XPK Start: Wed Apr 22 22:43:03 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. 2026-04-22 22:43:28.521886: 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) I0422 22:43:28.733226 132286463223616 max_utils.py:273] Attempting to initialize the jax distributed system... I0422 22:43:37.775903 132286463223616 distributed.py:149] Starting JAX distributed service on [::]:8482 I0422 22:43:37.778324 132286463223616 distributed.py:172] Connecting to JAX distributed service on mt-16-zarr3-true-sv-8xrns-slice-job-0-0.mt-16-zarr3-true-sv-8xrns:8482 I0422 22:43:38.725471 132286463223616 max_utils.py:284] Jax distributed system initialized! I0422 22:43:43.814937 132286463223616 max_utils.py:800] System Information: Jax Version: 0.9.2 I0422 22:43:43.815044 132286463223616 max_utils.py:801] System Information: Jaxlib Version: 0.9.2 I0422 22:43:43.815083 132286463223616 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 I0422 22:43:43.815245 132286463223616 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0422 22:43:44.503963 132286463223616 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0422 22:43:44.504258 132286463223616 checkpointing.py:677] Setting up checkpoint logger... I0422 22:43:44.504320 132286463223616 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True I0422 22:43:44.504366 132286463223616 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None I0422 22:43:44.504705 132286463223616 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 0x784fa0148c80>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0422 22:43:47.403411 132286463223616 checkpointing.py:265] Enabling policy for fixed interval checkpointing. I0422 22:43:47.403647 132286463223616 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 0x783c30527230>}, handler_registry=None I0422 22:43:47.403885 132286463223616 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x783c30527230>` 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`. I0422 22:43:47.403933 132286463223616 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x783c3052b680>` 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`. I0422 22:43:47.403969 132286463223616 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 0x783c30527230>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x783c30527230>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x783c3052b680>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x783c3052b680>}). I0422 22:43:47.404312 132286463223616 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34 I0422 22:43:47.404381 132286463223616 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x783c3036f2e0> timeout: 1200 secs and primary_host=0 for async checkpoint writes I0422 22:43:48.957859 132286463223616 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints I0422 22:43:49.388776 132286463223616 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_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x783c30526f60> I0422 22:43:49.388954 132286463223616 checkpointing.py:301] Checkpoint manager created! I0422 22:43:50.322912 132286463223616 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0422 22:43:50.323029 132286463223616 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 22:43:50.711428 132286463223616 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 22:43:50.711518 132286463223616 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 22:43:50.727984 132286463223616 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0422 22:43:50.728044 132286463223616 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0422 22:43:50.751636 132286463223616 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 22:43:50.751705 132286463223616 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 22:43:50.768247 132286463223616 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 22:43:50.768306 132286463223616 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 22:43:50.784941 132286463223616 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0422 22:43:50.785003 132286463223616 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 22:43:50.809937 132286463223616 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0422 22:43:50.810006 132286463223616 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0422 22:43:50.831084 132286463223616 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0422 22:43:50.831171 132286463223616 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0422 22:43:51.043643 132286463223616 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint I0422 22:43:51.043758 132286463223616 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0422 22:43:52.492295 132286463223616 maxtext_utils.py:1668] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0422 22:43:52.492420 132286463223616 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) I0422 22:43:52.492472 132286463223616 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) I0422 22:43:52.492540 132286463223616 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') I0422 22:43:52.492595 132286463223616 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) I0422 22:43:52.492633 132286463223616 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) I0422 22:43:52.492685 132286463223616 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) I0422 22:43:52.492736 132286463223616 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') I0422 22:43:52.492775 132286463223616 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) I0422 22:43:52.492810 132286463223616 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) I0422 22:43:52.492873 132286463223616 maxtext_utils.py:1668] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: P('embed_vocab', 'vocab') Physical: ('fsdp', None) I0422 22:43:52.492949 132286463223616 maxtext_utils.py:1668] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: P('vocab', 'embed_vocab') Physical: (None, 'fsdp') I0422 22:43:52.983434 132286463223616 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0422 22:43:52.983534 132286463223616 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0422 22:43:52.999022 132286463223616 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0422 22:43:52.999084 132286463223616 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0422 22:44:03.819467 132286463223616 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. I0422 22:44:03.820387 132286463223616 metric_logger.py:301] number parameters: 1.104 billion I0422 22:44:15.297818 132286463223616 checkpointing.py:772] Waiting for step 0 to finish before checkpoint... I0422 22:44:15.574434 132286463223616 checkpointing.py:776] Waited 0.2765936851501465 seconds for step 0 to finish before starting checkpointing. I0422 22:44:15.576849 132286463223616 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 22:44:15.578586 132286463223616 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0 I0422 22:44:15.580008 132286463223616 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints/0. I0422 22:44:16.010082 132286463223616 signaling_client.py:364] Using JaxDistributedSignalingClient I0422 22:44:16.011125 132286463223616 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0422 22:44:16.011179 132286463223616 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 I0422 22:44:16.288136 132286463223616 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.278125s I0422 22:44:16.288321 132286463223616 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.436 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28379178047180176 s) (per-host) I0422 22:44:16.288375 132286463223616 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.283857s (batch_requests_ready=0.002465s, total_serialization_initiated=0.281319s, others=0.000073s) I0422 22:44:16.288469 132286463223616 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.287874s (all_items=0.000018s, per_item={'items': '0.00001788'}, temp_paths=0.287856) I0422 22:44:16.289222 132286463223616 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.71 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints/0. I0422 22:44:16.289551 132160938866432 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 23:04:16.289510 I0422 22:44:16.300360 132286463223616 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0422 22:44:16.300717 132159312242432 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0422 22:44:16.300881 132286463223616 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776897855.5768318, 'wait_for_prev_duration_secs': 6.0558319091796875e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776897855.5786242, 'checkpointer_blocking_duration_secs': 0.7110834121704102, 'get_old_steps_start_time': 1776897856.2897346, 'get_old_steps_duration_secs': 2.9802322387695312e-05, 'checkpoint_manager_blocking_start_time': 1776897855.5749626, 'checkpoint_manager_blocking_duration_secs': 0.7258801460266113} I0422 22:44:16.300987 132286463223616 checkpointing.py:408] Started an asynchronous checkpoint save for step 0 I0422 22:44:16.301036 132286463223616 max_utils.py:750] Memstats: After params initialized: I0422 22:44:16.301087 132286463223616 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0)) I0422 22:44:16.301146 132286463223616 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0)) I0422 22:44:16.301176 132286463223616 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0)) I0422 22:44:16.301203 132286463223616 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0)) I0422 22:44:16.620259 132286463223616 metric_logger.py:196] completed step: 0, seconds: 11.477, TFLOP/s/device: 1.184, Tokens/s/device: 178.439, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0422 22:44:16.796900 132286463223616 metric_logger.py:196] completed step: 1, seconds: 1.321, TFLOP/s/device: 10.287, Tokens/s/device: 1550.540, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0422 22:44:17.202816 132286463223616 metric_logger.py:196] completed step: 2, seconds: 0.029, TFLOP/s/device: 462.777, Tokens/s/device: 69754.768, total_weights: 65536, loss: 10.560, lm_loss: 10.560, perplexity: 38547.719 I0422 22:44:17.350304 132286463223616 metric_logger.py:196] completed step: 3, seconds: 0.406, TFLOP/s/device: 33.442, Tokens/s/device: 5040.734, total_weights: 65536, loss: 9.980, lm_loss: 9.980, perplexity: 21590.506 I0422 22:44:17.354705 132286463223616 checkpointing.py:772] Waiting for step 5 to finish before checkpoint... I0422 22:44:17.644442 132286463223616 checkpointing.py:776] Waited 0.2897028923034668 seconds for step 5 to finish before starting checkpointing. I0422 22:44:17.647412 132286463223616 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 22:44:19.409062 2758 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0422 22:44:21.977382 132159320635136 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_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints/0/items/array_metadatas/process_6 I0422 22:44:53.804428 132160938866432 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 41.788 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.79985737800598 s) (per-host) I0422 22:44:53.804566 132160938866432 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.514900s. I0422 22:45:02.561617 132160938866432 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 46.271935s. I0422 22:45:02.561909 132159312242432 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 22:45:02.562027 132159312242432 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0422 22:45:02.562077 132159312242432 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0422 22:45:02.564372 132159312242432 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0422 22:45:02.564568 132286463223616 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. W0422 22:45:02.564709 132286463223616 checkpoint_manager.py:1452] Waiting for previous save to complete took 44.917295 seconds. If this number is high, consider checkpointing less frequently. I0422 22:45:02.566715 132286463223616 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 5 I0422 22:45:02.569255 132286463223616 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints/5. I0422 22:45:03.273591 132286463223616 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0422 22:45:03.273687 132286463223616 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 I0422 22:45:03.310174 132286463223616 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.037677s I0422 22:45:03.310343 132286463223616 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 37.547 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04108381271362305 s) (per-host) I0422 22:45:03.310391 132286463223616 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.041143s (batch_requests_ready=0.001748s, total_serialization_initiated=0.039330s, others=0.000065s) I0422 22:45:03.310483 132286463223616 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.046187s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.046172) I0422 22:45:03.311178 132286463223616 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.74 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints/5. I0422 22:45:03.311526 132159312242432 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 23:05:03.311483 I0422 22:45:03.316075 132286463223616 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=5] Starting CheckpointManager Save Finalize thread=save_finalize I0422 22:45:03.316324 132137969760000 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0422 22:45:03.316442 132286463223616 standard_logger.py:34] {'step': 5, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776897857.6473827, 'wait_for_prev_duration_secs': 44.91729545593262, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776897902.566754, 'checkpointer_blocking_duration_secs': 0.7449121475219727, 'get_old_steps_start_time': 1776897903.3116906, 'get_old_steps_duration_secs': 2.9325485229492188e-05, 'checkpoint_manager_blocking_start_time': 1776897857.645078, 'checkpoint_manager_blocking_duration_secs': 45.671329975128174} I0422 22:45:03.316619 132286463223616 checkpointing.py:408] Started an asynchronous checkpoint save for step 5 I0422 22:45:03.317411 132286463223616 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.739, Tokens/s/device: 13375.742, total_weights: 65536, loss: 9.460, lm_loss: 9.460, perplexity: 12840.355 I0422 22:45:03.323080 132286463223616 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.315, Tokens/s/device: 13914.745, total_weights: 65536, loss: 8.959, lm_loss: 8.959, perplexity: 7776.467 I0422 22:45:08.893220 132159320635136 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_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints/5/items/array_metadatas/process_6 I0422 22:45:26.767219 132286463223616 metric_logger.py:196] completed step: 6, seconds: 45.967, TFLOP/s/device: 0.296, Tokens/s/device: 44.554, total_weights: 65536, loss: 8.469, lm_loss: 8.469, perplexity: 4762.740 I0422 22:45:26.914633 132286463223616 metric_logger.py:196] completed step: 7, seconds: 23.298, TFLOP/s/device: 0.583, Tokens/s/device: 87.904, total_weights: 65536, loss: 8.003, lm_loss: 8.003, perplexity: 2989.151 I0422 22:45:27.061923 132286463223616 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.463, Tokens/s/device: 13484.862, total_weights: 65536, loss: 7.572, lm_loss: 7.572, perplexity: 1943.098 I0422 22:45:27.208554 132286463223616 checkpointing.py:772] Waiting for step 9 to finish before checkpoint... I0422 22:45:27.209224 132286463223616 checkpointing.py:776] Waited 0.0006937980651855469 seconds for step 9 to finish before starting checkpointing. I0422 22:45:27.211290 132286463223616 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=5][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 22:45:45.955296 132159312242432 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.005 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.68599557876587 s) (per-host) I0422 22:45:45.955420 132159312242432 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.643782s. I0422 22:45:53.224923 132159312242432 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 49.913269s. I0422 22:45:53.225167 132137969760000 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 22:45:53.225224 132137969760000 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0422 22:45:53.225276 132137969760000 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is syncing with other hosts... I0422 22:45:53.227839 132137969760000 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is done on all hosts. I0422 22:45:53.227992 132286463223616 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. W0422 22:45:53.228132 132286463223616 checkpoint_manager.py:1452] Waiting for previous save to complete took 26.016856 seconds. If this number is high, consider checkpointing less frequently. I0422 22:45:53.229628 132286463223616 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9 I0422 22:45:53.231848 132286463223616 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints/9. I0422 22:45:53.522388 132286463223616 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0422 22:45:53.522479 132286463223616 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 I0422 22:45:53.557685 132286463223616 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.036268s I0422 22:45:53.557853 132286463223616 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 38.894 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.03966069221496582 s) (per-host) I0422 22:45:53.557903 132286463223616 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.039722s (batch_requests_ready=0.001631s, total_serialization_initiated=0.038024s, others=0.000067s) I0422 22:45:53.557991 132286463223616 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.044026s (all_items=0.000029s, per_item={'items': '0.00002909'}, temp_paths=0.043997) I0422 22:45:53.558674 132286463223616 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.33 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints/9. I0422 22:45:53.558995 132137969760000 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 23:05:53.558957 I0422 22:45:53.564948 132286463223616 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0422 22:45:53.565231 132159320635136 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0422 22:45:53.565387 132286463223616 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776897927.211253, 'wait_for_prev_duration_secs': 26.01685643196106, 'time_between_consecutive_saves_sec': 24.646841049194336, 'checkpointer_blocking_start_time': 1776897953.2296667, 'checkpointer_blocking_duration_secs': 0.32949280738830566, 'get_old_steps_start_time': 1776897953.5591788, 'get_old_steps_duration_secs': 2.4557113647460938e-05, 'checkpoint_manager_blocking_start_time': 1776897927.2094572, 'checkpoint_manager_blocking_duration_secs': 26.35589861869812} I0422 22:45:53.565495 132286463223616 checkpointing.py:408] Started an asynchronous checkpoint save for step 9 I0422 22:45:53.565538 132286463223616 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0422 22:45:59.416164 132159876613888 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_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_16_zarr3_true/checkpoints/9/items/array_metadatas/process_6 I0422 22:46:36.904806 132137969760000 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 36.408 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.38657355308533 s) (per-host) I0422 22:46:36.904936 132137969760000 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.345813s. I0422 22:46:43.987189 132137969760000 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 50.428051s. I0422 22:46:43.987449 132159320635136 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0422 22:46:43.987514 132159320635136 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0422 22:46:43.987556 132159320635136 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0422 22:46:43.990251 132159320635136 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0422 22:46:43.990404 132286463223616 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. I0422 22:46:43.990555 132286463223616 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0422 22:46:43.991511 132286463223616 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.175, Tokens/s/device: 13893.694, 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: Wed Apr 22 22:46:53 UTC 2026 EXIT_CODE=0