XPK Start: Mon Apr 20 22:49:30 UTC 2026 2026-04-20 22:49:34.545267: 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 00:00:1776725374.558000 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776725374.561685 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776725374.572927 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776725374.572946 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776725374.572949 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776725374.572951 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-20 22:49:53.555374: 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 22:49:54.064901 134209225172800 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-20 22:50:03,105:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0420 22:50:03.105524 134209225172800 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-20 22:50:03,107:jax._src.distributed:157: Connecting to JAX distributed service on mt-07-eval-54tmv-slice-job-0-0.mt-07-eval-54tmv:8482 I0420 22:50:03.107855 134209225172800 distributed.py:157] Connecting to JAX distributed service on mt-07-eval-54tmv-slice-job-0-0.mt-07-eval-54tmv:8482 I0420 22:50:04.929539 134209225172800 max_utils.py:284] Jax distributed system initialized! I0420 22:50:12.114168 134209225172800 max_utils.py:800] System Information: Jax Version: 0.8.1 I0420 22:50:12.114277 134209225172800 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0420 22:50:12.114318 134209225172800 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 22:50:12.114352 134209225172800 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0420 22:50:12.779209 134209225172800 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0420 22:50:12.779507 134209225172800 checkpointing.py:677] Setting up checkpoint logger... I0420 22:50:12.779564 134209225172800 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0420 22:50:12.779609 134209225172800 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0420 22:50:12.779972 134209225172800 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 0x7a0f7bd712b0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0420 22:50:16.019806 134209225172800 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0420 22:50:16.020041 134209225172800 checkpoint_manager.py:709] [process=4][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a0e3ffbbef0>}, handler_registry=None I0420 22:50:16.020276 134209225172800 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a0e3ffbbef0>` 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 22:50:16.020325 134209225172800 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79fae051b470>` 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 22:50:16.020360 134209225172800 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 0x7a0e3ffbbef0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a0e3ffbbef0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79fae051b470>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79fae051b470>}). I0420 22:50:16.020865 134209225172800 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0420 22:50:16.020944 134209225172800 async_checkpointer.py:177] [process=4][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x79fae01faac0> timeout: 600 secs and primary_host=0 for async checkpoint writes I0420 22:50:17.703850 134209225172800 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_07_eval/checkpoints I0420 22:50:18.114457 134209225172800 checkpoint_manager.py:929] [process=4][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_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x79fae073fd70> I0420 22:50:18.114657 134209225172800 checkpointing.py:301] Checkpoint manager created! I0420 22:50:19.007342 134209225172800 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0420 22:50:19.007447 134209225172800 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 22:50:19.377712 134209225172800 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0420 22:50:19.377813 134209225172800 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 22:50:19.393744 134209225172800 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0420 22:50:19.393801 134209225172800 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 22:50:19.416793 134209225172800 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 22:50:19.416861 134209225172800 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 22:50:19.432837 134209225172800 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 22:50:19.432898 134209225172800 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 22:50:19.448766 134209225172800 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 22:50:19.448827 134209225172800 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 22:50:19.473709 134209225172800 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0420 22:50:19.473777 134209225172800 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 22:50:19.493188 134209225172800 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0420 22:50:19.493252 134209225172800 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0420 22:50:19.681070 134209225172800 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0420 22:50:19.681185 134209225172800 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0420 22:50:21.098134 134209225172800 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0420 22:50:21.098265 134209225172800 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 22:50:21.098320 134209225172800 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 22:50:21.098376 134209225172800 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 22:50:21.098435 134209225172800 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 22:50:21.098480 134209225172800 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 22:50:21.098536 134209225172800 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 22:50:21.098588 134209225172800 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 22:50:21.098629 134209225172800 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 22:50:21.098666 134209225172800 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 22:50:21.098737 134209225172800 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0420 22:50:21.098794 134209225172800 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0420 22:50:21.594852 134209225172800 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0420 22:50:21.594950 134209225172800 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0420 22:50:21.609676 134209225172800 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0420 22:50:21.609748 134209225172800 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0420 22:50:32.153787 134209225172800 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. I0420 22:50:32.154624 134209225172800 metric_logger.py:289] number parameters: 1.104 billion I0420 22:50:43.319450 134209225172800 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0420 22:50:43.481242 134209225172800 checkpointing.py:776] Waited 0.16177821159362793 seconds for step 0 to finish before starting checkpointing. I0420 22:50:43.483680 134209225172800 checkpoint_manager.py:2013] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0420 22:50:43.485332 134209225172800 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 0 I0420 22:50:43.488067 134209225172800 async_checkpointer.py:452] [process=4] 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_07_eval/checkpoints/0. I0420 22:50:44.079859 134209225172800 signaling_client.py:364] Using JaxDistributedSignalingClient I0420 22:50:44.080793 134209225172800 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0420 22:50:44.080849 134209225172800 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 22:50:44.334564 134209225172800 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.254785s I0420 22:50:44.334745 134209225172800 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.000 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5141952037811279 s) (per-host) I0420 22:50:44.334800 134209225172800 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.514260s (batch_requests_ready=0.255832s, total_serialization_initiated=0.258354s, others=0.000074s) I0420 22:50:44.334891 134209225172800 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.518247s (all_items=0.000021s, per_item={'items': '0.00002050'}, temp_paths=0.518226) I0420 22:50:44.335816 134083350873856 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started. I0420 22:50:44.335986 134209225172800 async_checkpointer.py:561] Finished blocking save. Time taken: 0.850580s. 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_07_eval/checkpoints/0. I0420 22:50:44.380006 134209225172800 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0420 22:50:44.380355 134081765439232 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save. I0420 22:50:44.380528 134209225172800 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_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776725443.4836612, 'wait_for_prev_duration_secs': 7.939338684082031e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776725443.4853714, 'checkpointer_blocking_duration_secs': 0.8507211208343506, 'get_old_steps_start_time': 1776725444.336112, 'get_old_steps_duration_secs': 2.5987625122070312e-05, 'checkpoint_manager_blocking_start_time': 1776725443.481775, 'checkpoint_manager_blocking_duration_secs': 0.898712158203125} I0420 22:50:44.380633 134209225172800 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0420 22:50:44.380685 134209225172800 max_utils.py:750] Memstats: After params initialized: I0420 22:50:44.380752 134209225172800 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_16(process=4,(0,4,0,0)) I0420 22:50:44.380786 134209225172800 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_17(process=4,(1,4,0,0)) I0420 22:50:44.380817 134209225172800 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_20(process=4,(0,5,0,0)) I0420 22:50:44.380843 134209225172800 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_21(process=4,(1,5,0,0)) I0420 22:50:44.648652 134209225172800 metric_logger.py:185] completed step: 0, seconds: 11.165, TFLOP/s/device: 1.217, Tokens/s/device: 183.435, total_weights: 65536, loss: 10.874 I0420 22:50:44.816320 134209225172800 metric_logger.py:185] completed step: 1, seconds: 1.328, TFLOP/s/device: 10.232, Tokens/s/device: 1542.330, total_weights: 65536, loss: 10.874 I0420 22:50:45.226289 134209225172800 metric_logger.py:185] completed step: 2, seconds: 0.020, TFLOP/s/device: 668.790, Tokens/s/device: 100807.246, total_weights: 65536, loss: 10.263 I0420 22:50:45.373673 134209225172800 metric_logger.py:185] completed step: 3, seconds: 0.411, TFLOP/s/device: 33.082, Tokens/s/device: 4986.438, total_weights: 65536, loss: 9.734 I0420 22:50:45.668864 134209225172800 metric_logger.py:185] completed step: 4, seconds: 0.152, TFLOP/s/device: 89.206, Tokens/s/device: 13446.084, total_weights: 65536, loss: 9.278 I0420 22:50:45.674197 134209225172800 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.202, Tokens/s/device: 13897.748, total_weights: 65536, loss: 8.892 I0420 22:51:04.506268 134209225172800 metric_logger.py:185] completed step: 6, seconds: 0.295, TFLOP/s/device: 45.989, Tokens/s/device: 6931.987, total_weights: 65536, loss: 8.592 I0420 22:51:04.653729 134209225172800 metric_logger.py:185] completed step: 7, seconds: 18.686, TFLOP/s/device: 0.727, Tokens/s/device: 109.603, total_weights: 65536, loss: 8.384 I0420 22:51:04.801132 134209225172800 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.608, Tokens/s/device: 13506.651, total_weights: 65536, loss: 8.255 I0420 22:51:04.948177 134209225172800 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0420 22:51:04.948775 134209225172800 checkpointing.py:776] Waited 0.0006158351898193359 seconds for step 9 to finish before starting checkpointing. I0420 22:51:04.950972 134209225172800 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0420 22:51:05.879045 2625 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0420 22:51:08.717866 134083359266560 array_metadata_store.py:203] [process=4][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_07_eval/checkpoints/0/items/array_metadatas/process_4 I0420 22:51:39.655539 134083350873856 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 28.290 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 55.8349506855011 s) (per-host) I0420 22:51:39.655647 134083350873856 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 55.319718s. I0420 22:51:50.470273 134083350873856 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 66.134318s. I0420 22:51:50.470566 134081765439232 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save. I0420 22:51:50.470726 134081765439232 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save. I0420 22:51:50.470786 134081765439232 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0420 22:51:50.472527 134081765439232 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0420 22:51:50.472724 134209225172800 checkpoint_manager.py:2036] [process=4][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0420 22:51:50.472853 134209225172800 checkpoint_manager.py:1458] Waiting for previous save to complete took 45.521885 seconds. If this number is high, consider checkpointing less frequently. I0420 22:51:50.474807 134209225172800 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 9 I0420 22:51:50.478432 134209225172800 async_checkpointer.py:452] [process=4] 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_07_eval/checkpoints/9. I0420 22:51:52.724746 134209225172800 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0420 22:51:52.724843 134209225172800 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 22:51:52.777646 134209225172800 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.054320s I0420 22:51:52.777832 134209225172800 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.331 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 1.1585984230041504 s) (per-host) I0420 22:51:52.777887 134209225172800 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 1.158663s (batch_requests_ready=1.102428s, total_serialization_initiated=0.056164s, others=0.000072s) I0420 22:51:52.778009 134209225172800 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 1.162671s (all_items=0.000016s, per_item={'items': '0.00001574'}, temp_paths=1.162655) I0420 22:51:52.779000 134082830788352 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started. I0420 22:51:52.779170 134209225172800 async_checkpointer.py:561] Finished blocking save. Time taken: 2.304287s. 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_07_eval/checkpoints/9. I0420 22:51:52.784730 134209225172800 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0420 22:51:52.785023 134081765439232 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save. I0420 22:51:52.785186 134209225172800 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_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776725464.9509432, 'wait_for_prev_duration_secs': 45.52188539505005, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776725510.4748478, 'checkpointer_blocking_duration_secs': 2.3044321537017822, 'get_old_steps_start_time': 1776725512.7792993, 'get_old_steps_duration_secs': 2.47955322265625e-05, 'checkpoint_manager_blocking_start_time': 1776725464.9489985, 'checkpoint_manager_blocking_duration_secs': 47.83615326881409} I0420 22:51:52.785289 134209225172800 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0420 22:51:52.785335 134209225172800 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0420 22:52:00.556666 134083350873856 array_metadata_store.py:203] [process=4][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_07_eval/checkpoints/9/items/array_metadatas/process_4 I0420 22:52:35.184940 134082830788352 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 36.258 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.56566762924194 s) (per-host) I0420 22:52:35.185060 134082830788352 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.405947s. I0420 22:52:47.237940 134082830788352 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 54.458808s. I0420 22:52:47.238276 134081765439232 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save. I0420 22:52:47.238393 134081765439232 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save. I0420 22:52:47.238441 134081765439232 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0420 22:52:47.240081 134081765439232 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0420 22:52:47.240285 134209225172800 checkpoint_manager.py:2036] [process=4][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0420 22:52:47.240437 134209225172800 checkpoint_manager.py:2013] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0420 22:52:47.241035 134209225172800 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.154, Tokens/s/device: 13890.490, 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: Mon Apr 20 22:52:56 UTC 2026 EXIT_CODE=0