XPK Start: Thu Apr 23 19:24:07 UTC 2026 2026-04-23 19:24:11.421212: 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 00:00:1776972251.434193 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776972251.438054 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776972251.449383 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776972251.449402 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776972251.449405 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776972251.449407 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-23 19:24:30.703085: 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) I0423 19:24:31.225023 136549911996224 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-23 19:24:40,266:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0423 19:24:40.266690 136549911996224 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-23 19:24:40,268:jax._src.distributed:157: Connecting to JAX distributed service on mt-07-eval-fyuxh-slice-job-0-0.mt-07-eval-fyuxh:8482 I0423 19:24:40.268993 136549911996224 distributed.py:157] Connecting to JAX distributed service on mt-07-eval-fyuxh-slice-job-0-0.mt-07-eval-fyuxh:8482 I0423 19:24:42.012187 136549911996224 max_utils.py:284] Jax distributed system initialized! I0423 19:24:48.943549 136549911996224 max_utils.py:800] System Information: Jax Version: 0.8.1 I0423 19:24:48.943668 136549911996224 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0423 19:24:48.943713 136549911996224 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 I0423 19:24:48.943749 136549911996224 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0423 19:24:49.561407 136549911996224 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0423 19:24:49.561719 136549911996224 checkpointing.py:677] Setting up checkpoint logger... I0423 19:24:49.561777 136549911996224 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0423 19:24:49.561821 136549911996224 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0423 19:24:49.562178 136549911996224 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 0x7c3077971070>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0423 19:24:52.361680 136549911996224 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0423 19:24:52.362029 136549911996224 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 0x7c1bec3f4440>}, handler_registry=None I0423 19:24:52.362270 136549911996224 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c1bec3f4440>` 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`. I0423 19:24:52.362319 136549911996224 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c1bec3f7ec0>` 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`. I0423 19:24:52.362356 136549911996224 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 0x7c1bec3f4440>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c1bec3f4440>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c1bec3f7ec0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c1bec3f7ec0>}). I0423 19:24:52.362874 136549911996224 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0423 19:24:52.362957 136549911996224 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7c1bec1ee840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0423 19:24:53.151198 136549911996224 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_07_eval/checkpoints I0423 19:24:53.589677 136549911996224 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_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7c2f3bbdcb60> I0423 19:24:53.589850 136549911996224 checkpointing.py:301] Checkpoint manager created! I0423 19:24:54.491201 136549911996224 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0423 19:24:54.491307 136549911996224 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0423 19:24:54.862976 136549911996224 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0423 19:24:54.863069 136549911996224 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0423 19:24:54.884097 136549911996224 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0423 19:24:54.884197 136549911996224 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0423 19:24:54.908049 136549911996224 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 19:24:54.908116 136549911996224 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 19:24:54.924266 136549911996224 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 19:24:54.924325 136549911996224 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 19:24:54.940378 136549911996224 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0423 19:24:54.940436 136549911996224 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 19:24:54.964625 136549911996224 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0423 19:24:54.964704 136549911996224 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0423 19:24:54.984610 136549911996224 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0423 19:24:54.984687 136549911996224 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0423 19:24:55.184205 136549911996224 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0423 19:24:55.184321 136549911996224 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0423 19:24:56.605686 136549911996224 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0423 19:24:56.605815 136549911996224 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) I0423 19:24:56.605870 136549911996224 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) I0423 19:24:56.605929 136549911996224 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') I0423 19:24:56.605982 136549911996224 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) I0423 19:24:56.606022 136549911996224 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) I0423 19:24:56.606075 136549911996224 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) I0423 19:24:56.606130 136549911996224 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') I0423 19:24:56.606172 136549911996224 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) I0423 19:24:56.606210 136549911996224 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) I0423 19:24:56.606265 136549911996224 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0423 19:24:56.606314 136549911996224 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0423 19:24:57.124883 136549911996224 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0423 19:24:57.124987 136549911996224 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0423 19:24:57.140063 136549911996224 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0423 19:24:57.140125 136549911996224 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0423 19:25:07.821344 136549911996224 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. I0423 19:25:07.822200 136549911996224 metric_logger.py:289] number parameters: 1.104 billion I0423 19:25:18.965131 136549911996224 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0423 19:25:19.118860 136549911996224 checkpointing.py:776] Waited 0.15370941162109375 seconds for step 0 to finish before starting checkpointing. I0423 19:25:19.121304 136549911996224 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0423 19:25:19.123149 136549911996224 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0423 19:25:19.126012 136549911996224 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_07_eval/checkpoints/0. I0423 19:25:19.690969 136549911996224 signaling_client.py:364] Using JaxDistributedSignalingClient I0423 19:25:19.691943 136549911996224 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0423 19:25:19.692001 136549911996224 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 I0423 19:25:19.940592 136549911996224 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.249698s I0423 19:25:19.940772 136549911996224 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.054 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5050439834594727 s) (per-host) I0423 19:25:19.940825 136549911996224 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.505106s (batch_requests_ready=0.251678s, total_serialization_initiated=0.253358s, others=0.000070s) I0423 19:25:19.940920 136549911996224 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.509142s (all_items=0.000019s, per_item={'items': '0.00001931'}, temp_paths=0.509123) I0423 19:25:19.941857 136424426809088 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0423 19:25:19.942036 136549911996224 async_checkpointer.py:561] Finished blocking save. Time taken: 0.818811s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_07_eval/checkpoints/0. I0423 19:25:19.963617 136549911996224 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0423 19:25:19.963935 136424451987200 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0423 19:25:19.964085 136549911996224 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776972319.1212864, 'wait_for_prev_duration_secs': 6.29425048828125e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776972319.1231873, 'checkpointer_blocking_duration_secs': 0.8189516067504883, 'get_old_steps_start_time': 1776972319.94216, 'get_old_steps_duration_secs': 2.4080276489257812e-05, 'checkpoint_manager_blocking_start_time': 1776972319.1194737, 'checkpoint_manager_blocking_duration_secs': 0.8445580005645752} I0423 19:25:19.964205 136549911996224 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0423 19:25:19.964270 136549911996224 max_utils.py:750] Memstats: After params initialized: I0423 19:25:19.964328 136549911996224 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0)) I0423 19:25:19.964363 136549911996224 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0)) I0423 19:25:19.964391 136549911996224 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0)) I0423 19:25:19.964417 136549911996224 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0)) I0423 19:25:20.233402 136549911996224 metric_logger.py:185] completed step: 0, seconds: 11.143, TFLOP/s/device: 1.219, Tokens/s/device: 183.796, total_weights: 65536, loss: 10.874 I0423 19:25:20.402005 136549911996224 metric_logger.py:185] completed step: 1, seconds: 1.267, TFLOP/s/device: 10.725, Tokens/s/device: 1616.533, total_weights: 65536, loss: 10.874 I0423 19:25:20.807039 136549911996224 metric_logger.py:185] completed step: 2, seconds: 0.021, TFLOP/s/device: 633.108, Tokens/s/device: 95428.918, total_weights: 65536, loss: 10.263 I0423 19:25:20.954476 136549911996224 metric_logger.py:185] completed step: 3, seconds: 0.406, TFLOP/s/device: 33.504, Tokens/s/device: 5050.032, total_weights: 65536, loss: 9.734 I0423 19:25:21.249855 136549911996224 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 89.021, Tokens/s/device: 13418.158, total_weights: 65536, loss: 9.278 I0423 19:25:21.255515 136549911996224 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.222, Tokens/s/device: 13900.767, total_weights: 65536, loss: 8.892 I0423 19:25:39.640064 136549911996224 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.942, Tokens/s/device: 6924.838, total_weights: 65536, loss: 8.592 I0423 19:25:39.787590 136549911996224 metric_logger.py:185] completed step: 7, seconds: 18.238, TFLOP/s/device: 0.745, Tokens/s/device: 112.294, total_weights: 65536, loss: 8.384 I0423 19:25:39.935184 136549911996224 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.303, Tokens/s/device: 13460.755, total_weights: 65536, loss: 8.255 I0423 19:25:40.082304 136549911996224 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0423 19:25:40.082962 136549911996224 checkpointing.py:776] Waited 0.0006783008575439453 seconds for step 9 to finish before starting checkpointing. I0423 19:25:40.084933 136549911996224 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0423 19:25:41.020997 2635 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0423 19:25:42.976964 136424435201792 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_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_07_eval/checkpoints/0/items/array_metadatas/process_5 I0423 19:26:14.722556 136424426809088 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 28.571 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 55.28679037094116 s) (per-host) I0423 19:26:14.722680 136424426809088 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 54.780714s. I0423 19:26:25.903255 136424426809088 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 65.961273s. I0423 19:26:25.903536 136424451987200 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0423 19:26:25.903705 136424451987200 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0423 19:26:25.903762 136424451987200 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0423 19:26:25.905187 136424451987200 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0423 19:26:25.905370 136549911996224 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. W0423 19:26:25.905503 136549911996224 checkpoint_manager.py:1458] Waiting for previous save to complete took 45.820574 seconds. If this number is high, consider checkpointing less frequently. I0423 19:26:25.907516 136549911996224 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0423 19:26:25.911054 136549911996224 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_07_eval/checkpoints/9. I0423 19:26:26.472991 136549911996224 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0423 19:26:26.473090 136549911996224 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 I0423 19:26:26.517909 136549911996224 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.046134s I0423 19:26:26.518076 136549911996224 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 4.996 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.30874037742614746 s) (per-host) I0423 19:26:26.518128 136549911996224 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.308801s (batch_requests_ready=0.260715s, total_serialization_initiated=0.048018s, others=0.000069s) I0423 19:26:26.518221 136549911996224 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.313211s (all_items=0.000016s, per_item={'items': '0.00001574'}, temp_paths=0.313195) I0423 19:26:26.519062 136423386638080 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0423 19:26:26.519219 136549911996224 async_checkpointer.py:561] Finished blocking save. Time taken: 0.611631s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_07_eval/checkpoints/9. I0423 19:26:26.533250 136549911996224 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0423 19:26:26.533498 136424451987200 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0423 19:26:26.533634 136549911996224 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776972340.0849047, 'wait_for_prev_duration_secs': 45.82057428359985, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776972385.9075541, 'checkpointer_blocking_duration_secs': 0.611778736114502, 'get_old_steps_start_time': 1776972386.5193517, 'get_old_steps_duration_secs': 2.5033950805664062e-05, 'checkpoint_manager_blocking_start_time': 1776972340.0832005, 'checkpoint_manager_blocking_duration_secs': 46.450398206710815} I0423 19:26:26.533754 136549911996224 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0423 19:26:26.533801 136549911996224 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0423 19:26:31.740602 136424435201792 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_20260423_190840/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260423_190840_07_eval/checkpoints/9/items/array_metadatas/process_5 I0423 19:27:08.428375 136423386638080 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.414 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.21899700164795 s) (per-host) I0423 19:27:08.428498 136423386638080 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.909327s. I0423 19:27:17.039846 136423386638080 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 50.520659s. I0423 19:27:17.040072 136424451987200 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0423 19:27:17.040138 136424451987200 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0423 19:27:17.040184 136424451987200 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0423 19:27:17.041799 136424451987200 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0423 19:27:17.041951 136549911996224 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. I0423 19:27:17.042098 136549911996224 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0423 19:27:17.042785 136549911996224 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.204, Tokens/s/device: 13897.937, 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: Thu Apr 23 19:27:27 UTC 2026 EXIT_CODE=0