:25.159499 137485416163136 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0419 04:03:25.159571 137485416163136 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0419 04:03:25.348808 137485416163136 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0419 04:03:25.348926 137485416163136 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0419 04:03:26.770448 137485416163136 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0419 04:03:26.770581 137485416163136 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) I0419 04:03:26.770637 137485416163136 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) I0419 04:03:26.770696 137485416163136 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') I0419 04:03:26.770749 137485416163136 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) I0419 04:03:26.770788 137485416163136 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) I0419 04:03:26.770841 137485416163136 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) I0419 04:03:26.770894 137485416163136 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') I0419 04:03:26.770936 137485416163136 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) I0419 04:03:26.770974 137485416163136 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) I0419 04:03:26.771023 137485416163136 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0419 04:03:26.771072 137485416163136 maxtext_utils.py:1740] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0419 04:03:27.266112 137485416163136 train.py:158] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 04:03:27.266207 137485416163136 train.py:158] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0419 04:03:27.281267 137485416163136 train.py:165] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 04:03:27.281327 137485416163136 train.py:165] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0419 04:03:38.080700 137485416163136 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 04:03:38.081641 137485416163136 metric_logger.py:289] number parameters: 1.104 billion I0419 04:03:49.394928 137485416163136 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0419 04:03:49.548568 137485416163136 checkpointing.py:798] Waited 0.15362048149108887 seconds for step 0 to finish before starting checkpointing. I0419 04:03:49.551229 137485416163136 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 04:03:49.552920 137485416163136 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 0 I0419 04:03:49.555772 137485416163136 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_13_scan_layers_true/checkpoints/0. I0419 04:03:50.954307 137485416163136 signaling_client.py:364] Using JaxDistributedSignalingClient I0419 04:03:50.955392 137485416163136 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 04:03:50.955449 137485416163136 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 04:03:51.198409 137485416163136 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.244194s I0419 04:03:51.198578 137485416163136 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.649 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.9354465007781982 s) (per-host) I0419 04:03:51.198628 137485416163136 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.935506s (batch_requests_ready=0.687688s, total_serialization_initiated=0.247751s, others=0.000067s) I0419 04:03:51.198715 137485416163136 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.939634s (all_items=0.000019s, per_item={'items': '0.00001884'}, temp_paths=0.939615) I0419 04:03:51.199641 137361761490688 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0419 04:03:51.199807 137485416163136 async_checkpointer.py:561] Finished blocking save. Time taken: 1.646819s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_13_scan_layers_true/checkpoints/0. I0419 04:03:51.219466 137485416163136 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0419 04:03:51.219789 137356401174272 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0419 04:03:51.219947 137485416163136 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_13_scan_layers_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776571429.5512033, 'wait_for_prev_duration_secs': 8.416175842285156e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776571429.5529563, 'checkpointer_blocking_duration_secs': 1.6469595432281494, 'get_old_steps_start_time': 1776571431.1999369, 'get_old_steps_duration_secs': 2.4318695068359375e-05, 'checkpoint_manager_blocking_start_time': 1776571429.5490415, 'checkpoint_manager_blocking_duration_secs': 1.670867681503296} I0419 04:03:51.220050 137485416163136 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0419 04:03:51.220098 137485416163136 max_utils.py:750] Memstats: After params initialized: I0419 04:03:51.220146 137485416163136 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0)) I0419 04:03:51.220180 137485416163136 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0)) I0419 04:03:51.220206 137485416163136 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0)) I0419 04:03:51.220230 137485416163136 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0)) I0419 04:03:51.483604 137485416163136 metric_logger.py:185] completed step: 0, seconds: 11.313, TFLOP/s/device: 1.201, Tokens/s/device: 181.028, total_weights: 65536, loss: 10.874 I0419 04:03:51.647835 137485416163136 metric_logger.py:185] completed step: 1, seconds: 2.087, TFLOP/s/device: 6.509, Tokens/s/device: 981.125, total_weights: 65536, loss: 10.874 I0419 04:03:52.040131 137485416163136 metric_logger.py:185] completed step: 2, seconds: 0.017, TFLOP/s/device: 803.925, Tokens/s/device: 121176.262, total_weights: 65536, loss: 10.263 I0419 04:03:52.187815 137485416163136 metric_logger.py:185] completed step: 3, seconds: 0.393, TFLOP/s/device: 34.599, Tokens/s/device: 5215.164, total_weights: 65536, loss: 9.734 I0419 04:03:52.482875 137485416163136 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.980, Tokens/s/device: 13412.094, total_weights: 65536, loss: 9.278 I0419 04:03:52.488187 137485416163136 metric_logger.py:185] completed step: 5, seconds: 0.148, TFLOP/s/device: 91.877, Tokens/s/device: 13848.692, total_weights: 65536, loss: 8.892 I0419 04:04:13.982093 137485416163136 metric_logger.py:185] completed step: 6, seconds: 0.295, TFLOP/s/device: 46.072, Tokens/s/device: 6944.444, total_weights: 65536, loss: 8.592 I0419 04:04:14.129835 137485416163136 metric_logger.py:185] completed step: 7, seconds: 21.347, TFLOP/s/device: 0.636, Tokens/s/device: 95.938, total_weights: 65536, loss: 8.384 I0419 04:04:14.277361 137485416163136 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.421, Tokens/s/device: 13478.561, total_weights: 65536, loss: 8.255 I0419 04:04:14.424396 137485416163136 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0419 04:04:14.424937 137485416163136 checkpointing.py:798] Waited 0.000560760498046875 seconds for step 9 to finish before starting checkpointing. I0419 04:04:14.427207 137485416163136 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 04:04:15.538822 2713 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0419 04:04:18.475647 137361769883392 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_13_scan_layers_true/checkpoints/0/items/array_metadatas/process_6 I0419 04:04:49.770287 137361761490688 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 26.545 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 59.507097244262695 s) (per-host) I0419 04:04:49.770411 137361761490688 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 58.570658s. I0419 04:04:59.662754 137361761490688 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 68.462985s. I0419 04:04:59.663052 137356401174272 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 04:04:59.663192 137356401174272 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0419 04:04:59.663259 137356401174272 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0419 04:04:59.664906 137356401174272 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0419 04:04:59.665052 137485416163136 checkpoint_manager.py:2036] [process=6][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0419 04:04:59.665144 137485416163136 checkpoint_manager.py:1458] Waiting for previous save to complete took 45.237942 seconds. If this number is high, consider checkpointing less frequently. I0419 04:04:59.667189 137485416163136 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 9 I0419 04:04:59.670746 137485416163136 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_13_scan_layers_true/checkpoints/9. I0419 04:05:01.051908 137485416163136 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 04:05:01.052010 137485416163136 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 04:05:01.096160 137485416163136 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.045487s I0419 04:05:01.096358 137485416163136 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.127 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.7252898216247559 s) (per-host) I0419 04:05:01.096413 137485416163136 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.725356s (batch_requests_ready=0.677871s, total_serialization_initiated=0.047409s, others=0.000076s) I0419 04:05:01.096508 137485416163136 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.729468s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.729453) I0419 04:05:01.097451 137357433829120 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0419 04:05:01.097628 137485416163136 async_checkpointer.py:561] Finished blocking save. Time taken: 1.430349s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_13_scan_layers_true/checkpoints/9. I0419 04:05:01.560169 137485416163136 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0419 04:05:01.560583 137356401174272 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0419 04:05:01.560765 137485416163136 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_13_scan_layers_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776571454.427176, 'wait_for_prev_duration_secs': 45.23794198036194, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776571499.667229, 'checkpointer_blocking_duration_secs': 1.4305224418640137, 'get_old_steps_start_time': 1776571501.0977705, 'get_old_steps_duration_secs': 3.170967102050781e-05, 'checkpoint_manager_blocking_start_time': 1776571454.4251595, 'checkpoint_manager_blocking_duration_secs': 47.13556146621704} I0419 04:05:01.560890 137485416163136 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0419 04:05:01.560947 137485416163136 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 04:05:06.724682 137361769883392 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260419_025828/linen_xpk_feat_nnx_set_defaults_true_20260419_025828_13_scan_layers_true/checkpoints/9/items/array_metadatas/process_6 I0419 04:05:42.357548 137357433829120 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 37.622 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.98644137382507 s) (per-host) I0419 04:05:42.357668 137357433829120 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.260101s. I0419 04:05:51.256787 137357433829120 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 50.159205s. I0419 04:05:51.257025 137356401174272 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 04:05:51.257093 137356401174272 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0419 04:05:51.257148 137356401174272 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0419 04:05:51.258528 137356401174272 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0419 04:05:51.258695 137485416163136 checkpoint_manager.py:2036] [process=6][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0419 04:05:51.258843 137485416163136 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 04:05:51.259553 137485416163136 metric_logger.py:185] completed step: 9, seconds: 0.148, TFLOP/s/device: 92.062, Tokens/s/device: 13876.561, 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: Sun Apr 19 04:06:04 UTC 2026 EXIT_CODE=0