XPK Start: Fri Apr 24 18:07:00 UTC 2026 2026-04-24 18:07:04.966876: 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 00:00:1777054024.980207 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1777054024.984045 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1777054024.995843 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777054024.995862 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777054024.995865 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777054024.995867 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-24 18:07:24.106403: 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) I0424 18:07:24.630344 138265942353728 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-24 18:07:33,671:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0424 18:07:33.671461 138265942353728 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-24 18:07:33,673:jax._src.distributed:157: Connecting to JAX distributed service on mt-04-int8-9wdfa-slice-job-0-0.mt-04-int8-9wdfa:8482 I0424 18:07:33.673709 138265942353728 distributed.py:157] Connecting to JAX distributed service on mt-04-int8-9wdfa-slice-job-0-0.mt-04-int8-9wdfa:8482 I0424 18:07:34.767501 138265942353728 max_utils.py:284] Jax distributed system initialized! I0424 18:07:40.824587 138265942353728 max_utils.py:800] System Information: Jax Version: 0.8.1 I0424 18:07:40.824690 138265942353728 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0424 18:07:40.824731 138265942353728 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 I0424 18:07:40.824765 138265942353728 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0424 18:07:41.439340 138265942353728 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0424 18:07:41.439734 138265942353728 checkpointing.py:677] Setting up checkpoint logger... I0424 18:07:41.439791 138265942353728 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0424 18:07:41.439835 138265942353728 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0424 18:07:41.440214 138265942353728 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 0x7dc012971190>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0424 18:07:44.492490 138265942353728 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0424 18:07:44.492722 138265942353728 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 0x7dac6c168500>}, handler_registry=None I0424 18:07:44.492963 138265942353728 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7dac6c168500>` 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`. I0424 18:07:44.493014 138265942353728 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7dadab9531a0>` 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`. I0424 18:07:44.493050 138265942353728 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 0x7dac6c168500>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7dac6c168500>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7dadab9531a0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7dadab9531a0>}). I0424 18:07:44.493608 138265942353728 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0424 18:07:44.493688 138265942353728 async_checkpointer.py:177] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7dac2c592840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0424 18:07:45.200484 138265942353728 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_04_int8/checkpoints I0424 18:07:45.225756 138265942353728 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_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_04_int8/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7dac6c16d610> I0424 18:07:45.225882 138265942353728 checkpointing.py:301] Checkpoint manager created! I0424 18:07:46.421607 138265942353728 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0424 18:07:46.421710 138265942353728 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0424 18:07:46.794240 138265942353728 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0424 18:07:46.794333 138265942353728 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0424 18:07:46.809987 138265942353728 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0424 18:07:46.810048 138265942353728 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0424 18:07:46.888073 138265942353728 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 18:07:46.888177 138265942353728 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:07:46.904232 138265942353728 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 18:07:46.904292 138265942353728 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:07:46.919939 138265942353728 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 18:07:46.920002 138265942353728 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:07:46.943884 138265942353728 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0424 18:07:46.943954 138265942353728 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0424 18:07:47.024310 138265942353728 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0424 18:07:47.024405 138265942353728 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0424 18:07:47.618177 138265942353728 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0424 18:07:47.618292 138265942353728 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0424 18:07:50.261538 138265942353728 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0424 18:07:50.261669 138265942353728 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) I0424 18:07:50.261722 138265942353728 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) I0424 18:07:50.261779 138265942353728 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') I0424 18:07:50.261832 138265942353728 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) I0424 18:07:50.261870 138265942353728 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) I0424 18:07:50.261922 138265942353728 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) I0424 18:07:50.261974 138265942353728 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') I0424 18:07:50.262013 138265942353728 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) I0424 18:07:50.262050 138265942353728 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) I0424 18:07:50.262114 138265942353728 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0424 18:07:50.262176 138265942353728 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0424 18:07:51.956456 138265942353728 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0424 18:07:51.956564 138265942353728 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0424 18:07:51.971430 138265942353728 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0424 18:07:51.971490 138265942353728 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0424 18:08:05.960605 138265942353728 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. I0424 18:08:05.961548 138265942353728 metric_logger.py:289] number parameters: 1.104 billion I0424 18:08:21.125377 138265942353728 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0424 18:08:21.367344 138265942353728 checkpointing.py:776] Waited 0.24195170402526855 seconds for step 0 to finish before starting checkpointing. I0424 18:08:21.369870 138265942353728 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0424 18:08:21.371610 138265942353728 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 0 I0424 18:08:21.374411 138265942353728 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_04_int8/checkpoints/0. I0424 18:08:21.969667 138265942353728 signaling_client.py:364] Using JaxDistributedSignalingClient I0424 18:08:21.970717 138265942353728 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0424 18:08:21.970776 138265942353728 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 I0424 18:08:22.225408 138265942353728 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.255820s I0424 18:08:22.225578 138265942353728 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.979 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5177552700042725 s) (per-host) I0424 18:08:22.225629 138265942353728 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.517816s (batch_requests_ready=0.258381s, total_serialization_initiated=0.259367s, others=0.000069s) I0424 18:08:22.225717 138265942353728 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.521885s (all_items=0.000018s, per_item={'items': '0.00001836'}, temp_paths=0.521867) I0424 18:08:22.226666 138138907313920 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0424 18:08:22.226793 138265942353728 async_checkpointer.py:561] Finished blocking save. Time taken: 0.855114s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_04_int8/checkpoints/0. I0424 18:08:22.275140 138265942353728 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0424 18:08:22.275465 138136784975616 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0424 18:08:22.275631 138265942353728 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_04_int8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777054101.369852, 'wait_for_prev_duration_secs': 6.699562072753906e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777054101.3716488, 'checkpointer_blocking_duration_secs': 0.8553371429443359, 'get_old_steps_start_time': 1777054102.2270062, 'get_old_steps_duration_secs': 2.4557113647460938e-05, 'checkpoint_manager_blocking_start_time': 1777054101.3679926, 'checkpoint_manager_blocking_duration_secs': 0.9075992107391357} I0424 18:08:22.275739 138265942353728 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0424 18:08:22.275790 138265942353728 max_utils.py:750] Memstats: After params initialized: I0424 18:08:22.275843 138265942353728 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0)) I0424 18:08:22.275876 138265942353728 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0)) I0424 18:08:22.275904 138265942353728 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0)) I0424 18:08:22.275928 138265942353728 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0)) I0424 18:08:22.552719 138265942353728 metric_logger.py:185] completed step: 0, seconds: 15.164, TFLOP/s/device: 0.896, Tokens/s/device: 135.059, total_weights: 65536, loss: 10.874 I0424 18:08:22.796073 138265942353728 metric_logger.py:185] completed step: 1, seconds: 1.426, TFLOP/s/device: 9.528, Tokens/s/device: 1436.238, total_weights: 65536, loss: 10.874 I0424 18:08:23.220417 138265942353728 metric_logger.py:185] completed step: 2, seconds: 0.014, TFLOP/s/device: 948.027, Tokens/s/device: 142897.014, total_weights: 65536, loss: 10.262 I0424 18:08:23.450158 138265942353728 metric_logger.py:185] completed step: 3, seconds: 0.425, TFLOP/s/device: 31.959, Tokens/s/device: 4817.191, total_weights: 65536, loss: 9.731 I0424 18:08:23.909605 138265942353728 metric_logger.py:185] completed step: 4, seconds: 0.234, TFLOP/s/device: 58.043, Tokens/s/device: 8748.809, total_weights: 65536, loss: 9.272 I0424 18:08:23.915202 138265942353728 metric_logger.py:185] completed step: 5, seconds: 0.230, TFLOP/s/device: 59.138, Tokens/s/device: 8913.921, total_weights: 65536, loss: 8.887 I0424 18:08:46.669862 138265942353728 metric_logger.py:185] completed step: 6, seconds: 0.460, TFLOP/s/device: 29.549, Tokens/s/device: 4453.946, total_weights: 65536, loss: 8.588 I0424 18:08:46.899497 138265942353728 metric_logger.py:185] completed step: 7, seconds: 22.526, TFLOP/s/device: 0.603, Tokens/s/device: 90.916, total_weights: 65536, loss: 8.380 I0424 18:08:47.128954 138265942353728 metric_logger.py:185] completed step: 8, seconds: 0.234, TFLOP/s/device: 58.132, Tokens/s/device: 8762.360, total_weights: 65536, loss: 8.251 I0424 18:08:47.358019 138265942353728 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0424 18:08:47.358745 138265942353728 checkpointing.py:776] Waited 0.000743865966796875 seconds for step 9 to finish before starting checkpointing. I0424 18:08:47.360764 138265942353728 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0424 18:08:47.457560 2703 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0424 18:08:49.409229 138138915706624 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_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_04_int8/checkpoints/0/items/array_metadatas/process_6 I0424 18:09:17.224009 138138907313920 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 28.453 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 55.516146421432495 s) (per-host) I0424 18:09:17.224167 138138907313920 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 54.997293s. I0424 18:09:26.037172 138138907313920 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 63.810284s. I0424 18:09:26.037513 138136784975616 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0424 18:09:26.037631 138136784975616 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0424 18:09:26.037680 138136784975616 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0424 18:09:26.039812 138136784975616 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0424 18:09:26.039984 138265942353728 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. W0424 18:09:26.040130 138265942353728 checkpoint_manager.py:1458] Waiting for previous save to complete took 38.679353 seconds. If this number is high, consider checkpointing less frequently. I0424 18:09:26.041946 138265942353728 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 9 I0424 18:09:26.045332 138265942353728 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_04_int8/checkpoints/9. I0424 18:09:27.388360 138265942353728 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0424 18:09:27.388459 138265942353728 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 I0424 18:09:27.432547 138265942353728 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.045387s I0424 18:09:27.432725 138265942353728 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.843 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.8368504047393799 s) (per-host) I0424 18:09:27.432776 138265942353728 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.836912s (batch_requests_ready=0.789415s, total_serialization_initiated=0.047428s, others=0.000069s) I0424 18:09:27.432868 138265942353728 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.841084s (all_items=0.000024s, per_item={'items': '0.00002360'}, temp_paths=0.841060) I0424 18:09:27.433749 138140031379200 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started. I0424 18:09:27.433908 138265942353728 async_checkpointer.py:561] Finished blocking save. Time taken: 1.391891s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_04_int8/checkpoints/9. I0424 18:09:27.435755 138265942353728 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0424 18:09:27.436024 138136784975616 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0424 18:09:27.436162 138265942353728 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_04_int8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777054127.3607368, 'wait_for_prev_duration_secs': 38.67935252189636, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777054166.0419848, 'checkpointer_blocking_duration_secs': 1.3920326232910156, 'get_old_steps_start_time': 1777054167.4340365, 'get_old_steps_duration_secs': 2.6464462280273438e-05, 'checkpoint_manager_blocking_start_time': 1777054127.3589902, 'checkpoint_manager_blocking_duration_secs': 40.07714056968689} I0424 18:09:27.436322 138265942353728 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0424 18:09:27.436369 138265942353728 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0424 18:09:31.697634 138140039771904 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_20260424_180038/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260424_180038_04_int8/checkpoints/9/items/array_metadatas/process_6 I0424 18:10:09.318164 138140031379200 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 36.974 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.722233057022095 s) (per-host) I0424 18:10:09.318285 138140031379200 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.884423s. I0424 18:10:16.642437 138140031379200 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 49.208561s. I0424 18:10:16.642730 138136784975616 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0424 18:10:16.642841 138136784975616 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0424 18:10:16.642888 138136784975616 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0424 18:10:16.644423 138136784975616 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0424 18:10:16.644592 138265942353728 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. I0424 18:10:16.644743 138265942353728 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0424 18:10:16.645379 138265942353728 metric_logger.py:185] completed step: 9, seconds: 0.230, TFLOP/s/device: 59.147, Tokens/s/device: 8915.318, 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: Fri Apr 24 18:10:26 UTC 2026 EXIT_CODE=0