XPK Start: Sat Apr 25 21:22:06 UTC 2026 PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. `rope_parameters`'s factor field must be a float >= 1, got 40 `rope_parameters`'s beta_fast field must be a float, got 32 `rope_parameters`'s beta_slow field must be a float, got 1 DeepseekV32Config got `key=rope_scaling` in kwargs but hasn't set it as attribute. For RoPE standardization you need to set `self.rope_parameters` in model's config. 2026-04-25 21:22:31.607262: 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 21:22:31.818147 135367305602880 max_utils.py:273] Attempting to initialize the jax distributed system... I0425 21:22:40.859216 135367305602880 distributed.py:149] Starting JAX distributed service on [::]:8482 I0425 21:22:40.861683 135367305602880 distributed.py:172] Connecting to JAX distributed service on mt-16-zarr3-true-sv-5v97t-slice-job-0-0.mt-16-zarr3-true-sv-5v97t:8482 I0425 21:22:41.783782 135367305602880 max_utils.py:284] Jax distributed system initialized! I0425 21:22:46.956195 135367305602880 max_utils.py:800] System Information: Jax Version: 0.9.2 I0425 21:22:46.956300 135367305602880 max_utils.py:801] System Information: Jaxlib Version: 0.9.2 I0425 21:22:46.956342 135367305602880 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 I0425 21:22:46.956379 135367305602880 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0425 21:22:47.654187 135367305602880 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0425 21:22:47.654466 135367305602880 checkpointing.py:677] Setting up checkpoint logger... I0425 21:22:47.654525 135367305602880 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0425 21:22:47.654569 135367305602880 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None I0425 21:22:47.654930 135367305602880 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 0x7b1d07314410>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0425 21:22:50.469777 135367305602880 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0425 21:22:50.470015 135367305602880 checkpoint_manager.py:708] [process=6][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b1be4599580>}, handler_registry=None I0425 21:22:50.470267 135367305602880 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b1be4599580>` 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 21:22:50.470318 135367305602880 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b09747a17c0>` 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 21:22:50.470355 135367305602880 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 0x7b1be4599580>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b1be4599580>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b09747a17c0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b09747a17c0>}). I0425 21:22:50.470692 135367305602880 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34 I0425 21:22:50.470762 135367305602880 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7b097417f2e0> timeout: 1200 secs and primary_host=0 for async checkpoint writes I0425 21:22:51.581279 135367305602880 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints I0425 21:22:51.673054 135367305602880 checkpoint_manager.py:929] [process=6][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=5), 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_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7b09747a0f50> I0425 21:22:51.673246 135367305602880 checkpointing.py:301] Checkpoint manager created! I0425 21:22:52.620402 135367305602880 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0425 21:22:52.620528 135367305602880 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0425 21:22:53.004929 135367305602880 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0425 21:22:53.005030 135367305602880 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0425 21:22:53.021704 135367305602880 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0425 21:22:53.021766 135367305602880 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0425 21:22:53.045712 135367305602880 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 21:22:53.045779 135367305602880 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 21:22:53.062541 135367305602880 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 21:22:53.062604 135367305602880 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 21:22:53.079332 135367305602880 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0425 21:22:53.079395 135367305602880 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 21:22:53.104555 135367305602880 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0425 21:22:53.104627 135367305602880 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0425 21:22:53.126634 135367305602880 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0425 21:22:53.126705 135367305602880 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0425 21:22:53.337851 135367305602880 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0425 21:22:53.337960 135367305602880 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0425 21:22:54.801320 135367305602880 maxtext_utils.py:1668] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0425 21:22:54.801446 135367305602880 maxtext_utils.py:1668] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: P('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0425 21:22:54.801506 135367305602880 maxtext_utils.py:1668] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: P('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0425 21:22:54.801571 135367305602880 maxtext_utils.py:1668] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: P('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0425 21:22:54.801622 135367305602880 maxtext_utils.py:1668] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: P('norm', 'layers') Physical: (None, None) I0425 21:22:54.801660 135367305602880 maxtext_utils.py:1668] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: P('norm', 'layers') Physical: (None, None) I0425 21:22:54.801711 135367305602880 maxtext_utils.py:1668] 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) I0425 21:22:54.801766 135367305602880 maxtext_utils.py:1668] 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') I0425 21:22:54.801806 135367305602880 maxtext_utils.py:1668] 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) I0425 21:22:54.801842 135367305602880 maxtext_utils.py:1668] 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) I0425 21:22:54.801888 135367305602880 maxtext_utils.py:1668] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: P('embed_vocab', 'vocab') Physical: ('fsdp', None) I0425 21:22:54.801935 135367305602880 maxtext_utils.py:1668] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: P('vocab', 'embed_vocab') Physical: (None, 'fsdp') I0425 21:22:55.298705 135367305602880 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0425 21:22:55.298799 135367305602880 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0425 21:22:55.314545 135367305602880 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0425 21:22:55.314604 135367305602880 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0425 21:23:06.420825 135367305602880 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. I0425 21:23:06.421648 135367305602880 metric_logger.py:301] number parameters: 1.104 billion I0425 21:23:17.751927 135367305602880 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0425 21:23:17.905078 135367305602880 checkpointing.py:776] Waited 0.15313482284545898 seconds for step 0 to finish before starting checkpointing. I0425 21:23:17.907533 135367305602880 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0425 21:23:17.909255 135367305602880 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0 I0425 21:23:17.910611 135367305602880 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints/0. I0425 21:23:18.252512 135367305602880 signaling_client.py:364] Using JaxDistributedSignalingClient I0425 21:23:18.253376 135367305602880 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0425 21:23:18.253433 135367305602880 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 21:23:18.529403 135367305602880 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.276964s I0425 21:23:18.529582 135367305602880 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.454 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2828238010406494 s) (per-host) I0425 21:23:18.529635 135367305602880 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.282888s (batch_requests_ready=0.002654s, total_serialization_initiated=0.280160s, others=0.000074s) I0425 21:23:18.529731 135367305602880 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.287032s (all_items=0.000019s, per_item={'items': '0.00001860'}, temp_paths=0.287013) I0425 21:23:18.530553 135367305602880 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.62 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints/0. I0425 21:23:18.530879 135240969611008 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 21:43:18.530842 I0425 21:23:18.953976 135367305602880 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0425 21:23:18.954380 135239079847680 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0425 21:23:18.954549 135367305602880 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777152197.9075134, 'wait_for_prev_duration_secs': 6.461143493652344e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777152197.9092987, 'checkpointer_blocking_duration_secs': 0.621732234954834, 'get_old_steps_start_time': 1777152198.5310566, 'get_old_steps_duration_secs': 3.075599670410156e-05, 'checkpoint_manager_blocking_start_time': 1777152197.9056263, 'checkpoint_manager_blocking_duration_secs': 1.048882007598877} I0425 21:23:18.954667 135367305602880 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0425 21:23:18.954718 135367305602880 max_utils.py:750] Memstats: After params initialized: I0425 21:23:18.954776 135367305602880 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0)) I0425 21:23:18.954809 135367305602880 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0)) I0425 21:23:18.954836 135367305602880 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0)) I0425 21:23:18.954861 135367305602880 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0)) I0425 21:23:19.274963 135367305602880 metric_logger.py:196] completed step: 0, seconds: 11.330, TFLOP/s/device: 1.199, Tokens/s/device: 180.757, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0425 21:23:19.447201 135367305602880 metric_logger.py:196] completed step: 1, seconds: 1.522, TFLOP/s/device: 8.930, Tokens/s/device: 1345.989, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0425 21:23:19.873507 135367305602880 metric_logger.py:196] completed step: 2, seconds: 0.025, TFLOP/s/device: 550.666, Tokens/s/device: 83002.351, total_weights: 65536, loss: 10.560, lm_loss: 10.560, perplexity: 38547.719 I0425 21:23:20.020951 135367305602880 metric_logger.py:196] completed step: 3, seconds: 0.427, TFLOP/s/device: 31.827, Tokens/s/device: 4797.354, total_weights: 65536, loss: 9.980, lm_loss: 9.980, perplexity: 21590.506 I0425 21:23:20.025525 135367305602880 checkpointing.py:772] Waiting for step 5 to finish before checkpoint... I0425 21:23:20.314769 135367305602880 checkpointing.py:776] Waited 0.28920936584472656 seconds for step 5 to finish before starting checkpointing. I0425 21:23:20.317904 135367305602880 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 21:23:21.705213 2758 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0425 21:23:23.772607 135239088240384 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints/0/items/array_metadatas/process_6 I0425 21:23:55.925178 135240969611008 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 41.923 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.67838168144226 s) (per-host) I0425 21:23:55.925304 135240969611008 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.394316s. I0425 21:24:01.636157 135240969611008 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 43.105153s. I0425 21:24:01.636470 135239079847680 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 21:24:01.636608 135239079847680 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0425 21:24:01.636658 135239079847680 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0425 21:24:01.638196 135239079847680 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0425 21:24:01.638347 135367305602880 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0425 21:24:01.638436 135367305602880 checkpoint_manager.py:1452] Waiting for previous save to complete took 41.320530 seconds. If this number is high, consider checkpointing less frequently. I0425 21:24:01.640602 135367305602880 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 5 I0425 21:24:01.643205 135367305602880 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints/5. I0425 21:24:02.281904 135367305602880 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0425 21:24:02.281999 135367305602880 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 21:24:02.326486 135367305602880 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.045695s I0425 21:24:02.326667 135367305602880 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 31.334 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.049230337142944336 s) (per-host) I0425 21:24:02.326717 135367305602880 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.049291s (batch_requests_ready=0.001898s, total_serialization_initiated=0.047324s, others=0.000069s) I0425 21:24:02.326806 135367305602880 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.054363s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.054348) I0425 21:24:02.327524 135367305602880 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.69 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints/5. I0425 21:24:02.327890 135239079847680 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 21:44:02.327853 I0425 21:24:02.329888 135367305602880 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=5] Starting CheckpointManager Save Finalize thread=save_finalize I0425 21:24:02.330195 135219680089856 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0425 21:24:02.330345 135367305602880 standard_logger.py:34] {'step': 5, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777152200.317873, 'wait_for_prev_duration_secs': 41.32052993774414, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777152241.6406405, 'checkpointer_blocking_duration_secs': 0.6873958110809326, 'get_old_steps_start_time': 1777152242.3280592, 'get_old_steps_duration_secs': 3.170967102050781e-05, 'checkpoint_manager_blocking_start_time': 1777152200.3154376, 'checkpoint_manager_blocking_duration_secs': 42.01487398147583} I0425 21:24:02.330503 135367305602880 checkpointing.py:408] Started an asynchronous checkpoint save for step 5 I0425 21:24:02.331291 135367305602880 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.777, Tokens/s/device: 13381.423, total_weights: 65536, loss: 9.460, lm_loss: 9.460, perplexity: 12840.355 I0425 21:24:02.337424 135367305602880 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.236, Tokens/s/device: 13902.843, total_weights: 65536, loss: 8.959, lm_loss: 8.959, perplexity: 7776.467 I0425 21:24:07.442829 135239088240384 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints/5/items/array_metadatas/process_6 I0425 21:24:25.082329 135367305602880 metric_logger.py:196] completed step: 6, seconds: 42.310, TFLOP/s/device: 0.321, Tokens/s/device: 48.404, total_weights: 65536, loss: 8.469, lm_loss: 8.469, perplexity: 4762.740 I0425 21:24:25.229763 135367305602880 metric_logger.py:196] completed step: 7, seconds: 22.599, TFLOP/s/device: 0.601, Tokens/s/device: 90.623, total_weights: 65536, loss: 8.003, lm_loss: 8.003, perplexity: 2989.151 I0425 21:24:25.377282 135367305602880 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.408, Tokens/s/device: 13476.610, total_weights: 65536, loss: 7.572, lm_loss: 7.572, perplexity: 1943.098 I0425 21:24:25.523619 135367305602880 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0425 21:24:25.524296 135367305602880 checkpointing.py:776] Waited 0.0006976127624511719 seconds for step 9 to finish before starting checkpointing. I0425 21:24:25.526405 135367305602880 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=5][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 21:24:44.208403 135239079847680 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.671 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.930927753448486 s) (per-host) I0425 21:24:44.208521 135239079847680 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.880519s. I0425 21:24:50.706603 135239079847680 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 48.378579s. I0425 21:24:50.706885 135219680089856 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 21:24:50.707008 135219680089856 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0425 21:24:50.707057 135219680089856 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is syncing with other hosts... I0425 21:24:50.708459 135219680089856 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is done on all hosts. I0425 21:24:50.710837 135367305602880 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=5][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=5. W0425 21:24:50.710958 135367305602880 checkpoint_manager.py:1452] Waiting for previous save to complete took 25.184563 seconds. If this number is high, consider checkpointing less frequently. I0425 21:24:50.712440 135367305602880 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9 I0425 21:24:50.714707 135367305602880 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints/9. I0425 21:24:51.007545 135367305602880 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0425 21:24:51.007639 135367305602880 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 21:24:51.051879 135367305602880 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.045195s I0425 21:24:51.052050 135367305602880 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 31.640 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.048753976821899414 s) (per-host) I0425 21:24:51.052114 135367305602880 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.048815s (batch_requests_ready=0.001739s, total_serialization_initiated=0.047010s, others=0.000066s) I0425 21:24:51.052206 135367305602880 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.053044s (all_items=0.000011s, per_item={'items': '0.00001097'}, temp_paths=0.053033) I0425 21:24:51.052865 135367305602880 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.34 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints/9. I0425 21:24:51.053175 135219680089856 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 21:44:51.053141 I0425 21:24:51.055090 135367305602880 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0425 21:24:51.055402 135239636600576 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0425 21:24:51.055535 135367305602880 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777152265.5263727, 'wait_for_prev_duration_secs': 25.184562921524048, 'time_between_consecutive_saves_sec': 23.88813591003418, 'checkpointer_blocking_start_time': 1777152290.7124786, 'checkpointer_blocking_duration_secs': 0.34084320068359375, 'get_old_steps_start_time': 1777152291.0533385, 'get_old_steps_duration_secs': 2.5272369384765625e-05, 'checkpoint_manager_blocking_start_time': 1777152265.5245192, 'checkpoint_manager_blocking_duration_secs': 25.530985355377197} I0425 21:24:51.055697 135367305602880 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0425 21:24:51.055742 135367305602880 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0425 21:24:56.133762 135239088240384 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_true/checkpoints/9/items/array_metadatas/process_6 I0425 21:25:32.599220 135219680089856 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.975 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.595887660980225 s) (per-host) I0425 21:25:32.599337 135219680089856 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.546048s. I0425 21:25:39.169881 135219680089856 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 48.116577s. I0425 21:25:39.170158 135239636600576 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0425 21:25:39.170276 135239636600576 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0425 21:25:39.170317 135239636600576 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0425 21:25:39.171758 135239636600576 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0425 21:25:39.171933 135367305602880 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0425 21:25:39.172079 135367305602880 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0425 21:25:39.173053 135367305602880 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.226, Tokens/s/device: 13901.333, total_weights: 65536, loss: 7.181, lm_loss: 7.181, perplexity: 1314.495 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Sat Apr 25 21:25:51 UTC 2026 EXIT_CODE=0