XPK Start: Sun Apr 19 06:43:44 UTC 2026 2026-04-19 06:43:48.377720: 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 00:00:1776581028.389995 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1776581028.393762 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1776581028.404877 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776581028.404895 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776581028.404897 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1776581028.404899 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-19 06:44:07.520336: 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) I0419 06:44:08.031818 137083250112320 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-19 06:44:17,072:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0419 06:44:17.072333 137083250112320 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-19 06:44:17,074:jax._src.distributed:157: Connecting to JAX distributed service on mt-04-int8-v6r00-slice-job-0-0.mt-04-int8-v6r00:8482 I0419 06:44:17.074751 137083250112320 distributed.py:157] Connecting to JAX distributed service on mt-04-int8-v6r00-slice-job-0-0.mt-04-int8-v6r00:8482 I0419 06:44:18.618546 137083250112320 max_utils.py:284] Jax distributed system initialized! I0419 06:44:25.444319 137083250112320 max_utils.py:800] System Information: Jax Version: 0.8.1 I0419 06:44:25.444427 137083250112320 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0419 06:44:25.444466 137083250112320 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 I0419 06:44:25.444500 137083250112320 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0419 06:44:26.056610 137083250112320 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 06:44:26.057016 137083250112320 checkpointing.py:677] Setting up checkpoint logger... I0419 06:44:26.057074 137083250112320 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0419 06:44:26.057120 137083250112320 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0419 06:44:26.057483 137083250112320 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 0x7cacbcbcf620>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0419 06:44:31.175249 137083250112320 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0419 06:44:31.175492 137083250112320 checkpoint_manager.py:709] [process=3][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c99f1225c70>}, handler_registry=None I0419 06:44:31.175724 137083250112320 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c99f1225c70>` 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`. I0419 06:44:31.175775 137083250112320 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c99f0c93d70>` 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`. I0419 06:44:31.175819 137083250112320 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 0x7c99f1225c70>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c99f1225c70>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c99f0c93d70>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c99f0c93d70>}). I0419 06:44:31.176344 137083250112320 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0419 06:44:31.176425 137083250112320 async_checkpointer.py:177] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7c992acfa840> timeout: 600 secs and primary_host=0 for async checkpoint writes I0419 06:44:32.704993 137083250112320 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_04_int8/checkpoints I0419 06:44:33.569827 137083250112320 checkpoint_manager.py:929] [process=3][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_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_04_int8/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7c99f1223f50> I0419 06:44:33.569997 137083250112320 checkpointing.py:301] Checkpoint manager created! I0419 06:44:34.748158 137083250112320 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0419 06:44:34.748276 137083250112320 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 06:44:35.123316 137083250112320 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 06:44:35.123414 137083250112320 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 06:44:35.139277 137083250112320 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 06:44:35.139340 137083250112320 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 06:44:35.217528 137083250112320 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 06:44:35.217625 137083250112320 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 06:44:35.233510 137083250112320 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 06:44:35.233578 137083250112320 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 06:44:35.249327 137083250112320 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 06:44:35.249387 137083250112320 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 06:44:35.273240 137083250112320 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0419 06:44:35.273309 137083250112320 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 06:44:35.353965 137083250112320 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0419 06:44:35.354059 137083250112320 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0419 06:44:35.960376 137083250112320 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0419 06:44:35.960495 137083250112320 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0419 06:44:38.622294 137083250112320 maxtext_utils.py:1620] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0419 06:44:38.622432 137083250112320 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) I0419 06:44:38.622488 137083250112320 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) I0419 06:44:38.622550 137083250112320 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') I0419 06:44:38.622604 137083250112320 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) I0419 06:44:38.622646 137083250112320 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) I0419 06:44:38.622701 137083250112320 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) I0419 06:44:38.622755 137083250112320 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') I0419 06:44:38.622798 137083250112320 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) I0419 06:44:38.622836 137083250112320 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) I0419 06:44:38.622884 137083250112320 maxtext_utils.py:1620] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0419 06:44:38.622933 137083250112320 maxtext_utils.py:1620] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0419 06:44:40.321912 137083250112320 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 06:44:40.322015 137083250112320 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0419 06:44:40.336808 137083250112320 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 06:44:40.336878 137083250112320 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0419 06:44:54.140514 137083250112320 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. I0419 06:44:54.141486 137083250112320 metric_logger.py:289] number parameters: 1.104 billion I0419 06:45:08.861340 137083250112320 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0419 06:45:09.100759 137083250112320 checkpointing.py:776] Waited 0.23939967155456543 seconds for step 0 to finish before starting checkpointing. I0419 06:45:09.103435 137083250112320 checkpoint_manager.py:2013] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 06:45:09.105305 137083250112320 checkpoint_manager.py:1518] [process=3] Saving checkpoint at step 0 I0419 06:45:09.107919 137083250112320 async_checkpointer.py:452] [process=3] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_04_int8/checkpoints/0. I0419 06:45:10.137669 137083250112320 signaling_client.py:364] Using JaxDistributedSignalingClient I0419 06:45:10.138699 137083250112320 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 06:45:10.138759 137083250112320 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 I0419 06:45:10.394604 137083250112320 base_pytree_checkpoint_handler.py:153] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.257015s I0419 06:45:10.394778 137083250112320 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.850 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5412642955780029 s) (per-host) I0419 06:45:10.394829 137083250112320 base_pytree_checkpoint_handler.py:737] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.541324s (batch_requests_ready=0.280643s, total_serialization_initiated=0.260612s, others=0.000069s) I0419 06:45:10.394918 137083250112320 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.545469s (all_items=0.000020s, per_item={'items': '0.00002003'}, temp_paths=0.545449) I0419 06:45:10.395921 136952432375552 async_checkpointer.py:79] [process=3][thread=async_save] Background save thread started. I0419 06:45:10.396093 137083250112320 async_checkpointer.py:561] Finished blocking save. Time taken: 1.290716s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_04_int8/checkpoints/0. I0419 06:45:10.838393 137083250112320 checkpoint_manager.py:1566] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0419 06:45:10.838780 136950828738304 async_checkpointer.py:265] [process=3][thread=save_finalize] Waiting for background save thread=async_save. I0419 06:45:10.838946 137083250112320 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_04_int8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776581109.1034188, 'wait_for_prev_duration_secs': 6.270408630371094e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776581109.1053448, 'checkpointer_blocking_duration_secs': 1.2908635139465332, 'get_old_steps_start_time': 1776581110.3962307, 'get_old_steps_duration_secs': 4.124641418457031e-05, 'checkpoint_manager_blocking_start_time': 1776581109.1014364, 'checkpoint_manager_blocking_duration_secs': 1.7374680042266846} I0419 06:45:10.839058 137083250112320 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0419 06:45:10.839110 137083250112320 max_utils.py:750] Memstats: After params initialized: I0419 06:45:10.839173 137083250112320 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_10(process=3,(2,2,0,0)) I0419 06:45:10.839208 137083250112320 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_11(process=3,(3,2,0,0)) I0419 06:45:10.839249 137083250112320 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_14(process=3,(2,3,0,0)) I0419 06:45:10.839276 137083250112320 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_15(process=3,(3,3,0,0)) I0419 06:45:11.104613 137083250112320 metric_logger.py:185] completed step: 0, seconds: 14.720, TFLOP/s/device: 0.923, Tokens/s/device: 139.133, total_weights: 65536, loss: 10.874 I0419 06:45:11.376404 137083250112320 metric_logger.py:185] completed step: 1, seconds: 2.242, TFLOP/s/device: 6.061, Tokens/s/device: 913.513, total_weights: 65536, loss: 10.874 I0419 06:45:11.784674 137083250112320 metric_logger.py:185] completed step: 2, seconds: 0.043, TFLOP/s/device: 316.820, Tokens/s/device: 47754.512, total_weights: 65536, loss: 10.262 I0419 06:45:12.014807 137083250112320 metric_logger.py:185] completed step: 3, seconds: 0.409, TFLOP/s/device: 33.241, Tokens/s/device: 5010.459, total_weights: 65536, loss: 9.731 I0419 06:45:12.474736 137083250112320 metric_logger.py:185] completed step: 4, seconds: 0.235, TFLOP/s/device: 57.926, Tokens/s/device: 8731.279, total_weights: 65536, loss: 9.272 I0419 06:45:12.480750 137083250112320 metric_logger.py:185] completed step: 5, seconds: 0.230, TFLOP/s/device: 59.046, Tokens/s/device: 8900.053, total_weights: 65536, loss: 8.887 I0419 06:45:37.274292 137083250112320 metric_logger.py:185] completed step: 6, seconds: 0.460, TFLOP/s/device: 29.509, Tokens/s/device: 4447.910, total_weights: 65536, loss: 8.588 I0419 06:45:37.504032 137083250112320 metric_logger.py:185] completed step: 7, seconds: 24.565, TFLOP/s/device: 0.553, Tokens/s/device: 83.369, total_weights: 65536, loss: 8.380 I0419 06:45:37.733748 137083250112320 metric_logger.py:185] completed step: 8, seconds: 0.234, TFLOP/s/device: 58.109, Tokens/s/device: 8758.762, total_weights: 65536, loss: 8.251 I0419 06:45:37.965575 137083250112320 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0419 06:45:37.966263 137083250112320 checkpointing.py:776] Waited 0.0006885528564453125 seconds for step 9 to finish before starting checkpointing. I0419 06:45:37.968211 137083250112320 checkpoint_manager.py:2024] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 06:45:39.568861 2819 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0419 06:45:43.823568 136961037678336 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_04_int8/checkpoints/0/items/array_metadatas/process_3 I0419 06:46:10.345736 136952432375552 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/gbytes_per_sec: 26.112 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 60.49217987060547 s) (per-host) I0419 06:46:10.345859 136952432375552 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 59.949826s. I0419 06:46:17.233537 136952432375552 async_checkpointer.py:144] [process=3][thread=async_save] Background save thread done. Time taken: 66.837488s. I0419 06:46:17.233835 136950828738304 async_checkpointer.py:273] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 06:46:17.233958 136950828738304 async_checkpointer.py:283] [process=3][thread=save_finalize] No errors found in background save thread=async_save. I0419 06:46:17.234007 136950828738304 checkpoint_manager.py:2133] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0419 06:46:17.235641 136950828738304 checkpoint_manager.py:2142] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0419 06:46:17.235840 137083250112320 checkpoint_manager.py:2036] [process=3][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0419 06:46:17.235970 137083250112320 checkpoint_manager.py:1458] Waiting for previous save to complete took 39.267766 seconds. If this number is high, consider checkpointing less frequently. I0419 06:46:17.237712 137083250112320 checkpoint_manager.py:1518] [process=3] Saving checkpoint at step 9 I0419 06:46:17.241151 137083250112320 async_checkpointer.py:452] [process=3] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_04_int8/checkpoints/9. I0419 06:46:19.165562 137083250112320 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0419 06:46:19.165663 137083250112320 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 I0419 06:46:19.211205 137083250112320 base_pytree_checkpoint_handler.py:153] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.046774s I0419 06:46:19.211388 137083250112320 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.214 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 1.2709074020385742 s) (per-host) I0419 06:46:19.211440 137083250112320 base_pytree_checkpoint_handler.py:737] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 1.270968s (batch_requests_ready=1.222050s, total_serialization_initiated=0.048850s, others=0.000068s) I0419 06:46:19.211533 137083250112320 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 1.275166s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=1.275151) I0419 06:46:19.212465 136954091927296 async_checkpointer.py:79] [process=3][thread=async_save] Background save thread started. I0419 06:46:19.212624 137083250112320 async_checkpointer.py:561] Finished blocking save. Time taken: 1.974839s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_04_int8/checkpoints/9. I0419 06:46:19.631301 137083250112320 checkpoint_manager.py:1566] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0419 06:46:19.631690 136950828738304 async_checkpointer.py:265] [process=3][thread=save_finalize] Waiting for background save thread=async_save. I0419 06:46:19.631881 137083250112320 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_04_int8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776581137.9681804, 'wait_for_prev_duration_secs': 39.267765522003174, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776581177.2377517, 'checkpointer_blocking_duration_secs': 1.9749796390533447, 'get_old_steps_start_time': 1776581179.2127516, 'get_old_steps_duration_secs': 2.6464462280273438e-05, 'checkpoint_manager_blocking_start_time': 1776581137.9664888, 'checkpoint_manager_blocking_duration_secs': 41.665342807769775} I0419 06:46:19.632003 137083250112320 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0419 06:46:19.632069 137083250112320 checkpoint_manager.py:2024] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 06:46:26.806598 136961037678336 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_04_int8/checkpoints/9/items/array_metadatas/process_3 I0419 06:47:02.979775 136954091927296 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/gbytes_per_sec: 35.072 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 45.0392529964447 s) (per-host) I0419 06:47:02.979901 136954091927296 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.767326s. I0419 06:47:09.736656 136954091927296 async_checkpointer.py:144] [process=3][thread=async_save] Background save thread done. Time taken: 50.524067s. I0419 06:47:09.736936 136950828738304 async_checkpointer.py:273] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 06:47:09.737084 136950828738304 async_checkpointer.py:283] [process=3][thread=save_finalize] No errors found in background save thread=async_save. I0419 06:47:09.737136 136950828738304 checkpoint_manager.py:2133] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0419 06:47:09.738755 136950828738304 checkpoint_manager.py:2142] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0419 06:47:09.738944 137083250112320 checkpoint_manager.py:2036] [process=3][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0419 06:47:09.739080 137083250112320 checkpoint_manager.py:2013] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 06:47:09.739681 137083250112320 metric_logger.py:185] completed step: 9, seconds: 0.230, TFLOP/s/device: 59.146, Tokens/s/device: 8915.085, 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: Sun Apr 19 06:47:20 UTC 2026 EXIT_CODE=0