XPK Start: Wed Apr 22 19:10:08 UTC 2026 2026-04-22 19:10:12.579138: 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 00:00:1776885012.591988 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776885012.595947 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776885012.607290 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776885012.607314 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776885012.607316 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776885012.607318 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-22 19:10:32.489274: 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 19:10:33.008001 132329347016512 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-22 19:10:42,048:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0422 19:10:42.048935 132329347016512 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-22 19:10:42,051:jax._src.distributed:157: Connecting to JAX distributed service on mt-02-synthetic-gzqkq-slice-job-0-0.mt-02-synthetic-gzqkq:8482 I0422 19:10:42.051389 132329347016512 distributed.py:157] Connecting to JAX distributed service on mt-02-synthetic-gzqkq-slice-job-0-0.mt-02-synthetic-gzqkq:8482 I0422 19:10:45.314634 132329347016512 max_utils.py:284] Jax distributed system initialized! I0422 19:10:52.392860 132329347016512 max_utils.py:800] System Information: Jax Version: 0.8.1 I0422 19:10:52.392966 132329347016512 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0422 19:10:52.393009 132329347016512 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 19:10:52.393047 132329347016512 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0422 19:10:53.018929 132329347016512 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0422 19:10:53.019236 132329347016512 checkpointing.py:677] Setting up checkpoint logger... I0422 19:10:53.019297 132329347016512 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0422 19:10:53.019346 132329347016512 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0422 19:10:53.019718 132329347016512 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 0x7859d9d70dd0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0422 19:10:56.288634 132329347016512 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0422 19:10:56.288977 132329347016512 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 0x78454c309af0>}, handler_registry=None I0422 19:10:56.289231 132329347016512 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78454c309af0>` 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 19:10:56.289282 132329347016512 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78454c0ebdd0>` 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 19:10:56.289324 132329347016512 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 0x78454c309af0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78454c309af0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78454c0ebdd0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78454c0ebdd0>}). I0422 19:10:56.289819 132329347016512 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0422 19:10:56.289894 132329347016512 async_checkpointer.py:177] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7844c421aac0> timeout: 600 secs and primary_host=0 for async checkpoint writes I0422 19:10:56.964631 132329347016512 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_02_synthetic/checkpoints I0422 19:10:57.426629 132329347016512 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_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x78454c0e81a0> I0422 19:10:57.426804 132329347016512 checkpointing.py:301] Checkpoint manager created! I0422 19:10:58.318299 132329347016512 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0422 19:10:58.318407 132329347016512 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 19:10:58.686325 132329347016512 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 19:10:58.686423 132329347016512 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 19:10:58.702257 132329347016512 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 19:10:58.702316 132329347016512 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 19:10:58.725184 132329347016512 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 19:10:58.725252 132329347016512 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 19:10:58.741180 132329347016512 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 19:10:58.741242 132329347016512 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 19:10:58.757145 132329347016512 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 19:10:58.757208 132329347016512 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 19:10:58.781957 132329347016512 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0422 19:10:58.782036 132329347016512 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 19:10:58.801481 132329347016512 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0422 19:10:58.801548 132329347016512 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0422 19:10:58.990983 132329347016512 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0422 19:10:58.991105 132329347016512 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0422 19:11:00.437633 132329347016512 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0422 19:11:00.437763 132329347016512 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 19:11:00.437818 132329347016512 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 19:11:00.437876 132329347016512 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 19:11:00.437928 132329347016512 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 19:11:00.437966 132329347016512 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 19:11:00.438017 132329347016512 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 19:11:00.438070 132329347016512 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 19:11:00.438135 132329347016512 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 19:11:00.438176 132329347016512 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 19:11:00.438224 132329347016512 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0422 19:11:00.438274 132329347016512 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0422 19:11:00.940066 132329347016512 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0422 19:11:00.940183 132329347016512 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0422 19:11:00.955300 132329347016512 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0422 19:11:00.955365 132329347016512 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0422 19:11:11.722669 132329347016512 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 19:11:11.723611 132329347016512 metric_logger.py:289] number parameters: 1.104 billion I0422 19:11:22.855897 132329347016512 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0422 19:11:23.008533 132329347016512 checkpointing.py:776] Waited 0.15260839462280273 seconds for step 0 to finish before starting checkpointing. I0422 19:11:23.011080 132329347016512 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 19:11:23.012683 132329347016512 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 0 I0422 19:11:23.015499 132329347016512 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_02_synthetic/checkpoints/0. I0422 19:11:24.008220 132329347016512 signaling_client.py:364] Using JaxDistributedSignalingClient I0422 19:11:24.009352 132329347016512 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0422 19:11:24.009411 132329347016512 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 19:11:24.258992 132329347016512 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.250851s I0422 19:11:24.259171 132329347016512 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.991 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5157427787780762 s) (per-host) I0422 19:11:24.259223 132329347016512 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.515805s (batch_requests_ready=0.261355s, total_serialization_initiated=0.254378s, others=0.000072s) I0422 19:11:24.259319 132329347016512 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.519705s (all_items=0.000021s, per_item={'items': '0.00002074'}, temp_paths=0.519684) I0422 19:11:24.260326 132206877996800 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0422 19:11:24.260497 132329347016512 async_checkpointer.py:561] Finished blocking save. Time taken: 1.247743s. 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_02_synthetic/checkpoints/0. I0422 19:11:24.682066 132329347016512 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0422 19:11:24.682489 132202574636800 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0422 19:11:24.682659 132329347016512 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_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776885083.0110633, 'wait_for_prev_duration_secs': 7.987022399902344e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776885083.0127218, 'checkpointer_blocking_duration_secs': 1.2478892803192139, 'get_old_steps_start_time': 1776885084.2606313, 'get_old_steps_duration_secs': 2.5510787963867188e-05, 'checkpoint_manager_blocking_start_time': 1776885083.0090804, 'checkpoint_manager_blocking_duration_secs': 1.6735358238220215} I0422 19:11:24.682824 132329347016512 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0422 19:11:24.682878 132329347016512 max_utils.py:750] Memstats: After params initialized: I0422 19:11:24.682937 132329347016512 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0)) I0422 19:11:24.682970 132329347016512 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0)) I0422 19:11:24.682999 132329347016512 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0)) I0422 19:11:24.683023 132329347016512 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0)) I0422 19:11:24.953052 132329347016512 metric_logger.py:185] completed step: 0, seconds: 11.132, TFLOP/s/device: 1.221, Tokens/s/device: 183.971, total_weights: 65536, loss: 10.874 I0422 19:11:25.117715 132329347016512 metric_logger.py:185] completed step: 1, seconds: 2.096, TFLOP/s/device: 6.483, Tokens/s/device: 977.183, total_weights: 65536, loss: 10.874 I0422 19:11:25.540919 132329347016512 metric_logger.py:185] completed step: 2, seconds: 0.017, TFLOP/s/device: 777.429, Tokens/s/device: 117182.583, total_weights: 65536, loss: 10.263 I0422 19:11:25.688324 132329347016512 metric_logger.py:185] completed step: 3, seconds: 0.424, TFLOP/s/device: 32.074, Tokens/s/device: 4834.590, total_weights: 65536, loss: 9.734 I0422 19:11:25.983781 132329347016512 metric_logger.py:185] completed step: 4, seconds: 0.152, TFLOP/s/device: 89.171, Tokens/s/device: 13440.878, total_weights: 65536, loss: 9.278 I0422 19:11:25.989856 132329347016512 metric_logger.py:185] completed step: 5, seconds: 0.148, TFLOP/s/device: 91.671, Tokens/s/device: 13817.578, total_weights: 65536, loss: 8.892 I0422 19:11:44.499019 132329347016512 metric_logger.py:185] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.958, Tokens/s/device: 6927.344, total_weights: 65536, loss: 8.592 I0422 19:11:44.646565 132329347016512 metric_logger.py:185] completed step: 7, seconds: 18.363, TFLOP/s/device: 0.740, Tokens/s/device: 111.530, total_weights: 65536, loss: 8.384 I0422 19:11:44.793881 132329347016512 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.405, Tokens/s/device: 13476.167, total_weights: 65536, loss: 8.255 I0422 19:11:44.941157 132329347016512 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0422 19:11:44.941803 132329347016512 checkpointing.py:776] Waited 0.0007081031799316406 seconds for step 9 to finish before starting checkpointing. I0422 19:11:44.943943 132329347016512 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 19:11:45.614644 2623 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0422 19:11:48.832777 132207364511488 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_02_synthetic/checkpoints/0/items/array_metadatas/process_6 I0422 19:12:19.668685 132206877996800 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 28.245 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 55.925217151641846 s) (per-host) I0422 19:12:19.668798 132206877996800 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 55.408362s. I0422 19:12:28.625208 132206877996800 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 64.364754s. I0422 19:12:28.625555 132202574636800 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 19:12:28.625691 132202574636800 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0422 19:12:28.625741 132202574636800 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0422 19:12:28.628299 132202574636800 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0422 19:12:28.628476 132329347016512 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 19:12:28.628622 132329347016512 checkpoint_manager.py:1458] Waiting for previous save to complete took 43.684683 seconds. If this number is high, consider checkpointing less frequently. I0422 19:12:28.630431 132329347016512 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 9 I0422 19:12:28.633857 132329347016512 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_02_synthetic/checkpoints/9. I0422 19:12:29.626224 132329347016512 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0422 19:12:29.626322 132329347016512 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 19:12:29.668606 132329347016512 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.043506s I0422 19:12:29.668777 132329347016512 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.181 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2977299690246582 s) (per-host) I0422 19:12:29.668830 132329347016512 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.297793s (batch_requests_ready=0.252409s, total_serialization_initiated=0.045314s, others=0.000070s) I0422 19:12:29.668919 132329347016512 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.301959s (all_items=0.000016s, per_item={'items': '0.00001645'}, temp_paths=0.301943) I0422 19:12:29.669738 132203598898944 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0422 19:12:29.669854 132329347016512 async_checkpointer.py:561] Finished blocking save. Time taken: 1.039349s. 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_02_synthetic/checkpoints/9. I0422 19:12:30.112717 132329347016512 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0422 19:12:30.113135 132202574636800 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0422 19:12:30.113307 132329347016512 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_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776885104.9439147, 'wait_for_prev_duration_secs': 43.684683084487915, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776885148.630471, 'checkpointer_blocking_duration_secs': 1.0394916534423828, 'get_old_steps_start_time': 1776885149.669981, 'get_old_steps_duration_secs': 2.5510787963867188e-05, 'checkpoint_manager_blocking_start_time': 1776885104.942027, 'checkpoint_manager_blocking_duration_secs': 45.17124271392822} I0422 19:12:30.113430 132329347016512 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0422 19:12:30.113478 132329347016512 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 19:12:35.192270 132203648915200 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_02_synthetic/checkpoints/9/items/array_metadatas/process_6 I0422 19:13:12.568245 132203598898944 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 36.567 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.19715690612793 s) (per-host) I0422 19:13:12.568376 132203598898944 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.898569s. I0422 19:13:20.287595 132203598898944 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 50.617773s. I0422 19:13:20.287893 132202574636800 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 19:13:20.288008 132202574636800 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0422 19:13:20.288057 132202574636800 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0422 19:13:20.289599 132202574636800 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0422 19:13:20.289800 132329347016512 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 19:13:20.289939 132329347016512 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 19:13:20.290738 132329347016512 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.135, Tokens/s/device: 13887.570, 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 19:13:33 UTC 2026 EXIT_CODE=0