XPK Start: Sat Apr 25 16:36:18 UTC 2026 2026-04-25 16:36:22.525991: 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 00:00:1777134982.538835 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1777134982.542613 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1777134982.554298 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777134982.554335 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777134982.554338 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777134982.554340 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-25 16:36:41.785108: 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) I0425 16:36:42.312527 140218733340480 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-25 16:36:51,351:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0425 16:36:51.351721 140218733340480 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-25 16:36:51,357:jax._src.distributed:157: Connecting to JAX distributed service on mt-02-synthetic-w7sk0-slice-job-0-0.mt-02-synthetic-w7sk0:8482 I0425 16:36:51.357034 140218733340480 distributed.py:157] Connecting to JAX distributed service on mt-02-synthetic-w7sk0-slice-job-0-0.mt-02-synthetic-w7sk0:8482 I0425 16:36:52.327682 140218733340480 max_utils.py:284] Jax distributed system initialized! I0425 16:36:59.284187 140218733340480 max_utils.py:800] System Information: Jax Version: 0.8.1 I0425 16:36:59.284292 140218733340480 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0425 16:36:59.284332 140218733340480 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 I0425 16:36:59.284372 140218733340480 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0425 16:36:59.903084 140218733340480 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0425 16:37:00.018564 140218733340480 checkpointing.py:688] Setting up checkpoint logger... I0425 16:37:00.018728 140218733340480 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0425 16:37:00.018780 140218733340480 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0425 16:37:00.018985 140218733340480 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 0x7f86c7f48dd0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0425 16:37:03.149699 140218733340480 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0425 16:37:03.149940 140218733340480 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 0x7f744e1e2240>}, handler_registry=None I0425 16:37:03.150172 140218733340480 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7f744e1e2240>` 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`. I0425 16:37:03.150219 140218733340480 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f72304e4560>` 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`. I0425 16:37:03.150283 140218733340480 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 0x7f744e1e2240>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7f744e1e2240>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f72304e4560>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f72304e4560>}). I0425 16:37:03.150629 140218733340480 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0425 16:37:03.150719 140218733340480 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7f7230404f40> timeout: 600 secs and primary_host=0 for async checkpoint writes I0425 16:37:03.865706 140218733340480 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/nnx_xpk_feat_nnx_set_defaults_true_20260425_150412_02_synthetic/checkpoints I0425 16:37:04.284034 140218733340480 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/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/nnx_xpk_feat_nnx_set_defaults_true_20260425_150412_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7f723069ee70> I0425 16:37:04.284212 140218733340480 checkpointing.py:302] Checkpoint manager created! I0425 16:37:05.216596 140218733340480 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0425 16:37:05.216729 140218733340480 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0425 16:37:06.665962 140218733340480 maxtext_utils.py:1749] decoder/decoder_norm/scale/value Shape: float32[2048] Physical: (None,) I0425 16:37:06.666085 140218733340480 maxtext_utils.py:1749] decoder/layers/mlp/wi_0/kernel/value Shape: float32[2048,16,7168] Physical: ('fsdp', None, None) I0425 16:37:06.666140 140218733340480 maxtext_utils.py:1749] decoder/layers/mlp/wi_1/kernel/value Shape: float32[2048,16,7168] Physical: ('fsdp', None, None) I0425 16:37:06.666186 140218733340480 maxtext_utils.py:1749] decoder/layers/mlp/wo/kernel/value Shape: float32[7168,16,2048] Physical: (None, None, 'fsdp') I0425 16:37:06.666221 140218733340480 maxtext_utils.py:1749] decoder/layers/post_self_attention_layer_norm/scale/value Shape: float32[2048,16] Physical: (None, None) I0425 16:37:06.666257 140218733340480 maxtext_utils.py:1749] decoder/layers/pre_self_attention_layer_norm/scale/value Shape: float32[2048,16] Physical: (None, None) I0425 16:37:06.666294 140218733340480 maxtext_utils.py:1749] decoder/layers/self_attention/key/kernel/value Shape: float32[2048,16,16,128] Physical: ('fsdp', None, None, None) I0425 16:37:06.666331 140218733340480 maxtext_utils.py:1749] decoder/layers/self_attention/out/kernel/value Shape: float32[16,16,128,2048] Physical: (None, None, None, 'fsdp') I0425 16:37:06.666365 140218733340480 maxtext_utils.py:1749] decoder/layers/self_attention/query/kernel/value Shape: float32[2048,16,16,128] Physical: ('fsdp', None, None, None) I0425 16:37:06.666400 140218733340480 maxtext_utils.py:1749] decoder/layers/self_attention/value/kernel/value Shape: float32[2048,16,16,128] Physical: ('fsdp', None, None, None) I0425 16:37:06.666433 140218733340480 maxtext_utils.py:1749] decoder/logits_dense/kernel/value Shape: float32[2048,32000] Physical: ('fsdp', None) I0425 16:37:06.666466 140218733340480 maxtext_utils.py:1749] token_embedder/embedding/value Shape: float32[32000,2048] Physical: (None, 'fsdp') I0425 16:37:07.059201 140218733340480 nnx_decoders.py:616] nnx_decoders/carry Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0425 16:37:07.059340 140218733340480 nnx_decoders.py:616] nnx_decoders/carry Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0425 16:37:07.064588 140218733340480 nnx_decoders.py:616] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0425 16:37:07.064698 140218733340480 nnx_decoders.py:616] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0425 16:37:07.210459 140218733340480 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0425 16:37:07.210586 140218733340480 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0425 16:37:07.355916 140218733340480 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0425 16:37:07.356046 140218733340480 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0425 16:37:07.510691 140218733340480 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 16:37:07.510828 140218733340480 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 16:37:07.662757 140218733340480 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 16:37:07.662907 140218733340480 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 16:37:07.810426 140218733340480 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 16:37:07.810557 140218733340480 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 16:37:07.964304 140218733340480 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0425 16:37:07.964438 140218733340480 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 16:37:07.986896 140218733340480 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0425 16:37:07.986991 140218733340480 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0425 16:37:18.859357 140218733340480 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. I0425 16:37:18.861752 140218733340480 metric_logger.py:289] number parameters: 1.104 billion I0425 16:37:29.841542 140218733340480 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0425 16:37:29.994192 140218733340480 checkpointing.py:798] Waited 0.15263795852661133 seconds for step 0 to finish before starting checkpointing. I0425 16:37:29.996597 140218733340480 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0425 16:37:29.998322 140218733340480 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0425 16:37:30.001102 140218733340480 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/nnx_xpk_feat_nnx_set_defaults_true_20260425_150412_02_synthetic/checkpoints/0. I0425 16:37:30.986001 140218733340480 signaling_client.py:364] Using JaxDistributedSignalingClient I0425 16:37:30.986892 140218733340480 jax_array_handlers.py:358] Scheduling D2H of 69 prioritized jax.Array. I0425 16:37:30.986955 140218733340480 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 I0425 16:37:31.237576 140218733340480 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.252675s I0425 16:37:31.237768 140218733340480 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.912 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5296878814697266 s) (per-host) I0425 16:37:31.237823 140218733340480 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.529754s (batch_requests_ready=0.270641s, total_serialization_initiated=0.259039s, others=0.000073s) I0425 16:37:31.237921 140218733340480 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.533798s (all_items=0.000018s, per_item={'items': '0.00001836'}, temp_paths=0.533779) I0425 16:37:31.238808 140096665810688 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0425 16:37:31.238937 140218733340480 async_checkpointer.py:561] Finished blocking save. Time taken: 1.240547s. Continuing background save to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/nnx_xpk_feat_nnx_set_defaults_true_20260425_150412_02_synthetic/checkpoints/0. I0425 16:37:31.253575 140218733340480 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0425 16:37:31.253880 140092488275712 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0425 16:37:31.254033 140218733340480 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/nnx_xpk_feat_nnx_set_defaults_true_20260425_150412_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777135049.9965773, 'wait_for_prev_duration_secs': 6.604194641113281e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777135049.9983604, 'checkpointer_blocking_duration_secs': 1.2407546043395996, 'get_old_steps_start_time': 1777135051.239133, 'get_old_steps_duration_secs': 2.574920654296875e-05, 'checkpoint_manager_blocking_start_time': 1777135049.9945824, 'checkpoint_manager_blocking_duration_secs': 1.2594139575958252} I0425 16:37:31.254137 140218733340480 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0425 16:37:31.254209 140218733340480 max_utils.py:750] Memstats: After params initialized: I0425 16:37:31.254258 140218733340480 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0)) I0425 16:37:31.254293 140218733340480 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0)) I0425 16:37:31.254324 140218733340480 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0)) I0425 16:37:31.254351 140218733340480 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0)) I0425 16:37:31.523487 140218733340480 metric_logger.py:185] completed step: 0, seconds: 10.979, TFLOP/s/device: 1.238, Tokens/s/device: 186.542, total_weights: 65536, loss: 10.874 I0425 16:37:31.690313 140218733340480 metric_logger.py:185] completed step: 1, seconds: 1.680, TFLOP/s/device: 8.086, Tokens/s/device: 1218.847, total_weights: 65536, loss: 10.874 I0425 16:37:32.119424 140218733340480 metric_logger.py:185] completed step: 2, seconds: 0.021, TFLOP/s/device: 645.102, Tokens/s/device: 97236.730, total_weights: 65536, loss: 10.262 I0425 16:37:32.267394 140218733340480 metric_logger.py:185] completed step: 3, seconds: 0.420, TFLOP/s/device: 32.355, Tokens/s/device: 4876.934, total_weights: 65536, loss: 9.732 I0425 16:37:32.562930 140218733340480 metric_logger.py:185] completed step: 4, seconds: 0.163, TFLOP/s/device: 83.533, Tokens/s/device: 12590.990, total_weights: 65536, loss: 9.274 I0425 16:37:32.569712 140218733340480 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.152, Tokens/s/device: 13890.113, total_weights: 65536, loss: 8.888 I0425 16:37:51.330952 140218733340480 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.920, Tokens/s/device: 6921.561, total_weights: 65536, loss: 8.589 I0425 16:37:51.478465 140218733340480 metric_logger.py:185] completed step: 7, seconds: 18.615, TFLOP/s/device: 0.730, Tokens/s/device: 110.017, total_weights: 65536, loss: 8.380 I0425 16:37:51.625873 140218733340480 metric_logger.py:185] completed step: 8, seconds: 0.153, TFLOP/s/device: 88.867, Tokens/s/device: 13395.076, total_weights: 65536, loss: 8.251 I0425 16:37:51.774371 140218733340480 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0425 16:37:51.774950 140218733340480 checkpointing.py:798] Waited 0.0005795955657958984 seconds for step 9 to finish before starting checkpointing. I0425 16:37:51.777069 140218733340480 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 16:37:52.421716 2586 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0425 16:37:54.833514 140096749672192 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 69 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/nnx_xpk_feat_nnx_set_defaults_true_20260425_150412_02_synthetic/checkpoints/0/items/array_metadatas/process_5 I0425 16:38:25.739911 140096665810688 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 28.703 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 55.0317759513855 s) (per-host) I0425 16:38:25.740040 140096665810688 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 54.501020s. I0425 16:38:33.606255 140096665810688 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 62.367220s. I0425 16:38:33.606553 140092488275712 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 16:38:33.606700 140092488275712 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0425 16:38:33.606749 140092488275712 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0425 16:38:33.609132 140092488275712 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0425 16:38:33.609309 140218733340480 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. W0425 16:38:33.609447 140218733340480 checkpoint_manager.py:1458] Waiting for previous save to complete took 41.832382 seconds. If this number is high, consider checkpointing less frequently. I0425 16:38:33.611669 140218733340480 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0425 16:38:33.615121 140218733340480 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/nnx_xpk_feat_nnx_set_defaults_true_20260425_150412_02_synthetic/checkpoints/9. I0425 16:38:34.634226 140218733340480 jax_array_handlers.py:358] Scheduling D2H of 69 prioritized jax.Array. I0425 16:38:34.634384 140218733340480 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 I0425 16:38:34.679126 140218733340480 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.046878s I0425 16:38:34.679301 140218733340480 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 4.642 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3322765827178955 s) (per-host) I0425 16:38:34.679356 140218733340480 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.332340s (batch_requests_ready=0.280701s, total_serialization_initiated=0.051566s, others=0.000073s) I0425 16:38:34.679452 140218733340480 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.336502s (all_items=0.000016s, per_item={'items': '0.00001597'}, temp_paths=0.336486) I0425 16:38:34.680359 140093545232128 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0425 16:38:34.680517 140218733340480 async_checkpointer.py:561] Finished blocking save. Time taken: 1.068779s. Continuing background save to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/nnx_xpk_feat_nnx_set_defaults_true_20260425_150412_02_synthetic/checkpoints/9. I0425 16:38:34.689081 140218733340480 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0425 16:38:34.689331 140092488275712 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0425 16:38:34.689488 140218733340480 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/nnx_xpk_feat_nnx_set_defaults_true_20260425_150412_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777135071.7770407, 'wait_for_prev_duration_secs': 41.83238244056702, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777135113.611706, 'checkpointer_blocking_duration_secs': 1.0689172744750977, 'get_old_steps_start_time': 1777135114.680643, 'get_old_steps_duration_secs': 3.886222839355469e-05, 'checkpoint_manager_blocking_start_time': 1777135071.7751665, 'checkpoint_manager_blocking_duration_secs': 42.91428780555725} I0425 16:38:34.689595 140218733340480 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0425 16:38:34.689639 140218733340480 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 16:38:39.407779 140096749672192 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 69 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/nnx_xpk_feat_nnx_set_defaults_true_20260425_150412_02_synthetic/checkpoints/9/items/array_metadatas/process_5 I0425 16:39:16.251013 140093545232128 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.696 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.90394639968872 s) (per-host) I0425 16:39:16.251131 140093545232128 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.570662s. I0425 16:39:24.143610 140093545232128 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 49.463127s. I0425 16:39:24.143900 140092488275712 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 16:39:24.144033 140092488275712 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0425 16:39:24.144082 140092488275712 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0425 16:39:24.146529 140092488275712 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0425 16:39:24.146728 140218733340480 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. I0425 16:39:24.146884 140218733340480 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0425 16:39:24.147573 140218733340480 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.355, Tokens/s/device: 13920.798, total_weights: 65536, loss: 8.175 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Sat Apr 25 16:39:34 UTC 2026 EXIT_CODE=0