XPK Start: Wed Apr 22 19:23:37 UTC 2026 2026-04-22 19:23:41.360722: 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 00:00:1776885821.374255 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776885821.378083 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776885821.390037 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776885821.390056 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776885821.390059 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776885821.390061 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-22 19:24:00.723603: 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) I0422 19:24:01.248250 137165445179200 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-22 19:24:10,290:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0422 19:24:10.290317 137165445179200 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-22 19:24:10,292:jax._src.distributed:157: Connecting to JAX distributed service on mt-07-eval-0bqdi-slice-job-0-0.mt-07-eval-0bqdi:8482 I0422 19:24:10.292644 137165445179200 distributed.py:157] Connecting to JAX distributed service on mt-07-eval-0bqdi-slice-job-0-0.mt-07-eval-0bqdi:8482 I0422 19:24:11.586260 137165445179200 max_utils.py:284] Jax distributed system initialized! I0422 19:24:18.390615 137165445179200 max_utils.py:800] System Information: Jax Version: 0.8.1 I0422 19:24:18.390719 137165445179200 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0422 19:24:18.390765 137165445179200 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 I0422 19:24:18.390808 137165445179200 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0422 19:24:19.015075 137165445179200 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0422 19:24:19.015394 137165445179200 checkpointing.py:677] Setting up checkpoint logger... I0422 19:24:19.015451 137165445179200 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0422 19:24:19.015496 137165445179200 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0422 19:24:19.015852 137165445179200 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 0x7cbfcfd711c0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0422 19:24:22.351607 137165445179200 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0422 19:24:22.351850 137165445179200 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 0x7cbe8c69e300>}, handler_registry=None I0422 19:24:22.352081 137165445179200 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7cbe8c69e300>` 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`. I0422 19:24:22.352143 137165445179200 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7caadc60b110>` 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`. I0422 19:24:22.352179 137165445179200 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 0x7cbe8c69e300>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7cbe8c69e300>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7caadc60b110>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7caadc60b110>}). I0422 19:24:22.352689 137165445179200 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0422 19:24:22.352767 137165445179200 async_checkpointer.py:177] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7caadc3be840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0422 19:24:23.898520 137165445179200 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_07_eval/checkpoints I0422 19:24:24.056811 137165445179200 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_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7caadc608770> I0422 19:24:24.056988 137165445179200 checkpointing.py:301] Checkpoint manager created! I0422 19:24:24.954517 137165445179200 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0422 19:24:24.954631 137165445179200 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 19:24:25.322472 137165445179200 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 19:24:25.322567 137165445179200 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 19:24:25.338207 137165445179200 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 19:24:25.338268 137165445179200 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 19:24:25.366591 137165445179200 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 19:24:25.366696 137165445179200 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 19:24:25.382571 137165445179200 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 19:24:25.382635 137165445179200 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 19:24:25.398374 137165445179200 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 19:24:25.398436 137165445179200 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 19:24:25.422486 137165445179200 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0422 19:24:25.422560 137165445179200 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 19:24:25.442076 137165445179200 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0422 19:24:25.442153 137165445179200 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0422 19:24:25.633427 137165445179200 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0422 19:24:25.633540 137165445179200 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0422 19:24:27.064374 137165445179200 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0422 19:24:27.064506 137165445179200 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) I0422 19:24:27.064560 137165445179200 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) I0422 19:24:27.064617 137165445179200 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') I0422 19:24:27.064668 137165445179200 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) I0422 19:24:27.064707 137165445179200 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) I0422 19:24:27.064758 137165445179200 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) I0422 19:24:27.064808 137165445179200 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') I0422 19:24:27.064848 137165445179200 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) I0422 19:24:27.064884 137165445179200 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) I0422 19:24:27.064931 137165445179200 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0422 19:24:27.064980 137165445179200 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0422 19:24:27.580077 137165445179200 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0422 19:24:27.580192 137165445179200 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0422 19:24:27.595184 137165445179200 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0422 19:24:27.595246 137165445179200 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0422 19:24:38.489840 137165445179200 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. I0422 19:24:38.491220 137165445179200 metric_logger.py:289] number parameters: 1.104 billion I0422 19:24:49.623497 137165445179200 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0422 19:24:49.776206 137165445179200 checkpointing.py:776] Waited 0.15268874168395996 seconds for step 0 to finish before starting checkpointing. I0422 19:24:49.778602 137165445179200 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 19:24:49.780455 137165445179200 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 0 I0422 19:24:49.783248 137165445179200 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_07_eval/checkpoints/0. I0422 19:24:50.750678 137165445179200 signaling_client.py:364] Using JaxDistributedSignalingClient I0422 19:24:50.751904 137165445179200 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0422 19:24:50.751961 137165445179200 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 I0422 19:24:51.001080 137165445179200 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.250481s I0422 19:24:51.001269 137165445179200 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.054 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5050485134124756 s) (per-host) I0422 19:24:51.001328 137165445179200 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.505118s (batch_requests_ready=0.250989s, total_serialization_initiated=0.254050s, others=0.000078s) I0422 19:24:51.001424 137165445179200 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.509209s (all_items=0.000038s, per_item={'items': '0.00003767'}, temp_paths=0.509172) I0422 19:24:51.002335 137043304752896 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0422 19:24:51.002511 137165445179200 async_checkpointer.py:561] Finished blocking save. Time taken: 1.221977s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_07_eval/checkpoints/0. I0422 19:24:51.046811 137165445179200 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0422 19:24:51.047164 137038514878208 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0422 19:24:51.047352 137165445179200 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776885889.778584, 'wait_for_prev_duration_secs': 6.29425048828125e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776885889.7804945, 'checkpointer_blocking_duration_secs': 1.222137689590454, 'get_old_steps_start_time': 1776885891.0026538, 'get_old_steps_duration_secs': 3.457069396972656e-05, 'checkpoint_manager_blocking_start_time': 1776885889.7768276, 'checkpoint_manager_blocking_duration_secs': 1.270465612411499} I0422 19:24:51.047460 137165445179200 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0422 19:24:51.047514 137165445179200 max_utils.py:750] Memstats: After params initialized: I0422 19:24:51.047569 137165445179200 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0)) I0422 19:24:51.047603 137165445179200 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0)) I0422 19:24:51.047632 137165445179200 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0)) I0422 19:24:51.047657 137165445179200 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0)) I0422 19:24:51.310271 137165445179200 metric_logger.py:185] completed step: 0, seconds: 11.132, TFLOP/s/device: 1.221, Tokens/s/device: 183.972, total_weights: 65536, loss: 10.874 I0422 19:24:51.487206 137165445179200 metric_logger.py:185] completed step: 1, seconds: 1.685, TFLOP/s/device: 8.061, Tokens/s/device: 1215.101, total_weights: 65536, loss: 10.874 I0422 19:24:51.899297 137165445179200 metric_logger.py:185] completed step: 2, seconds: 0.030, TFLOP/s/device: 457.310, Tokens/s/device: 68930.699, total_weights: 65536, loss: 10.263 I0422 19:24:52.046690 137165445179200 metric_logger.py:185] completed step: 3, seconds: 0.412, TFLOP/s/device: 32.949, Tokens/s/device: 4966.462, total_weights: 65536, loss: 9.734 I0422 19:24:52.342008 137165445179200 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.987, Tokens/s/device: 13413.149, total_weights: 65536, loss: 9.278 I0422 19:24:52.347666 137165445179200 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.227, Tokens/s/device: 13901.427, total_weights: 65536, loss: 8.892 I0422 19:25:11.125813 137165445179200 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.926, Tokens/s/device: 6922.450, total_weights: 65536, loss: 8.592 I0422 19:25:11.273360 137165445179200 metric_logger.py:185] completed step: 7, seconds: 18.631, TFLOP/s/device: 0.729, Tokens/s/device: 109.922, total_weights: 65536, loss: 8.384 I0422 19:25:11.420600 137165445179200 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.394, Tokens/s/device: 13474.482, total_weights: 65536, loss: 8.255 I0422 19:25:11.567671 137165445179200 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0422 19:25:11.568331 137165445179200 checkpointing.py:776] Waited 0.0006816387176513672 seconds for step 9 to finish before starting checkpointing. I0422 19:25:11.570723 137165445179200 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 19:25:12.305874 2635 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0422 19:25:14.218223 137043313145600 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_07_eval/checkpoints/0/items/array_metadatas/process_6 I0422 19:25:46.379369 137043304752896 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 28.266 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 55.88308310508728 s) (per-host) I0422 19:25:46.379485 137043304752896 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 55.377042s. I0422 19:25:54.986187 137043304752896 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 63.983727s. I0422 19:25:54.986485 137038514878208 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 19:25:54.986601 137038514878208 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0422 19:25:54.986654 137038514878208 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0422 19:25:54.987989 137038514878208 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0422 19:25:54.988209 137165445179200 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. W0422 19:25:54.988357 137165445179200 checkpoint_manager.py:1458] Waiting for previous save to complete took 43.417644 seconds. If this number is high, consider checkpointing less frequently. I0422 19:25:54.990427 137165445179200 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 9 I0422 19:25:54.993760 137165445179200 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_07_eval/checkpoints/9. I0422 19:25:55.546071 137165445179200 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0422 19:25:55.546183 137165445179200 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 I0422 19:25:55.588116 137165445179200 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.043301s I0422 19:25:55.588289 137165445179200 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.349 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28839707374572754 s) (per-host) I0422 19:25:55.588344 137165445179200 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.288461s (batch_requests_ready=0.243221s, total_serialization_initiated=0.045167s, others=0.000073s) I0422 19:25:55.588437 137165445179200 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.292745s (all_items=0.000016s, per_item={'items': '0.00001621'}, temp_paths=0.292729) I0422 19:25:55.589376 137039530747648 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0422 19:25:55.589508 137165445179200 async_checkpointer.py:561] Finished blocking save. Time taken: 0.599006s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_07_eval/checkpoints/9. I0422 19:25:55.613494 137165445179200 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0422 19:25:55.613778 137038514878208 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0422 19:25:55.613941 137165445179200 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776885911.5706844, 'wait_for_prev_duration_secs': 43.417643547058105, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776885954.990467, 'checkpointer_blocking_duration_secs': 0.599224328994751, 'get_old_steps_start_time': 1776885955.589712, 'get_old_steps_duration_secs': 3.075599670410156e-05, 'checkpoint_manager_blocking_start_time': 1776885911.5685608, 'checkpoint_manager_blocking_duration_secs': 44.04534554481506} I0422 19:25:55.614043 137165445179200 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0422 19:25:55.614089 137165445179200 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 19:26:00.874075 137039539140352 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260422_190741_07_eval/checkpoints/9/items/array_metadatas/process_6 I0422 19:26:38.096267 137039530747648 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 36.910 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.79634141921997 s) (per-host) I0422 19:26:38.096386 137039530747648 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.506792s. I0422 19:26:45.665666 137039530747648 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 50.076058s. I0422 19:26:45.665949 137038514878208 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 19:26:45.666076 137038514878208 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0422 19:26:45.666143 137038514878208 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0422 19:26:45.667740 137038514878208 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0422 19:26:45.667916 137165445179200 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. I0422 19:26:45.668067 137165445179200 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 19:26:45.668706 137165445179200 metric_logger.py:185] completed step: 9, seconds: 0.148, TFLOP/s/device: 92.008, Tokens/s/device: 13868.480, 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: Wed Apr 22 19:26:57 UTC 2026 EXIT_CODE=0