XPK Start: Fri Apr 24 15:04:41 UTC 2026 2026-04-24 15:04:45.553885: 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 00:00:1777043085.567256 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1777043085.571017 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1777043085.582505 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777043085.582525 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777043085.582527 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777043085.582529 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-24 15:05:04.772794: 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 15:05:05.292093 138159297029952 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-24 15:05:14,332:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0424 15:05:14.332560 138159297029952 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-24 15:05:14,334:jax._src.distributed:157: Connecting to JAX distributed service on mt-03-dropout-j67si-slice-job-0-0.mt-03-dropout-j67si:8482 I0424 15:05:14.334961 138159297029952 distributed.py:157] Connecting to JAX distributed service on mt-03-dropout-j67si-slice-job-0-0.mt-03-dropout-j67si:8482 I0424 15:06:40.346294 138159297029952 max_utils.py:284] Jax distributed system initialized! I0424 15:06:47.429237 138159297029952 max_utils.py:800] System Information: Jax Version: 0.8.1 I0424 15:06:47.429345 138159297029952 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0424 15:06:47.429386 138159297029952 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 15:06:47.429423 138159297029952 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0424 15:06:48.050544 138159297029952 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0424 15:06:48.051152 138159297029952 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0424 15:06:48.051338 138159297029952 checkpointing.py:688] Setting up checkpoint logger... I0424 15:06:48.051389 138159297029952 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0424 15:06:48.051432 138159297029952 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0424 15:06:48.051790 138159297029952 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 0x7da73374cf50>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0424 15:06:51.022559 138159297029952 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0424 15:06:51.022907 138159297029952 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 0x7d92943b64e0>}, handler_registry=None I0424 15:06:51.023164 138159297029952 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d92943b64e0>` 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 15:06:51.023214 138159297029952 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7da5f2a03380>` 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 15:06:51.023251 138159297029952 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 0x7d92943b64e0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d92943b64e0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7da5f2a03380>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7da5f2a03380>}). I0424 15:06:51.023745 138159297029952 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0424 15:06:51.023826 138159297029952 async_checkpointer.py:177] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d92943144a0> timeout: 600 secs and primary_host=0 for async checkpoint writes I0424 15:06:51.969140 138159297029952 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260424_145744/linen_xpk_feat_nnx_set_defaults_true_20260424_145744_03_dropout/checkpoints I0424 15:06:51.971651 138159297029952 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_nnx_set_defaults_true_20260424_145744/linen_xpk_feat_nnx_set_defaults_true_20260424_145744_03_dropout/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7da5f32eb5f0> I0424 15:06:51.971763 138159297029952 checkpointing.py:302] Checkpoint manager created! I0424 15:06:52.290489 138159297029952 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0424 15:06:52.290593 138159297029952 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0424 15:06:52.661646 138159297029952 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0424 15:06:52.661747 138159297029952 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0424 15:06:52.677700 138159297029952 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0424 15:06:52.677769 138159297029952 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None). I0424 15:06:52.700648 138159297029952 attentions.py:1150] attentions/query Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 15:06:52.700719 138159297029952 attentions.py:1150] attentions/query Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0424 15:06:52.716961 138159297029952 attentions.py:1151] attentions/key Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 15:06:52.717025 138159297029952 attentions.py:1151] attentions/key Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0424 15:06:52.732900 138159297029952 attentions.py:1152] attentions/value Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0424 15:06:52.732973 138159297029952 attentions.py:1152] attentions/value Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0424 15:06:52.756957 138159297029952 attentions.py:1193] attentions/out Logical: bfloat16[32,128,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0424 15:06:52.757027 138159297029952 attentions.py:1193] attentions/out Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None). I0424 15:06:52.782985 138159297029952 linears.py:541] linears/x Logical: bfloat16[32,128,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0424 15:06:52.783057 138159297029952 linears.py:541] linears/x Physical: bfloat16[32,128,7168]....................................... ('fsdp', None, None). I0424 15:06:52.988758 138159297029952 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0424 15:06:52.988871 138159297029952 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0424 15:06:54.468679 138159297029952 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: (None,) I0424 15:06:54.468817 138159297029952 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0424 15:06:54.468873 138159297029952 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: PartitionSpec('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0424 15:06:54.468955 138159297029952 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: PartitionSpec('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0424 15:06:54.469017 138159297029952 maxtext_utils.py:1740] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0424 15:06:54.469060 138159297029952 maxtext_utils.py:1740] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: (None, None) I0424 15:06:54.469114 138159297029952 maxtext_utils.py:1740] 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 15:06:54.469166 138159297029952 maxtext_utils.py:1740] 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 15:06:54.469207 138159297029952 maxtext_utils.py:1740] 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 15:06:54.469245 138159297029952 maxtext_utils.py:1740] 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 15:06:54.469295 138159297029952 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', None) I0424 15:06:54.469343 138159297029952 maxtext_utils.py:1740] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: (None, 'fsdp') I0424 15:06:55.002312 138159297029952 train.py:158] train/xent Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0424 15:06:55.002406 138159297029952 train.py:158] train/xent Physical: float32[32,128]............................................. ('fsdp', None). I0424 15:06:55.017364 138159297029952 train.py:165] train/z_loss Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0424 15:06:55.017424 138159297029952 train.py:165] train/z_loss Physical: float32[32,128]............................................. ('fsdp', None). I0424 15:06:58.346425 138159297029952 max_utils.py:791] Total memory size: 0.8 GB, Output size: 0.4 GB, Temp size: 0.4 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0424 15:06:58.347299 138159297029952 metric_logger.py:289] number parameters: 1.104 billion I0424 15:07:02.287696 138159297029952 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0424 15:07:02.393697 138159297029952 checkpointing.py:798] Waited 0.10598635673522949 seconds for step 0 to finish before starting checkpointing. I0424 15:07:02.396147 138159297029952 checkpoint_manager.py:2013] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0424 15:07:02.398054 138159297029952 checkpoint_manager.py:1518] [process=3] Saving checkpoint at step 0 I0424 15:07:02.400767 138159297029952 async_checkpointer.py:452] [process=3] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260424_145744/linen_xpk_feat_nnx_set_defaults_true_20260424_145744_03_dropout/checkpoints/0. I0424 15:07:03.017327 138159297029952 signaling_client.py:364] Using JaxDistributedSignalingClient I0424 15:07:03.019750 138159297029952 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0424 15:07:03.019811 138159297029952 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 15:07:03.269043 138159297029952 base_pytree_checkpoint_handler.py:153] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.251795s I0424 15:07:03.269217 138159297029952 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.855 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5403604507446289 s) (per-host) I0424 15:07:03.269270 138159297029952 base_pytree_checkpoint_handler.py:737] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.540424s (batch_requests_ready=0.285059s, total_serialization_initiated=0.255295s, others=0.000070s) I0424 15:07:03.269362 138159297029952 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.544473s (all_items=0.000018s, per_item={'items': '0.00001836'}, temp_paths=0.544455) I0424 15:07:03.270373 138028974601984 async_checkpointer.py:79] [process=3][thread=async_save] Background save thread started. I0424 15:07:03.270502 138159297029952 async_checkpointer.py:561] Finished blocking save. Time taken: 0.872377s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260424_145744/linen_xpk_feat_nnx_set_defaults_true_20260424_145744_03_dropout/checkpoints/0. I0424 15:07:03.361606 138159297029952 checkpoint_manager.py:1566] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0424 15:07:03.362014 138031071729408 async_checkpointer.py:265] [process=3][thread=save_finalize] Waiting for background save thread=async_save. I0424 15:07:03.362183 138159297029952 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260424_145744/linen_xpk_feat_nnx_set_defaults_true_20260424_145744_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777043222.3961298, 'wait_for_prev_duration_secs': 6.628036499023438e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777043222.398093, 'checkpointer_blocking_duration_secs': 0.872596025466919, 'get_old_steps_start_time': 1777043223.27071, 'get_old_steps_duration_secs': 2.5272369384765625e-05, 'checkpoint_manager_blocking_start_time': 1777043222.3942788, 'checkpoint_manager_blocking_duration_secs': 0.9678628444671631} I0424 15:07:03.362289 138159297029952 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0424 15:07:03.362342 138159297029952 max_utils.py:750] Memstats: After params initialized: I0424 15:07:03.362399 138159297029952 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_10(process=3,(2,2,0,0)) I0424 15:07:03.362434 138159297029952 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_11(process=3,(3,2,0,0)) I0424 15:07:03.362462 138159297029952 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_14(process=3,(2,3,0,0)) I0424 15:07:03.362488 138159297029952 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_15(process=3,(3,3,0,0)) I0424 15:07:03.630409 138159297029952 metric_logger.py:185] completed step: 0, seconds: 3.940, TFLOP/s/device: 0.203, Tokens/s/device: 32.485, total_weights: 4096, loss: 10.889 I0424 15:07:03.729377 138159297029952 metric_logger.py:185] completed step: 1, seconds: 1.341, TFLOP/s/device: 0.597, Tokens/s/device: 95.419, total_weights: 4096, loss: 10.902 I0424 15:07:04.154680 138159297029952 metric_logger.py:185] completed step: 2, seconds: 0.029, TFLOP/s/device: 27.745, Tokens/s/device: 4434.283, total_weights: 4096, loss: 9.956 I0424 15:07:04.226433 138159297029952 metric_logger.py:185] completed step: 3, seconds: 0.426, TFLOP/s/device: 1.880, Tokens/s/device: 300.483, total_weights: 4096, loss: 9.139 I0424 15:07:04.369213 138159297029952 metric_logger.py:185] completed step: 4, seconds: 0.075, TFLOP/s/device: 10.633, Tokens/s/device: 1699.461, total_weights: 4096, loss: 8.458 I0424 15:07:04.375135 138159297029952 metric_logger.py:185] completed step: 5, seconds: 0.072, TFLOP/s/device: 11.152, Tokens/s/device: 1782.333, total_weights: 4096, loss: 7.924 I0424 15:07:13.167492 138159297029952 metric_logger.py:185] completed step: 6, seconds: 0.143, TFLOP/s/device: 5.591, Tokens/s/device: 893.636, total_weights: 4096, loss: 7.553 I0424 15:07:13.239155 138159297029952 metric_logger.py:185] completed step: 7, seconds: 8.723, TFLOP/s/device: 0.092, Tokens/s/device: 14.674, total_weights: 4096, loss: 7.267 I0424 15:07:13.309889 138159297029952 metric_logger.py:185] completed step: 8, seconds: 0.075, TFLOP/s/device: 10.674, Tokens/s/device: 1706.030, total_weights: 4096, loss: 7.111 I0424 15:07:13.381323 138159297029952 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0424 15:07:13.381888 138159297029952 checkpointing.py:798] Waited 0.0005824565887451172 seconds for step 9 to finish before starting checkpointing. I0424 15:07:13.383852 138159297029952 checkpoint_manager.py:2024] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0424 15:07:15.019575 2697 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0424 15:07:16.461880 138031088514816 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260424_145744/linen_xpk_feat_nnx_set_defaults_true_20260424_145744_03_dropout/checkpoints/0/items/array_metadatas/process_3 I0424 15:07:44.336295 138028974601984 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/gbytes_per_sec: 37.964 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.607402324676514 s) (per-host) I0424 15:07:44.336404 138028974601984 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.065817s. I0424 15:07:51.392541 138028974601984 async_checkpointer.py:144] [process=3][thread=async_save] Background save thread done. Time taken: 48.121939s. I0424 15:07:51.392826 138031071729408 async_checkpointer.py:273] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save. I0424 15:07:51.392963 138031071729408 async_checkpointer.py:283] [process=3][thread=save_finalize] No errors found in background save thread=async_save. I0424 15:07:51.393016 138031071729408 checkpoint_manager.py:2133] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0424 15:07:51.394620 138031071729408 checkpoint_manager.py:2142] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0424 15:07:51.394793 138159297029952 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. W0424 15:07:51.394943 138159297029952 checkpoint_manager.py:1458] Waiting for previous save to complete took 38.011083 seconds. If this number is high, consider checkpointing less frequently. I0424 15:07:51.396861 138159297029952 checkpoint_manager.py:1518] [process=3] Saving checkpoint at step 9 I0424 15:07:51.400390 138159297029952 async_checkpointer.py:452] [process=3] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260424_145744/linen_xpk_feat_nnx_set_defaults_true_20260424_145744_03_dropout/checkpoints/9. I0424 15:07:51.937186 138159297029952 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0424 15:07:51.937284 138159297029952 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 15:07:51.980327 138159297029952 base_pytree_checkpoint_handler.py:153] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.044101s I0424 15:07:51.980494 138159297029952 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.215 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.29580116271972656 s) (per-host) I0424 15:07:51.980547 138159297029952 base_pytree_checkpoint_handler.py:737] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.295863s (batch_requests_ready=0.249790s, total_serialization_initiated=0.046003s, others=0.000070s) I0424 15:07:51.980641 138159297029952 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.299809s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=0.299793) I0424 15:07:51.981550 138031088514816 async_checkpointer.py:79] [process=3][thread=async_save] Background save thread started. I0424 15:07:51.981714 138159297029952 async_checkpointer.py:561] Finished blocking save. Time taken: 0.584781s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260424_145744/linen_xpk_feat_nnx_set_defaults_true_20260424_145744_03_dropout/checkpoints/9. I0424 15:07:52.000663 138159297029952 checkpoint_manager.py:1566] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0424 15:07:52.000973 138031071729408 async_checkpointer.py:265] [process=3][thread=save_finalize] Waiting for background save thread=async_save. I0424 15:07:52.001159 138159297029952 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260424_145744/linen_xpk_feat_nnx_set_defaults_true_20260424_145744_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777043233.3838217, 'wait_for_prev_duration_secs': 38.011083126068115, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777043271.3968997, 'checkpointer_blocking_duration_secs': 0.584925651550293, 'get_old_steps_start_time': 1777043271.9818451, 'get_old_steps_duration_secs': 2.4318695068359375e-05, 'checkpoint_manager_blocking_start_time': 1777043233.3821208, 'checkpoint_manager_blocking_duration_secs': 38.61900329589844} I0424 15:07:52.001262 138159297029952 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0424 15:07:52.001309 138159297029952 checkpoint_manager.py:2024] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0424 15:07:56.764737 138007059306240 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260424_145744/linen_xpk_feat_nnx_set_defaults_true_20260424_145744_03_dropout/checkpoints/9/items/array_metadatas/process_3 I0424 15:08:32.359827 138031088514816 base_pytree_checkpoint_handler.py:129] [process=3] /jax/checkpoint/write/gbytes_per_sec: 38.835 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 40.675069093704224 s) (per-host) I0424 15:08:32.359980 138031088514816 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 40.378314s. I0424 15:08:41.090394 138031088514816 async_checkpointer.py:144] [process=3][thread=async_save] Background save thread done. Time taken: 49.108713s. I0424 15:08:41.090701 138031071729408 async_checkpointer.py:273] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save. I0424 15:08:41.090830 138031071729408 async_checkpointer.py:283] [process=3][thread=save_finalize] No errors found in background save thread=async_save. I0424 15:08:41.090882 138031071729408 checkpoint_manager.py:2133] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0424 15:08:41.092547 138031071729408 checkpoint_manager.py:2142] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0424 15:08:41.092717 138159297029952 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. I0424 15:08:41.092855 138159297029952 checkpoint_manager.py:2013] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0424 15:08:41.093491 138159297029952 metric_logger.py:185] completed step: 9, seconds: 0.071, TFLOP/s/device: 11.226, Tokens/s/device: 1794.250, total_weights: 4096, loss: 6.981 Per train step: Total TFLOPs: 0.80 split as 99.60% learnable weight flops and 0.40% attention flops XPK End: Fri Apr 24 15:08:51 UTC 2026 EXIT_CODE=0