XPK Start: Sat Apr 18 13:53:08 UTC 2026 PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. 2026-04-18 13:53:32.856615: 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) I0418 13:53:33.389420 134904225339200 max_utils.py:273] Attempting to initialize the jax distributed system... I0418 13:53:42.432156 134904225339200 distributed.py:149] Starting JAX distributed service on [::]:8482 I0418 13:53:42.434605 134904225339200 distributed.py:172] Connecting to JAX distributed service on mt-13-scan-layers-true-c448i-slice-job-0-0.mt-13-scan-layers-true-c448i:8482 I0418 13:53:43.845241 134904225339200 max_utils.py:284] Jax distributed system initialized! I0418 13:53:48.943924 134904225339200 max_utils.py:800] System Information: Jax Version: 0.9.2 I0418 13:53:48.944024 134904225339200 max_utils.py:801] System Information: Jaxlib Version: 0.9.2 I0418 13:53:48.944063 134904225339200 max_utils.py:802] System Information: Jax Backend: PJRT C API TFRT TPU v6 lite Built on Mar 4 2026 11:32:08 (1772652728) cl/878335365 I0418 13:53:48.944124 134904225339200 train_utils.py:348] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0418 13:53:49.641209 134904225339200 maxtext_utils.py:1551] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0418 13:53:49.641487 134904225339200 checkpointing.py:677] Setting up checkpoint logger... I0418 13:53:49.641541 134904225339200 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0418 13:53:49.641586 134904225339200 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None I0418 13:53:49.641949 134904225339200 base_pytree_checkpoint_handler.py:441] 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 0x7ab13571c110>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0418 13:53:53.442666 134904225339200 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0418 13:53:53.442899 134904225339200 checkpoint_manager.py:708] [process=7][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a9da819e630>}, handler_registry=None I0418 13:53:53.443146 134904225339200 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a9da819e630>` 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`. I0418 13:53:53.443194 134904225339200 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a9da0797230>` 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`. I0418 13:53:53.443230 134904225339200 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 0x7a9da819e630>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a9da819e630>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a9da0797230>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a9da0797230>}). I0418 13:53:53.443544 134904225339200 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34 I0418 13:53:53.443614 134904225339200 async_checkpointer.py:192] [process=7][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7a9da03c7c40> timeout: 1200 secs and primary_host=0 for async checkpoint writes I0418 13:53:55.772903 134904225339200 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_13_scan_layers_true/checkpoints I0418 13:53:56.218443 134904225339200 checkpoint_manager.py:929] [process=7][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_main_20260418_125719/linen_xpk_main_20260418_125719_13_scan_layers_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7a9da0793c50> I0418 13:53:56.218608 134904225339200 checkpointing.py:301] Checkpoint manager created! I0418 13:53:57.142573 134904225339200 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0418 13:53:57.142683 134904225339200 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0418 13:53:57.518963 134904225339200 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0418 13:53:57.519052 134904225339200 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0418 13:53:57.535205 134904225339200 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0418 13:53:57.535260 134904225339200 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0418 13:53:57.558798 134904225339200 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0418 13:53:57.558867 134904225339200 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0418 13:53:57.575425 134904225339200 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0418 13:53:57.575487 134904225339200 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0418 13:53:57.591599 134904225339200 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0418 13:53:57.591661 134904225339200 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0418 13:53:57.616921 134904225339200 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0418 13:53:57.616987 134904225339200 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0418 13:53:57.638034 134904225339200 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0418 13:53:57.638118 134904225339200 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0418 13:53:57.857830 134904225339200 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0418 13:53:57.857938 134904225339200 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0418 13:53:59.286585 134904225339200 maxtext_utils.py:1654] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0418 13:53:59.286710 134904225339200 maxtext_utils.py:1654] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: P('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0418 13:53:59.286763 134904225339200 maxtext_utils.py:1654] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: P('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0418 13:53:59.286821 134904225339200 maxtext_utils.py:1654] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: P('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0418 13:53:59.286873 134904225339200 maxtext_utils.py:1654] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: P('norm', 'layers') Physical: (None, None) I0418 13:53:59.286917 134904225339200 maxtext_utils.py:1654] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: P('norm', 'layers') Physical: (None, None) I0418 13:53:59.286968 134904225339200 maxtext_utils.py:1654] params/params/decoder/layers/self_attention/key/kernel Shape: float32[2048,16,16,128] Logical: P('embed', 'layers', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None, None) I0418 13:53:59.287018 134904225339200 maxtext_utils.py:1654] params/params/decoder/layers/self_attention/out/kernel Shape: float32[16,16,128,2048] Logical: P('heads', 'layers', 'kv', 'embed') Physical: (None, None, None, 'fsdp') I0418 13:53:59.287057 134904225339200 maxtext_utils.py:1654] params/params/decoder/layers/self_attention/query/kernel Shape: float32[2048,16,16,128] Logical: P('embed', 'layers', 'q_heads', 'kv') Physical: ('fsdp', None, None, None) I0418 13:53:59.287109 134904225339200 maxtext_utils.py:1654] params/params/decoder/layers/self_attention/value/kernel Shape: float32[2048,16,16,128] Logical: P('embed', 'layers', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None, None) I0418 13:53:59.287158 134904225339200 maxtext_utils.py:1654] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: P('embed_vocab', 'vocab') Physical: ('fsdp', None) I0418 13:53:59.287206 134904225339200 maxtext_utils.py:1654] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: P('vocab', 'embed_vocab') Physical: (None, 'fsdp') I0418 13:53:59.775805 134904225339200 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0418 13:53:59.775898 134904225339200 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0418 13:53:59.791150 134904225339200 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0418 13:53:59.791208 134904225339200 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0418 13:54:10.886161 134904225339200 max_utils.py:791] Total memory size: 1.5 GB, Output size: 0.4 GB, Temp size: 1.1 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0418 13:54:10.886972 134904225339200 metric_logger.py:301] number parameters: 1.104 billion I0418 13:54:22.275099 134904225339200 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0418 13:54:22.428277 134904225339200 checkpointing.py:776] Waited 0.1531682014465332 seconds for step 0 to finish before starting checkpointing. I0418 13:54:22.430805 134904225339200 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0418 13:54:22.432466 134904225339200 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 0 I0418 13:54:22.433899 134904225339200 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_13_scan_layers_true/checkpoints/0. I0418 13:54:23.169906 134904225339200 signaling_client.py:364] Using JaxDistributedSignalingClient I0418 13:54:23.170911 134904225339200 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0418 13:54:23.170968 134904225339200 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 I0418 13:54:23.444356 134904225339200 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.274523s I0418 13:54:23.444531 134904225339200 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 5.507 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28011417388916016 s) (per-host) I0418 13:54:23.444586 134904225339200 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.280181s (batch_requests_ready=0.002351s, total_serialization_initiated=0.277755s, others=0.000074s) I0418 13:54:23.444677 134904225339200 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.284440s (all_items=0.000019s, per_item={'items': '0.00001931'}, temp_paths=0.284421) I0418 13:54:23.445441 134904225339200 event_tracking.py:125] [process=7] [async] Finished blocking save in 1.01 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_13_scan_layers_true/checkpoints/0. I0418 13:54:23.445741 134778125887232 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-18 14:14:23.445705 I0418 13:54:23.457742 134904225339200 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0418 13:54:23.458031 134776489596672 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save. I0418 13:54:23.458172 134904225339200 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_13_scan_layers_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776520462.430787, 'wait_for_prev_duration_secs': 6.0558319091796875e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776520462.4325032, 'checkpointer_blocking_duration_secs': 1.01338529586792, 'get_old_steps_start_time': 1776520463.4459143, 'get_old_steps_duration_secs': 3.0994415283203125e-05, 'checkpoint_manager_blocking_start_time': 1776520462.4288256, 'checkpoint_manager_blocking_duration_secs': 1.029303789138794} I0418 13:54:23.458282 134904225339200 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0418 13:54:23.458333 134904225339200 max_utils.py:750] Memstats: After params initialized: I0418 13:54:23.458386 134904225339200 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_26(process=7,(2,6,0,0)) I0418 13:54:23.458420 134904225339200 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_27(process=7,(3,6,0,0)) I0418 13:54:23.458448 134904225339200 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_30(process=7,(2,7,0,0)) I0418 13:54:23.458473 134904225339200 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_31(process=7,(3,7,0,0)) I0418 13:54:23.771928 134904225339200 metric_logger.py:196] completed step: 0, seconds: 11.388, TFLOP/s/device: 1.193, Tokens/s/device: 179.838, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0418 13:54:23.944504 134904225339200 metric_logger.py:196] completed step: 1, seconds: 1.495, TFLOP/s/device: 9.086, Tokens/s/device: 1369.596, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0418 13:54:24.369333 134904225339200 metric_logger.py:196] completed step: 2, seconds: 0.025, TFLOP/s/device: 536.087, Tokens/s/device: 80804.892, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129 I0418 13:54:24.516687 134904225339200 metric_logger.py:196] completed step: 3, seconds: 0.425, TFLOP/s/device: 31.981, Tokens/s/device: 4820.457, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203 I0418 13:54:24.811680 134904225339200 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.864, Tokens/s/device: 13394.551, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614 I0418 13:54:24.817182 134904225339200 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.254, Tokens/s/device: 13905.486, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717 I0418 13:54:27.214677 2785 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0418 13:54:29.613729 134776497989376 array_metadata_store.py:203] [process=7][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_13_scan_layers_true/checkpoints/0/items/array_metadatas/process_7 I0418 13:54:47.387987 134904225339200 metric_logger.py:196] completed step: 6, seconds: 0.295, TFLOP/s/device: 45.990, Tokens/s/device: 6932.174, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759 I0418 13:54:47.535530 134904225339200 metric_logger.py:196] completed step: 7, seconds: 22.424, TFLOP/s/device: 0.606, Tokens/s/device: 91.331, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905 I0418 13:54:47.682917 134904225339200 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.251, Tokens/s/device: 13452.885, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406 I0418 13:54:47.829584 134904225339200 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0418 13:54:47.830293 134904225339200 checkpointing.py:776] Waited 0.0007281303405761719 seconds for step 9 to finish before starting checkpointing. I0418 13:54:47.832404 134904225339200 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0418 13:55:00.055947 134778125887232 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 42.817 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.89147996902466 s) (per-host) I0418 13:55:00.056066 134778125887232 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.610222s. I0418 13:55:10.328424 134778125887232 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 46.882562s. I0418 13:55:10.328706 134776489596672 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save. I0418 13:55:10.328826 134776489596672 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save. I0418 13:55:10.328876 134776489596672 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0418 13:55:10.330738 134776489596672 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0418 13:55:10.330915 134904225339200 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0418 13:55:10.331047 134904225339200 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.498642 seconds. If this number is high, consider checkpointing less frequently. I0418 13:55:10.333013 134904225339200 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 9 I0418 13:55:10.335196 134904225339200 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_13_scan_layers_true/checkpoints/9. I0418 13:55:11.048717 134904225339200 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0418 13:55:11.048813 134904225339200 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 I0418 13:55:11.080966 134904225339200 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.033271s I0418 13:55:11.081155 134904225339200 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 41.790 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.03691220283508301 s) (per-host) I0418 13:55:11.081209 134904225339200 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.036977s (batch_requests_ready=0.001950s, total_serialization_initiated=0.034955s, others=0.000072s) I0418 13:55:11.081298 134904225339200 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.041118s (all_items=0.000020s, per_item={'items': '0.00002027'}, temp_paths=0.041097) I0418 13:55:11.082003 134904225339200 event_tracking.py:125] [process=7] [async] Finished blocking save in 0.75 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_13_scan_layers_true/checkpoints/9. I0418 13:55:11.082340 134776489596672 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-18 14:15:11.082303 I0418 13:55:11.088903 134904225339200 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0418 13:55:11.089205 134776497989376 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save. I0418 13:55:11.089371 134904225339200 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_13_scan_layers_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776520487.8323739, 'wait_for_prev_duration_secs': 22.49864172935486, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776520510.3330514, 'checkpointer_blocking_duration_secs': 0.7494254112243652, 'get_old_steps_start_time': 1776520511.0825005, 'get_old_steps_duration_secs': 3.0279159545898438e-05, 'checkpoint_manager_blocking_start_time': 1776520487.8305392, 'checkpoint_manager_blocking_duration_secs': 23.25879693031311} I0418 13:55:11.089482 134904225339200 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0418 13:55:11.089529 134904225339200 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0418 13:55:18.643489 134750790457088 array_metadata_store.py:203] [process=7][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_13_scan_layers_true/checkpoints/9/items/array_metadatas/process_7 I0418 13:55:54.626799 134776489596672 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 36.244 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.582512855529785 s) (per-host) I0418 13:55:54.626925 134776489596672 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.544476s. I0418 13:56:03.966576 134776489596672 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 52.884110s. I0418 13:56:03.966847 134776497989376 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save. I0418 13:56:03.966980 134776497989376 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save. I0418 13:56:03.967029 134776497989376 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0418 13:56:03.968533 134776497989376 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0418 13:56:03.968696 134904225339200 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0418 13:56:03.968852 134904225339200 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0418 13:56:03.969925 134904225339200 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.163, Tokens/s/device: 13891.809, total_weights: 65536, loss: 8.179, lm_loss: 8.179, perplexity: 3564.635 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Sat Apr 18 13:56:14 UTC 2026 EXIT_CODE=0