XPK Start: Sat Apr 25 18:20:15 UTC 2026 2026-04-25 18:20:20.040778: 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 00:00:1777141220.053380 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1777141220.057085 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1777141220.068549 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777141220.068565 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777141220.068568 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777141220.068570 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-25 18:20:39.138478: 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 18:20:39.655810 140092765345600 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-25 18:20:48,696:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0425 18:20:48.696625 140092765345600 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-25 18:20:48,698:jax._src.distributed:157: Connecting to JAX distributed service on mt-07-eval-469c8-slice-job-0-0.mt-07-eval-469c8:8482 I0425 18:20:48.698994 140092765345600 distributed.py:157] Connecting to JAX distributed service on mt-07-eval-469c8-slice-job-0-0.mt-07-eval-469c8:8482 I0425 18:20:52.119508 140092765345600 max_utils.py:284] Jax distributed system initialized! I0425 18:20:57.809797 140092765345600 max_utils.py:800] System Information: Jax Version: 0.8.1 I0425 18:20:57.809906 140092765345600 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0425 18:20:57.809945 140092765345600 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 18:20:57.809980 140092765345600 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0425 18:20:58.424158 140092765345600 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0425 18:20:58.424453 140092765345600 checkpointing.py:677] Setting up checkpoint logger... I0425 18:20:58.424508 140092765345600 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0425 18:20:58.424553 140092765345600 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0425 18:20:58.424937 140092765345600 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 0x7f6958379100>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0425 18:21:01.366552 140092765345600 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0425 18:21:01.366914 140092765345600 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 0x7f5721a380b0>}, handler_registry=None I0425 18:21:01.367152 140092765345600 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7f5721a380b0>` 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 18:21:01.367200 140092765345600 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f5721a3bda0>` 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 18:21:01.367237 140092765345600 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 0x7f5721a380b0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7f5721a380b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f5721a3bda0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f5721a3bda0>}). I0425 18:21:01.367752 140092765345600 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0425 18:21:01.367830 140092765345600 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7f5720ae6840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0425 18:21:02.445908 140092765345600 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_07_eval/checkpoints I0425 18:21:02.535160 140092765345600 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_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7f5721a3ba40> I0425 18:21:02.535330 140092765345600 checkpointing.py:301] Checkpoint manager created! I0425 18:21:03.431315 140092765345600 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0425 18:21:03.431415 140092765345600 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0425 18:21:03.799783 140092765345600 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0425 18:21:03.799878 140092765345600 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0425 18:21:03.815670 140092765345600 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0425 18:21:03.815726 140092765345600 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0425 18:21:03.839455 140092765345600 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 18:21:03.839524 140092765345600 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 18:21:03.855297 140092765345600 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 18:21:03.855361 140092765345600 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 18:21:03.870978 140092765345600 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 18:21:03.871051 140092765345600 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 18:21:03.894875 140092765345600 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0425 18:21:03.894950 140092765345600 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 18:21:03.914479 140092765345600 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0425 18:21:03.914544 140092765345600 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0425 18:21:04.103665 140092765345600 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0425 18:21:04.103774 140092765345600 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0425 18:21:05.523357 140092765345600 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0425 18:21:05.523484 140092765345600 maxtext_utils.py:1620] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0425 18:21:05.523539 140092765345600 maxtext_utils.py:1620] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0425 18:21:05.523596 140092765345600 maxtext_utils.py:1620] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: PartitionSpec('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0425 18:21:05.523661 140092765345600 maxtext_utils.py:1620] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0425 18:21:05.523705 140092765345600 maxtext_utils.py:1620] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0425 18:21:05.523762 140092765345600 maxtext_utils.py:1620] 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) I0425 18:21:05.523816 140092765345600 maxtext_utils.py:1620] 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') I0425 18:21:05.523857 140092765345600 maxtext_utils.py:1620] 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) I0425 18:21:05.523896 140092765345600 maxtext_utils.py:1620] 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) I0425 18:21:05.523945 140092765345600 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0425 18:21:05.523993 140092765345600 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0425 18:21:06.025890 140092765345600 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0425 18:21:06.025988 140092765345600 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0425 18:21:06.040859 140092765345600 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0425 18:21:06.040924 140092765345600 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0425 18:21:16.895729 140092765345600 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 18:21:16.896584 140092765345600 metric_logger.py:289] number parameters: 1.104 billion I0425 18:21:27.976142 140092765345600 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0425 18:21:28.132419 140092765345600 checkpointing.py:776] Waited 0.156266450881958 seconds for step 0 to finish before starting checkpointing. I0425 18:21:28.134812 140092765345600 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0425 18:21:28.136492 140092765345600 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0425 18:21:28.139991 140092765345600 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_07_eval/checkpoints/0. I0425 18:21:29.342977 140092765345600 signaling_client.py:364] Using JaxDistributedSignalingClient I0425 18:21:29.344042 140092765345600 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0425 18:21:29.344097 140092765345600 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 18:21:29.591852 140092765345600 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.248953s I0425 18:21:29.592031 140092765345600 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.044 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.75467848777771 s) (per-host) I0425 18:21:29.592086 140092765345600 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.754743s (batch_requests_ready=0.502159s, total_serialization_initiated=0.252510s, others=0.000074s) I0425 18:21:29.592201 140092765345600 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.758877s (all_items=0.000020s, per_item={'items': '0.00002027'}, temp_paths=0.758857) I0425 18:21:29.593215 139972405339904 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0425 18:21:29.593337 140092765345600 async_checkpointer.py:561] Finished blocking save. Time taken: 1.456730s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_07_eval/checkpoints/0. I0425 18:21:29.766627 140092765345600 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0425 18:21:29.767057 139967598679808 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0425 18:21:29.767204 140092765345600 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777141288.134793, 'wait_for_prev_duration_secs': 6.318092346191406e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777141288.1365778, 'checkpointer_blocking_duration_secs': 1.4569406509399414, 'get_old_steps_start_time': 1777141289.5935378, 'get_old_steps_duration_secs': 2.6464462280273438e-05, 'checkpoint_manager_blocking_start_time': 1777141288.1330025, 'checkpoint_manager_blocking_duration_secs': 1.6341562271118164} I0425 18:21:29.767368 140092765345600 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0425 18:21:29.767421 140092765345600 max_utils.py:750] Memstats: After params initialized: I0425 18:21:29.767477 140092765345600 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0)) I0425 18:21:29.767510 140092765345600 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0)) I0425 18:21:29.767538 140092765345600 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0)) I0425 18:21:29.767563 140092765345600 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0)) I0425 18:21:30.034901 140092765345600 metric_logger.py:185] completed step: 0, seconds: 11.079, TFLOP/s/device: 1.226, Tokens/s/device: 184.847, total_weights: 65536, loss: 10.874 I0425 18:21:30.206319 140092765345600 metric_logger.py:185] completed step: 1, seconds: 2.057, TFLOP/s/device: 6.604, Tokens/s/device: 995.404, total_weights: 65536, loss: 10.874 I0425 18:21:30.621676 140092765345600 metric_logger.py:185] completed step: 2, seconds: 0.024, TFLOP/s/device: 559.878, Tokens/s/device: 84390.968, total_weights: 65536, loss: 10.263 I0425 18:21:30.769198 140092765345600 metric_logger.py:185] completed step: 3, seconds: 0.416, TFLOP/s/device: 32.683, Tokens/s/device: 4926.274, total_weights: 65536, loss: 9.734 I0425 18:21:31.064691 140092765345600 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 89.008, Tokens/s/device: 13416.312, total_weights: 65536, loss: 9.278 I0425 18:21:31.070432 140092765345600 metric_logger.py:185] completed step: 5, seconds: 0.148, TFLOP/s/device: 91.915, Tokens/s/device: 13854.501, total_weights: 65536, loss: 8.892 I0425 18:21:50.010603 140092765345600 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.967, Tokens/s/device: 6928.656, total_weights: 65536, loss: 8.592 I0425 18:21:50.158032 140092765345600 metric_logger.py:185] completed step: 7, seconds: 18.793, TFLOP/s/device: 0.723, Tokens/s/device: 108.974, total_weights: 65536, loss: 8.384 I0425 18:21:50.305666 140092765345600 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.375, Tokens/s/device: 13471.646, total_weights: 65536, loss: 8.255 I0425 18:21:50.452814 140092765345600 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0425 18:21:50.453463 140092765345600 checkpointing.py:776] Waited 0.0006694793701171875 seconds for step 9 to finish before starting checkpointing. I0425 18:21:50.455358 140092765345600 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 18:21:51.922450 2584 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0425 18:21:54.132436 139972413732608 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_07_eval/checkpoints/0/items/array_metadatas/process_5 I0425 18:22:25.017595 139972405339904 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 28.117 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 56.1802020072937 s) (per-host) I0425 18:22:25.017727 139972405339904 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 55.424311s. I0425 18:22:34.020406 139972405339904 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 64.426975s. I0425 18:22:34.020711 139967598679808 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 18:22:34.020834 139967598679808 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0425 18:22:34.020889 139967598679808 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0425 18:22:34.022739 139967598679808 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0425 18:22:34.022950 140092765345600 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 18:22:34.023103 140092765345600 checkpoint_manager.py:1458] Waiting for previous save to complete took 43.567744 seconds. If this number is high, consider checkpointing less frequently. I0425 18:22:34.025029 140092765345600 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0425 18:22:34.028693 140092765345600 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_07_eval/checkpoints/9. I0425 18:22:35.028215 140092765345600 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0425 18:22:35.028315 140092765345600 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 18:22:35.070638 140092765345600 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.043718s I0425 18:22:35.070813 140092765345600 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.222 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2953801155090332 s) (per-host) I0425 18:22:35.070866 140092765345600 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.295441s (batch_requests_ready=0.249785s, total_serialization_initiated=0.045589s, others=0.000068s) I0425 18:22:35.070956 140092765345600 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.299654s (all_items=0.000016s, per_item={'items': '0.00001645'}, temp_paths=0.299638) I0425 18:22:35.071841 139968135550720 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0425 18:22:35.072009 140092765345600 async_checkpointer.py:561] Finished blocking save. Time taken: 1.046902s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_07_eval/checkpoints/9. I0425 18:22:35.495289 140092765345600 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0425 18:22:35.495696 139967598679808 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0425 18:22:35.495871 140092765345600 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777141310.4553282, 'wait_for_prev_duration_secs': 43.56774401664734, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777141354.0250738, 'checkpointer_blocking_duration_secs': 1.0470404624938965, 'get_old_steps_start_time': 1777141355.072133, 'get_old_steps_duration_secs': 2.765655517578125e-05, 'checkpoint_manager_blocking_start_time': 1777141310.4537072, 'checkpoint_manager_blocking_duration_secs': 45.04212689399719} I0425 18:22:35.495982 140092765345600 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0425 18:22:35.496032 140092765345600 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 18:22:39.903023 139972413732608 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_07_eval/checkpoints/9/items/array_metadatas/process_5 I0425 18:23:17.533460 139968135550720 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 36.943 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.757986545562744 s) (per-host) I0425 18:23:17.533582 139968135550720 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.461629s. I0425 18:23:25.114151 139968135550720 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 50.042182s. I0425 18:23:25.114512 139967598679808 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 18:23:25.114626 139967598679808 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0425 18:23:25.114700 139967598679808 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0425 18:23:25.115898 139967598679808 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0425 18:23:25.116080 140092765345600 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 18:23:25.116219 140092765345600 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0425 18:23:25.116895 140092765345600 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.240, Tokens/s/device: 13903.409, 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: Sat Apr 25 18:23:36 UTC 2026 EXIT_CODE=0