XPK Start: Mon Apr 20 22:41:47 UTC 2026 2026-04-20 22:41:51.083415: 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 00:00:1776724911.096208 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776724911.099879 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776724911.111176 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776724911.111196 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776724911.111199 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776724911.111200 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-20 22:42:10.585439: 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 22:42:11.105032 137534388549440 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-20 22:42:20,144:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0420 22:42:20.144107 137534388549440 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-20 22:42:20,146:jax._src.distributed:157: Connecting to JAX distributed service on mt-04-int8-4447v-slice-job-0-0.mt-04-int8-4447v:8482 I0420 22:42:20.146491 137534388549440 distributed.py:157] Connecting to JAX distributed service on mt-04-int8-4447v-slice-job-0-0.mt-04-int8-4447v:8482 I0420 22:42:21.090442 137534388549440 max_utils.py:284] Jax distributed system initialized! I0420 22:42:27.854346 137534388549440 max_utils.py:800] System Information: Jax Version: 0.8.1 I0420 22:42:27.854470 137534388549440 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0420 22:42:27.854511 137534388549440 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 22:42:27.854545 137534388549440 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0420 22:42:28.474976 137534388549440 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0420 22:42:28.475383 137534388549440 checkpointing.py:677] Setting up checkpoint logger... I0420 22:42:28.475452 137534388549440 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0420 22:42:28.475497 137534388549440 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0420 22:42:28.475840 137534388549440 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 0x7d15c5f78bf0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0420 22:42:31.803738 137534388549440 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0420 22:42:31.803985 137534388549440 checkpoint_manager.py:709] [process=1][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d034f035610>}, handler_registry=None I0420 22:42:31.804215 137534388549440 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d034f035610>` 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 22:42:31.804263 137534388549440 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d034f0ccf50>` 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 22:42:31.804297 137534388549440 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 0x7d034f035610>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d034f035610>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d034f0ccf50>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d034f0ccf50>}). I0420 22:42:31.804813 137534388549440 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0420 22:42:31.804897 137534388549440 async_checkpointer.py:177] [process=1][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d034d57a840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0420 22:42:32.476615 137534388549440 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_04_int8/checkpoints I0420 22:42:33.349706 137534388549440 checkpoint_manager.py:929] [process=1][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_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_04_int8/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d034f0d0d70> I0420 22:42:33.349886 137534388549440 checkpointing.py:301] Checkpoint manager created! I0420 22:42:34.521459 137534388549440 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0420 22:42:34.521567 137534388549440 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 22:42:34.896623 137534388549440 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0420 22:42:34.896716 137534388549440 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 22:42:34.912857 137534388549440 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0420 22:42:34.912921 137534388549440 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 22:42:34.990937 137534388549440 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 22:42:34.991026 137534388549440 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 22:42:35.007321 137534388549440 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 22:42:35.007386 137534388549440 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 22:42:35.023051 137534388549440 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 22:42:35.023121 137534388549440 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 22:42:35.047073 137534388549440 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0420 22:42:35.047149 137534388549440 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 22:42:35.128084 137534388549440 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0420 22:42:35.128183 137534388549440 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0420 22:42:35.731246 137534388549440 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0420 22:42:35.731362 137534388549440 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0420 22:42:38.383873 137534388549440 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0420 22:42:38.384006 137534388549440 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) I0420 22:42:38.384062 137534388549440 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) I0420 22:42:38.384121 137534388549440 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') I0420 22:42:38.384183 137534388549440 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) I0420 22:42:38.384221 137534388549440 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) I0420 22:42:38.384276 137534388549440 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) I0420 22:42:38.384328 137534388549440 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') I0420 22:42:38.384368 137534388549440 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) I0420 22:42:38.384425 137534388549440 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) I0420 22:42:38.384487 137534388549440 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0420 22:42:38.384539 137534388549440 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0420 22:42:40.091787 137534388549440 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0420 22:42:40.091892 137534388549440 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0420 22:42:40.106849 137534388549440 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0420 22:42:40.106912 137534388549440 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0420 22:42:54.143175 137534388549440 max_utils.py:791] Total memory size: 1.8 GB, Output size: 0.4 GB, Temp size: 1.4 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0420 22:42:54.144096 137534388549440 metric_logger.py:289] number parameters: 1.104 billion I0420 22:43:08.749639 137534388549440 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0420 22:43:08.990437 137534388549440 checkpointing.py:776] Waited 0.24075865745544434 seconds for step 0 to finish before starting checkpointing. I0420 22:43:08.993037 137534388549440 checkpoint_manager.py:2013] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0420 22:43:08.995110 137534388549440 checkpoint_manager.py:1518] [process=1] Saving checkpoint at step 0 I0420 22:43:08.998447 137534388549440 async_checkpointer.py:452] [process=1] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_04_int8/checkpoints/0. I0420 22:43:10.876633 137534388549440 signaling_client.py:364] Using JaxDistributedSignalingClient I0420 22:43:10.877525 137534388549440 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0420 22:43:10.877582 137534388549440 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 22:43:11.131828 137534388549440 base_pytree_checkpoint_handler.py:153] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.255275s I0420 22:43:11.131997 137534388549440 base_pytree_checkpoint_handler.py:129] [process=1] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.601 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.9637954235076904 s) (per-host) I0420 22:43:11.132048 137534388549440 base_pytree_checkpoint_handler.py:737] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.963856s (batch_requests_ready=0.704937s, total_serialization_initiated=0.258850s, others=0.000069s) I0420 22:43:11.132140 137534388549440 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.967976s (all_items=0.000020s, per_item={'items': '0.00001955'}, temp_paths=0.967956) I0420 22:43:11.133063 137412369966848 async_checkpointer.py:79] [process=1][thread=async_save] Background save thread started. I0420 22:43:11.133235 137534388549440 async_checkpointer.py:561] Finished blocking save. Time taken: 2.138063s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_04_int8/checkpoints/0. I0420 22:43:11.134992 137534388549440 checkpoint_manager.py:1566] [process=1][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0420 22:43:11.135243 137402152634112 async_checkpointer.py:265] [process=1][thread=save_finalize] Waiting for background save thread=async_save. I0420 22:43:11.135390 137534388549440 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_04_int8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776724988.9930184, 'wait_for_prev_duration_secs': 6.628036499023438e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776724988.9951494, 'checkpointer_blocking_duration_secs': 2.1381618976593018, 'get_old_steps_start_time': 1776724991.1333299, 'get_old_steps_duration_secs': 2.288818359375e-05, 'checkpoint_manager_blocking_start_time': 1776724988.990991, 'checkpoint_manager_blocking_duration_secs': 2.14436411857605} I0420 22:43:11.135506 137534388549440 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0420 22:43:11.135555 137534388549440 max_utils.py:750] Memstats: After params initialized: I0420 22:43:11.135602 137534388549440 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_2(process=1,(2,0,0,0)) I0420 22:43:11.135633 137534388549440 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_3(process=1,(3,0,0,0)) I0420 22:43:11.135660 137534388549440 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_6(process=1,(2,1,0,0)) I0420 22:43:11.135684 137534388549440 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_7(process=1,(3,1,0,0)) I0420 22:43:11.405337 137534388549440 metric_logger.py:185] completed step: 0, seconds: 14.605, TFLOP/s/device: 0.930, Tokens/s/device: 140.222, total_weights: 65536, loss: 10.874 I0420 22:43:11.659611 137534388549440 metric_logger.py:185] completed step: 1, seconds: 2.654, TFLOP/s/device: 5.119, Tokens/s/device: 771.550, total_weights: 65536, loss: 10.874 I0420 22:43:12.474443 137534388549440 metric_logger.py:185] completed step: 2, seconds: 0.025, TFLOP/s/device: 536.489, Tokens/s/device: 80865.514, total_weights: 65536, loss: 10.262 I0420 22:43:12.704051 137534388549440 metric_logger.py:185] completed step: 3, seconds: 0.816, TFLOP/s/device: 16.659, Tokens/s/device: 2511.099, total_weights: 65536, loss: 9.731 I0420 22:43:13.163791 137534388549440 metric_logger.py:185] completed step: 4, seconds: 0.234, TFLOP/s/device: 58.094, Tokens/s/device: 8756.552, total_weights: 65536, loss: 9.272 I0420 22:43:13.169133 137534388549440 metric_logger.py:185] completed step: 5, seconds: 0.230, TFLOP/s/device: 59.161, Tokens/s/device: 8917.414, total_weights: 65536, loss: 8.887 I0420 22:43:41.260393 137534388549440 metric_logger.py:185] completed step: 6, seconds: 0.460, TFLOP/s/device: 29.545, Tokens/s/device: 4453.307, total_weights: 65536, loss: 8.588 I0420 22:43:41.491081 137534388549440 metric_logger.py:185] completed step: 7, seconds: 27.863, TFLOP/s/device: 0.488, Tokens/s/device: 73.502, total_weights: 65536, loss: 8.380 I0420 22:43:41.721794 137534388549440 metric_logger.py:185] completed step: 8, seconds: 0.233, TFLOP/s/device: 58.202, Tokens/s/device: 8772.907, total_weights: 65536, loss: 8.251 I0420 22:43:41.951824 137534388549440 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0420 22:43:41.952533 137534388549440 checkpointing.py:776] Waited 0.000728607177734375 seconds for step 9 to finish before starting checkpointing. I0420 22:43:41.954735 137534388549440 checkpoint_manager.py:2024] [process=1][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0420 22:43:45.411508 2814 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0420 22:43:51.325796 137412378359552 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_04_int8/checkpoints/0/items/array_metadatas/process_1 I0420 22:44:14.781025 137412369966848 base_pytree_checkpoint_handler.py:129] [process=1] /jax/checkpoint/write/gbytes_per_sec: 24.447 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 64.61278796195984 s) (per-host) I0420 22:44:14.781147 137412369966848 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 63.647953s. I0420 22:44:24.775424 137412369966848 async_checkpointer.py:144] [process=1][thread=async_save] Background save thread done. Time taken: 73.642208s. I0420 22:44:24.775736 137402152634112 async_checkpointer.py:273] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save. I0420 22:44:24.775879 137402152634112 async_checkpointer.py:283] [process=1][thread=save_finalize] No errors found in background save thread=async_save. I0420 22:44:24.775933 137402152634112 checkpoint_manager.py:2133] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0420 22:44:24.778548 137402152634112 checkpoint_manager.py:2142] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0420 22:44:24.778683 137534388549440 checkpoint_manager.py:2036] [process=1][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0420 22:44:24.778785 137534388549440 checkpoint_manager.py:1458] Waiting for previous save to complete took 42.824052 seconds. If this number is high, consider checkpointing less frequently. I0420 22:44:24.780699 137534388549440 checkpoint_manager.py:1518] [process=1] Saving checkpoint at step 9 I0420 22:44:24.784272 137534388549440 async_checkpointer.py:452] [process=1] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_04_int8/checkpoints/9. I0420 22:44:26.255168 137534388549440 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0420 22:44:26.255270 137534388549440 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 22:44:26.302557 137534388549440 base_pytree_checkpoint_handler.py:153] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.048579s I0420 22:44:26.302722 137534388549440 base_pytree_checkpoint_handler.py:129] [process=1] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.872 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.8241181373596191 s) (per-host) I0420 22:44:26.302772 137534388549440 base_pytree_checkpoint_handler.py:737] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.824177s (batch_requests_ready=0.773474s, total_serialization_initiated=0.050636s, others=0.000067s) I0420 22:44:26.302861 137534388549440 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.828279s (all_items=0.000015s, per_item={'items': '0.00001478'}, temp_paths=0.828264) I0420 22:44:26.303758 137405910730496 async_checkpointer.py:79] [process=1][thread=async_save] Background save thread started. I0420 22:44:26.303924 137534388549440 async_checkpointer.py:561] Finished blocking save. Time taken: 1.523151s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_04_int8/checkpoints/9. I0420 22:44:26.309873 137534388549440 checkpoint_manager.py:1566] [process=1][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0420 22:44:26.310122 137402152634112 async_checkpointer.py:265] [process=1][thread=save_finalize] Waiting for background save thread=async_save. I0420 22:44:26.310282 137534388549440 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_04_int8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776725021.9547071, 'wait_for_prev_duration_secs': 42.82405185699463, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776725064.7807393, 'checkpointer_blocking_duration_secs': 1.523294448852539, 'get_old_steps_start_time': 1776725066.3040528, 'get_old_steps_duration_secs': 2.47955322265625e-05, 'checkpoint_manager_blocking_start_time': 1776725021.9527614, 'checkpoint_manager_blocking_duration_secs': 44.357484102249146} I0420 22:44:26.310393 137534388549440 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0420 22:44:26.310449 137534388549440 checkpoint_manager.py:2024] [process=1][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0420 22:44:31.923614 137412378359552 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_04_int8/checkpoints/9/items/array_metadatas/process_1 I0420 22:45:09.248015 137405910730496 base_pytree_checkpoint_handler.py:129] [process=1] /jax/checkpoint/write/gbytes_per_sec: 36.089 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.769368410110474 s) (per-host) I0420 22:45:09.248141 137405910730496 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.944270s. I0420 22:45:16.556999 137405910730496 async_checkpointer.py:144] [process=1][thread=async_save] Background save thread done. Time taken: 50.253114s. I0420 22:45:16.557321 137402152634112 async_checkpointer.py:273] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save. I0420 22:45:16.557453 137402152634112 async_checkpointer.py:283] [process=1][thread=save_finalize] No errors found in background save thread=async_save. I0420 22:45:16.557502 137402152634112 checkpoint_manager.py:2133] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0420 22:45:16.559692 137402152634112 checkpoint_manager.py:2142] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0420 22:45:16.559864 137534388549440 checkpoint_manager.py:2036] [process=1][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0420 22:45:16.560004 137534388549440 checkpoint_manager.py:2013] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0420 22:45:16.560717 137534388549440 metric_logger.py:185] completed step: 9, seconds: 0.231, TFLOP/s/device: 58.924, Tokens/s/device: 8881.680, 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: Mon Apr 20 22:45:27 UTC 2026 EXIT_CODE=0