XPK Start: Sat Apr 25 15:34:11 UTC 2026 2026-04-25 15:34:15.297113: 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 00:00:1777131255.309952 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1777131255.313875 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered W0000 00:00:1777131255.325201 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777131255.325223 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777131255.325225 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. W0000 00:00:1777131255.325227 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once. 2026-04-25 15:34:34.491082: 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) I0425 15:34:35.021082 132584161597248 max_utils.py:273] Attempting to initialize the jax distributed system... INFO:2026-04-25 15:34:44,061:jax._src.distributed:140: Starting JAX distributed service on [::]:8482 I0425 15:34:44.061827 132584161597248 distributed.py:140] Starting JAX distributed service on [::]:8482 INFO:2026-04-25 15:34:44,064:jax._src.distributed:157: Connecting to JAX distributed service on mt-09-pdb-lt-1-56gwa-slice-job-0-0.mt-09-pdb-lt-1-56gwa:8482 I0425 15:34:44.064066 132584161597248 distributed.py:157] Connecting to JAX distributed service on mt-09-pdb-lt-1-56gwa-slice-job-0-0.mt-09-pdb-lt-1-56gwa:8482 I0425 15:34:45.284380 132584161597248 max_utils.py:284] Jax distributed system initialized! I0425 15:34:52.362979 132584161597248 max_utils.py:800] System Information: Jax Version: 0.8.1 I0425 15:34:52.363084 132584161597248 max_utils.py:801] System Information: Jaxlib Version: 0.8.1 I0425 15:34:52.363126 132584161597248 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 I0425 15:34:52.363159 132584161597248 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0425 15:34:52.990896 132584161597248 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 8, 1, 1, 1, 1, 4, 1, 1, 1, 1) I0425 15:34:52.991509 132584161597248 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 8, 1, 1, 1, 1, 4, 1, 1, 1, 1) I0425 15:34:52.991707 132584161597248 checkpointing.py:688] Setting up checkpoint logger... I0425 15:34:52.991758 132584161597248 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0425 15:34:52.991801 132584161597248 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None I0425 15:34:52.992147 132584161597248 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 0x789514f7d3a0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0425 15:34:55.859435 132584161597248 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0425 15:34:55.859827 132584161597248 checkpoint_manager.py:709] [process=5][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7880c04ed490>}, handler_registry=None I0425 15:34:55.860074 132584161597248 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7880c04ed490>` 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`. I0425 15:34:55.860124 132584161597248 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7880c0690860>` 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`. I0425 15:34:55.860161 132584161597248 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 0x7880c04ed490>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7880c04ed490>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7880c0690860>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7880c0690860>}). I0425 15:34:55.860703 132584161597248 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33 I0425 15:34:55.860798 132584161597248 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7880a831c180> timeout: 600 secs and primary_host=0 for async checkpoint writes I0425 15:34:56.974981 132584161597248 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_09_pdb_lt_1/checkpoints I0425 15:34:57.402003 132584161597248 checkpoint_manager.py:929] [process=5][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_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_09_pdb_lt_1/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7882ba9f1c70> I0425 15:34:57.402172 132584161597248 checkpointing.py:302] Checkpoint manager created! I0425 15:34:58.298560 132584161597248 nnx_wrappers.py:455] Unknown Logical: bfloat16[8,2048,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0425 15:34:58.298682 132584161597248 nnx_wrappers.py:455] Unknown Physical: bfloat16[8,2048,2048]....................................... ('fsdp', None, 'tensor'). I0425 15:34:58.668078 132584161597248 attentions.py:1084] attentions/inputs_q Logical: bfloat16[8,2048,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0425 15:34:58.668171 132584161597248 attentions.py:1084] attentions/inputs_q Physical: bfloat16[8,2048,2048]....................................... ('fsdp', None, 'tensor'). I0425 15:34:58.683773 132584161597248 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[8,2048,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0425 15:34:58.683833 132584161597248 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[8,2048,2048]....................................... ('fsdp', None, 'tensor'). I0425 15:34:58.706346 132584161597248 attentions.py:1150] attentions/query Logical: bfloat16[8,2048,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 15:34:58.706407 132584161597248 attentions.py:1150] attentions/query Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0425 15:34:58.722195 132584161597248 attentions.py:1151] attentions/key Logical: bfloat16[8,2048,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 15:34:58.722259 132584161597248 attentions.py:1151] attentions/key Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0425 15:34:58.737984 132584161597248 attentions.py:1152] attentions/value Logical: bfloat16[8,2048,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 15:34:58.738044 132584161597248 attentions.py:1152] attentions/value Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0425 15:34:58.763027 132584161597248 attentions.py:1193] attentions/out Logical: bfloat16[8,2048,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0425 15:34:58.763099 132584161597248 attentions.py:1193] attentions/out Physical: bfloat16[8,2048,16,128]..................................... ('fsdp', None, 'tensor', None). I0425 15:34:58.782665 132584161597248 linears.py:541] linears/x Logical: bfloat16[8,2048,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0425 15:34:58.782730 132584161597248 linears.py:541] linears/x Physical: bfloat16[8,2048,7168]....................................... ('fsdp', None, 'tensor'). I0425 15:34:58.971445 132584161597248 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0425 15:34:58.971553 132584161597248 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 8 tensor: 4 I0425 15:35:00.418945 132584161597248 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: PartitionSpec('norm',) Physical: ('tensor',) I0425 15:35:00.419080 132584161597248 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, 'tensor') I0425 15:35:00.419135 132584161597248 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, 'tensor') I0425 15:35:00.419193 132584161597248 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: PartitionSpec('mlp', 'layers', 'embed') Physical: ('tensor', None, 'fsdp') I0425 15:35:00.419245 132584161597248 maxtext_utils.py:1740] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: ('tensor', None) I0425 15:35:00.419289 132584161597248 maxtext_utils.py:1740] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: PartitionSpec('norm', 'layers') Physical: ('tensor', None) I0425 15:35:00.419343 132584161597248 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, 'tensor', None) I0425 15:35:00.419394 132584161597248 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: ('tensor', None, None, 'fsdp') I0425 15:35:00.419434 132584161597248 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, 'tensor', None) I0425 15:35:00.419471 132584161597248 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, 'tensor', None) I0425 15:35:00.419520 132584161597248 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: PartitionSpec('embed', 'vocab') Physical: ('fsdp', 'tensor') I0425 15:35:00.419568 132584161597248 maxtext_utils.py:1740] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: PartitionSpec('vocab', 'embed') Physical: ('tensor', 'fsdp') I0425 15:35:00.919051 132584161597248 train.py:158] train/xent Logical: float32[8,2048]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0425 15:35:00.919142 132584161597248 train.py:158] train/xent Physical: float32[8,2048]............................................. ('fsdp', None). I0425 15:35:00.934302 132584161597248 train.py:165] train/z_loss Logical: float32[8,2048]............................................. ('activation_embed_and_logits_batch', 'activation_length'). I0425 15:35:00.934362 132584161597248 train.py:165] train/z_loss Physical: float32[8,2048]............................................. ('fsdp', None). I0425 15:35:10.599879 132584161597248 max_utils.py:791] Total memory size: 0.9 GB, Output size: 0.4 GB, Temp size: 0.5 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0425 15:35:10.600712 132584161597248 metric_logger.py:289] number parameters: 1.104 billion I0425 15:35:12.115342 132584161597248 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0425 15:35:21.203890 132584161597248 checkpointing.py:798] Waited 9.088526964187622 seconds for step 0 to finish before starting checkpointing. I0425 15:35:21.206584 132584161597248 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0425 15:35:21.208303 132584161597248 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0 I0425 15:35:21.210929 132584161597248 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_09_pdb_lt_1/checkpoints/0. I0425 15:35:21.764630 132584161597248 signaling_client.py:364] Using JaxDistributedSignalingClient I0425 15:35:21.765677 132584161597248 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0425 15:35:21.765734 132584161597248 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 I0425 15:35:21.996489 132584161597248 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.231940s I0425 15:35:21.996680 132584161597248 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.076 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5014069080352783 s) (per-host) I0425 15:35:21.996739 132584161597248 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.501491s (batch_requests_ready=0.265907s, total_serialization_initiated=0.235493s, others=0.000091s) I0425 15:35:21.996830 132584161597248 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.505590s (all_items=0.000019s, per_item={'items': '0.00001860'}, temp_paths=0.505572) I0425 15:35:21.997755 132461338023680 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0425 15:35:21.997921 132584161597248 async_checkpointer.py:561] Finished blocking save. Time taken: 0.789550s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_09_pdb_lt_1/checkpoints/0. I0425 15:35:21.999754 132584161597248 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0425 15:35:22.000028 132458125190912 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0425 15:35:22.000163 132584161597248 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_09_pdb_lt_1/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777131321.2065663, 'wait_for_prev_duration_secs': 6.246566772460938e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777131321.208341, 'checkpointer_blocking_duration_secs': 0.7896850109100342, 'get_old_steps_start_time': 1777131321.9980445, 'get_old_steps_duration_secs': 2.5510787963867188e-05, 'checkpoint_manager_blocking_start_time': 1777131321.2043848, 'checkpoint_manager_blocking_duration_secs': 0.7957429885864258} I0425 15:35:22.000322 132584161597248 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0425 15:35:22.000373 132584161597248 max_utils.py:750] Memstats: After params initialized: I0425 15:35:22.000420 132584161597248 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0)) I0425 15:35:22.000455 132584161597248 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0)) I0425 15:35:22.000482 132584161597248 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0)) I0425 15:35:22.000507 132584161597248 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0)) I0425 15:35:22.264386 132584161597248 metric_logger.py:185] completed step: 0, seconds: 1.515, TFLOP/s/device: 2.243, Tokens/s/device: 338.061, total_weights: 16384, loss: 10.862 I0425 15:35:22.340054 132584161597248 metric_logger.py:185] completed step: 1, seconds: 10.148, TFLOP/s/device: 0.335, Tokens/s/device: 50.454, total_weights: 16384, loss: 10.862 I0425 15:35:22.758184 132584161597248 metric_logger.py:185] completed step: 2, seconds: 0.018, TFLOP/s/device: 193.131, Tokens/s/device: 29110.757, total_weights: 16384, loss: 9.763 I0425 15:35:22.816782 132584161597248 metric_logger.py:185] completed step: 3, seconds: 0.418, TFLOP/s/device: 8.119, Tokens/s/device: 1223.797, total_weights: 16384, loss: 8.823 I0425 15:35:22.934319 132584161597248 metric_logger.py:185] completed step: 4, seconds: 0.064, TFLOP/s/device: 53.202, Tokens/s/device: 8019.171, total_weights: 16384, loss: 7.947 I0425 15:35:22.939910 132584161597248 metric_logger.py:185] completed step: 5, seconds: 0.058, TFLOP/s/device: 58.584, Tokens/s/device: 8830.479, total_weights: 16384, loss: 7.212 I0425 15:35:35.392370 132584161597248 metric_logger.py:185] completed step: 6, seconds: 0.118, TFLOP/s/device: 28.753, Tokens/s/device: 4333.951, total_weights: 16384, loss: 6.652 I0425 15:35:35.450716 132584161597248 metric_logger.py:185] completed step: 7, seconds: 12.395, TFLOP/s/device: 0.274, Tokens/s/device: 41.308, total_weights: 16384, loss: 6.269 I0425 15:35:35.509397 132584161597248 metric_logger.py:185] completed step: 8, seconds: 0.063, TFLOP/s/device: 54.058, Tokens/s/device: 8148.195, total_weights: 16384, loss: 6.031 I0425 15:35:35.567581 132584161597248 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0425 15:35:35.568224 132584161597248 checkpointing.py:798] Waited 0.0006589889526367188 seconds for step 9 to finish before starting checkpointing. I0425 15:35:35.570091 132584161597248 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 15:35:37.373645 2475 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0425 15:35:39.310493 132461346416384 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_09_pdb_lt_1/checkpoints/0/items/array_metadatas/process_5 I0425 15:36:10.503501 132461338023680 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 32.231 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 49.00820302963257 s) (per-host) I0425 15:36:10.503616 132461338023680 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 48.505749s. I0425 15:36:18.234261 132461338023680 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 56.236377s. I0425 15:36:18.234532 132458125190912 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 15:36:18.234667 132458125190912 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0425 15:36:18.234718 132458125190912 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0425 15:36:18.236356 132458125190912 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0425 15:36:18.236540 132584161597248 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0425 15:36:18.236686 132584161597248 checkpoint_manager.py:1458] Waiting for previous save to complete took 42.666586 seconds. If this number is high, consider checkpointing less frequently. I0425 15:36:18.238604 132584161597248 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 9 I0425 15:36:18.242167 132584161597248 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_09_pdb_lt_1/checkpoints/9. I0425 15:36:18.775878 132584161597248 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array. I0425 15:36:18.775976 132584161597248 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 I0425 15:36:18.817283 132584161597248 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.042540s I0425 15:36:18.817449 132584161597248 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.083 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3034992218017578 s) (per-host) I0425 15:36:18.817501 132584161597248 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.303561s (batch_requests_ready=0.259067s, total_serialization_initiated=0.044424s, others=0.000069s) I0425 15:36:18.817591 132584161597248 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.307508s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=0.307493) I0425 15:36:18.818483 132461346416384 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started. I0425 15:36:18.818665 132584161597248 async_checkpointer.py:561] Finished blocking save. Time taken: 0.579974s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_09_pdb_lt_1/checkpoints/9. I0425 15:36:18.820609 132584161597248 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0425 15:36:18.820855 132458125190912 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save. I0425 15:36:18.820991 132584161597248 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_09_pdb_lt_1/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777131335.570061, 'wait_for_prev_duration_secs': 42.66658592224121, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777131378.238644, 'checkpointer_blocking_duration_secs': 0.5801324844360352, 'get_old_steps_start_time': 1777131378.8187954, 'get_old_steps_duration_secs': 2.5987625122070312e-05, 'checkpoint_manager_blocking_start_time': 1777131335.5684624, 'checkpoint_manager_blocking_duration_secs': 43.25249242782593} I0425 15:36:18.821100 132584161597248 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0425 15:36:18.821146 132584161597248 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 15:36:23.512916 132432523126528 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_09_pdb_lt_1/checkpoints/9/items/array_metadatas/process_5 I0425 15:36:59.735467 132461346416384 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 38.320 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.22147798538208 s) (per-host) I0425 15:36:59.735583 132461346416384 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 40.916986s. I0425 15:37:07.495001 132461346416384 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 48.676389s. I0425 15:37:07.495320 132458125190912 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 15:37:07.495436 132458125190912 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save. I0425 15:37:07.495486 132458125190912 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0425 15:37:07.496862 132458125190912 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0425 15:37:07.497038 132584161597248 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0425 15:37:07.497179 132584161597248 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0425 15:37:07.497797 132584161597248 metric_logger.py:185] completed step: 9, seconds: 0.059, TFLOP/s/device: 57.956, Tokens/s/device: 8735.711, total_weights: 16384, loss: 5.892 Per train step: Total TFLOPs: 3.40 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Sat Apr 25 15:37:17 UTC 2026 EXIT_CODE=0