XPK Start: Mon Apr 20 19:52:34 UTC 2026 2026-04-20 19:52:38.700366: 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 00:00:1776714758.713182 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776714758.716908 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776714758.728444 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776714758.728466 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776714758.728469 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776714758.728471 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-20 19:52:57.705079: 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) I0420 19:52:58.232890 134328841770816 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-20 19:53:07,273:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0420 19:53:07.273229 134328841770816 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-20 19:53:07,275:jax._src.distributed:157: Connecting to JAX distributed service on mt-10-shardy-false-045w1-slice-job-0-0.mt-10-shardy-false-045w1:8482 I0420 19:53:07.275556 134328841770816 distributed.py:157] Connecting to JAX distributed service on mt-10-shardy-false-045w1-slice-job-0-0.mt-10-shardy-false-045w1:8482 I0420 19:53:08.958704 134328841770816 max_utils.py:284] Jax distributed system initialized! I0420 19:53:15.913797 134328841770816 max_utils.py:800] System Information: Jax Version: 0.8.1 I0420 19:53:15.913903 134328841770816 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0420 19:53:15.913945 134328841770816 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 I0420 19:53:15.913982 134328841770816 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0420 19:53:16.535851 134328841770816 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0420 19:53:16.536450 134328841770816 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0420 19:53:16.536643 134328841770816 checkpointing.py:688] Setting up checkpoint logger... I0420 19:53:16.536692 134328841770816 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0420 19:53:16.536735 134328841770816 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0420 19:53:16.537079 134328841770816 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 0x7a2b5f9792b0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0420 19:53:19.370213 134328841770816 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0420 19:53:19.370588 134328841770816 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 0x7a18f0b79880>}, handler_registry=None I0420 19:53:19.370819 134328841770816 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a18f0b79880>` 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`. I0420 19:53:19.370867 134328841770816 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a18f0b7ed50>` 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`. I0420 19:53:19.370900 134328841770816 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 0x7a18f0b79880>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a18f0b79880>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a18f0b7ed50>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a18f0b7ed50>}). I0420 19:53:19.371393 134328841770816 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0420 19:53:19.371480 134328841770816 async_checkpointer.py:177] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7a173054c180> timeout: 600 secs and primary_host=0 for async checkpoint writes I0420 19:53:20.488239 134328841770816 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_10_shardy_false/checkpoints I0420 19:53:20.508780 134328841770816 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_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_10_shardy_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7a17303077a0> I0420 19:53:20.508914 134328841770816 checkpointing.py:302] Checkpoint manager created! I0420 19:53:21.395251 134328841770816 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0420 19:53:21.395362 134328841770816 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 19:53:21.766295 134328841770816 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0420 19:53:21.766396 134328841770816 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 19:53:21.782066 134328841770816 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0420 19:53:21.782128 134328841770816 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 19:53:21.804636 134328841770816 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 19:53:21.804705 134328841770816 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 19:53:21.820313 134328841770816 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 19:53:21.820383 134328841770816 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 19:53:21.835972 134328841770816 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 19:53:21.836033 134328841770816 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 19:53:21.860679 134328841770816 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0420 19:53:21.860752 134328841770816 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 19:53:21.880072 134328841770816 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0420 19:53:21.880140 134328841770816 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0420 19:53:22.068588 134328841770816 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0420 19:53:22.068698 134328841770816 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0420 19:53:23.479151 134328841770816 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0420 19:53:23.479285 134328841770816 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) I0420 19:53:23.479347 134328841770816 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) I0420 19:53:23.479410 134328841770816 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') I0420 19:53:23.479483 134328841770816 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) I0420 19:53:23.479525 134328841770816 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) I0420 19:53:23.479579 134328841770816 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) I0420 19:53:23.479634 134328841770816 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') I0420 19:53:23.479677 134328841770816 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) I0420 19:53:23.479715 134328841770816 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) I0420 19:53:23.479765 134328841770816 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0420 19:53:23.479815 134328841770816 maxtext_utils.py:1740] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0420 19:53:23.978791 134328841770816 train.py:158] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0420 19:53:23.978893 134328841770816 train.py:158] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0420 19:53:23.994030 134328841770816 train.py:165] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0420 19:53:23.994095 134328841770816 train.py:165] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0420 19:53:34.720006 134328841770816 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. I0420 19:53:34.720872 134328841770816 metric_logger.py:289] number parameters: 1.104 billion I0420 19:53:45.757004 134328841770816 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0420 19:53:45.909826 134328841770816 checkpointing.py:798] Waited 0.15280675888061523 seconds for step 0 to finish before starting checkpointing. I0420 19:53:45.912333 134328841770816 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0420 19:53:45.914047 134328841770816 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 0 I0420 19:53:45.916841 134328841770816 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_10_shardy_false/checkpoints/0. I0420 19:53:46.457056 134328841770816 signaling_client.py:364] Using JaxDistributedSignalingClient I0420 19:53:46.458109 134328841770816 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0420 19:53:46.458169 134328841770816 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 I0420 19:53:46.704887 134328841770816 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.247915s I0420 19:53:46.705056 134328841770816 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.031 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5089032649993896 s) (per-host) I0420 19:53:46.705107 134328841770816 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.508964s (batch_requests_ready=0.257419s, total_serialization_initiated=0.251476s, others=0.000068s) I0420 19:53:46.705196 134328841770816 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.512982s (all_items=0.000018s, per_item={'items': '0.00001836'}, temp_paths=0.512964) I0420 19:53:46.706100 134203652769536 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0420 19:53:46.706274 134328841770816 async_checkpointer.py:561] Finished blocking save. Time taken: 0.792160s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_10_shardy_false/checkpoints/0. I0420 19:53:46.710450 134328841770816 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0420 19:53:46.710771 134204197152512 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0420 19:53:46.710901 134328841770816 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_10_shardy_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776714825.912316, 'wait_for_prev_duration_secs': 6.532669067382812e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776714825.9140856, 'checkpointer_blocking_duration_secs': 0.7922954559326172, 'get_old_steps_start_time': 1776714826.7064016, 'get_old_steps_duration_secs': 4.267692565917969e-05, 'checkpoint_manager_blocking_start_time': 1776714825.910379, 'checkpoint_manager_blocking_duration_secs': 0.8004856109619141} I0420 19:53:46.711056 134328841770816 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0420 19:53:46.711105 134328841770816 max_utils.py:750] Memstats: After params initialized: I0420 19:53:46.711154 134328841770816 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0)) I0420 19:53:46.711187 134328841770816 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0)) I0420 19:53:46.711213 134328841770816 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0)) I0420 19:53:46.711237 134328841770816 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0)) I0420 19:53:46.975494 134328841770816 metric_logger.py:185] completed step: 0, seconds: 11.036, TFLOP/s/device: 1.231, Tokens/s/device: 185.574, total_weights: 65536, loss: 10.874 I0420 19:53:47.144015 134328841770816 metric_logger.py:185] completed step: 1, seconds: 1.217, TFLOP/s/device: 11.163, Tokens/s/device: 1682.557, total_weights: 65536, loss: 10.874 I0420 19:53:47.564130 134328841770816 metric_logger.py:185] completed step: 2, seconds: 0.021, TFLOP/s/device: 641.356, Tokens/s/device: 96672.174, total_weights: 65536, loss: 10.263 I0420 19:53:47.711531 134328841770816 metric_logger.py:185] completed step: 3, seconds: 0.420, TFLOP/s/device: 32.326, Tokens/s/device: 4872.455, total_weights: 65536, loss: 9.734 I0420 19:53:48.007012 134328841770816 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.922, Tokens/s/device: 13403.229, total_weights: 65536, loss: 9.278 I0420 19:53:48.012338 134328841770816 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.257, Tokens/s/device: 13905.958, total_weights: 65536, loss: 8.892 I0420 19:54:07.523812 134328841770816 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.932, Tokens/s/device: 6923.410, total_weights: 65536, loss: 8.592 I0420 19:54:07.671250 134328841770816 metric_logger.py:185] completed step: 7, seconds: 19.365, TFLOP/s/device: 0.702, Tokens/s/device: 105.760, total_weights: 65536, loss: 8.384 I0420 19:54:07.818756 134328841770816 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.391, Tokens/s/device: 13474.039, total_weights: 65536, loss: 8.255 I0420 19:54:07.965878 134328841770816 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0420 19:54:07.966552 134328841770816 checkpointing.py:798] Waited 0.0006959438323974609 seconds for step 9 to finish before starting checkpointing. I0420 19:54:07.968537 134328841770816 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0420 19:54:09.780921 2643 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0420 19:54:11.729045 134203702245120 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_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_10_shardy_false/checkpoints/0/items/array_metadatas/process_6 I0420 19:54:43.449868 134203652769536 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 27.589 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 57.2536678314209 s) (per-host) I0420 19:54:43.449985 134203652769536 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 56.743764s. I0420 19:54:51.706156 134203652769536 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 64.999919s. I0420 19:54:51.706495 134204197152512 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0420 19:54:51.706627 134204197152512 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0420 19:54:51.706705 134204197152512 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0420 19:54:51.708137 134204197152512 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0420 19:54:51.708350 134328841770816 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. W0420 19:54:51.708492 134328841770816 checkpoint_manager.py:1458] Waiting for previous save to complete took 43.739961 seconds. If this number is high, consider checkpointing less frequently. I0420 19:54:51.710368 134328841770816 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 9 I0420 19:54:51.713917 134328841770816 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_10_shardy_false/checkpoints/9. I0420 19:54:52.684687 134328841770816 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0420 19:54:52.684785 134328841770816 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 I0420 19:54:52.727950 134328841770816 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.044431s I0420 19:54:52.728121 134328841770816 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.159 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.29901695251464844 s) (per-host) I0420 19:54:52.728172 134328841770816 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.299078s (batch_requests_ready=0.252738s, total_serialization_initiated=0.046271s, others=0.000069s) I0420 19:54:52.728259 134328841770816 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.303386s (all_items=0.000015s, per_item={'items': '0.00001478'}, temp_paths=0.303371) I0420 19:54:52.729181 134203702245120 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0420 19:54:52.729345 134328841770816 async_checkpointer.py:561] Finished blocking save. Time taken: 1.018890s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_10_shardy_false/checkpoints/9. I0420 19:54:52.743811 134328841770816 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0420 19:54:52.744103 134204197152512 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0420 19:54:52.744263 134328841770816 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_10_shardy_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776714847.9685085, 'wait_for_prev_duration_secs': 43.73996138572693, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776714891.7104146, 'checkpointer_blocking_duration_secs': 1.019059658050537, 'get_old_steps_start_time': 1776714892.7294984, 'get_old_steps_duration_secs': 2.8133392333984375e-05, 'checkpoint_manager_blocking_start_time': 1776714847.9667804, 'checkpoint_manager_blocking_duration_secs': 44.777448415756226} I0420 19:54:52.744371 134328841770816 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0420 19:54:52.744417 134328841770816 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0420 19:54:57.956930 134204188759808 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_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_10_shardy_false/checkpoints/9/items/array_metadatas/process_6 I0420 19:55:34.603127 134203702245120 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 37.454 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.17398500442505 s) (per-host) I0420 19:55:34.603248 134203702245120 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.873956s. I0420 19:55:42.732915 134203702245120 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 50.003608s. I0420 19:55:42.733203 134204197152512 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0420 19:55:42.733320 134204197152512 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0420 19:55:42.733371 134204197152512 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0420 19:55:42.734737 134204197152512 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0420 19:55:42.734900 134328841770816 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. I0420 19:55:42.735042 134328841770816 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0420 19:55:42.735778 134328841770816 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.321, Tokens/s/device: 13915.690, 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: Mon Apr 20 19:55:55 UTC 2026 EXIT_CODE=0