XPK Start: Sat Apr 25 21:31:55 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 21:32:20.796490: 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 21:32:21.011051 132778434664256 max_utils.py:273] Attempting to initialize the jax distributed system...
I0425 21:32:30.053408 132778434664256 distributed.py:149] Starting JAX distributed service on [::]:8482
I0425 21:32:30.055825 132778434664256 distributed.py:172] Connecting to JAX distributed service on mt-16-zarr3-false-sv-kbeyw-slice-job-0-0.mt-16-zarr3-false-sv-kbeyw:8482
I0425 21:32:31.153337 132778434664256 max_utils.py:284] Jax distributed system initialized!
I0425 21:32:37.318595 132778434664256 max_utils.py:800] System Information: Jax Version: 0.9.2
I0425 21:32:37.318699 132778434664256 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0425 21:32:37.318740 132778434664256 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 21:32:37.318775 132778434664256 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0425 21:32:38.013194 132778434664256 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0425 21:32:38.013473 132778434664256 checkpointing.py:677] Setting up checkpoint logger...
I0425 21:32:38.013535 132778434664256 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=False
I0425 21:32:38.013579 132778434664256 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0425 21:32:38.013979 132778434664256 base_pytree_checkpoint_handler.py:441] Created BasePyTreeCheckpointHandler: use_ocdbt=True, use_zarr3=False, pytree_metadata_options=PyTreeMetadataOptions(support_rich_types=False), array_metadata_store=<orbax.checkpoint._src.metadata.array_metadata_store.Store object at 0x78c256f1fa10>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0425 21:32:40.934448 132778434664256 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0425 21:32:40.934695 132778434664256 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 0x78ad0044ac30>}, handler_registry=None
I0425 21:32:40.934947 132778434664256 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78ad0044ac30>` 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 21:32:40.934996 132778434664256 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78ad0044be90>` 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 21:32:40.935030 132778434664256 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 0x78ad0044ac30>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78ad0044ac30>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78ad0044be90>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78ad0044be90>}).
I0425 21:32:40.935369 132778434664256 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0425 21:32:40.935439 132778434664256 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x78ad001d7560> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0425 21:32:42.438785 132778434664256 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_false/checkpoints
I0425 21:32:42.448507 132778434664256 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_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x78ad00449bb0>
I0425 21:32:42.448678 132778434664256 checkpointing.py:301] Checkpoint manager created!
I0425 21:32:43.380824 132778434664256 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0425 21:32:43.380948 132778434664256 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 21:32:43.762676 132778434664256 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0425 21:32:43.762773 132778434664256 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 21:32:43.779443 132778434664256 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0425 21:32:43.779506 132778434664256 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 21:32:43.803556 132778434664256 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 21:32:43.803625 132778434664256 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 21:32:43.820286 132778434664256 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 21:32:43.820347 132778434664256 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 21:32:43.837024 132778434664256 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 21:32:43.837087 132778434664256 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 21:32:43.862280 132778434664256 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0425 21:32:43.862353 132778434664256 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 21:32:43.883578 132778434664256 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0425 21:32:43.883642 132778434664256 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0425 21:32:44.095487 132778434664256 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0425 21:32:44.095604 132778434664256 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0425 21:32:45.528866 132778434664256 maxtext_utils.py:1668] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0425 21:32:45.528995 132778434664256 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 21:32:45.529049 132778434664256 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 21:32:45.529152 132778434664256 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 21:32:45.529211 132778434664256 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 21:32:45.529252 132778434664256 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 21:32:45.529309 132778434664256 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 21:32:45.529363 132778434664256 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 21:32:45.529403 132778434664256 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 21:32:45.529439 132778434664256 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 21:32:45.529488 132778434664256 maxtext_utils.py:1668] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0425 21:32:45.529536 132778434664256 maxtext_utils.py:1668] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0425 21:32:46.023987 132778434664256 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0425 21:32:46.024085 132778434664256 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0425 21:32:46.039628 132778434664256 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0425 21:32:46.039689 132778434664256 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0425 21:32:57.162414 132778434664256 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 21:32:57.163222 132778434664256 metric_logger.py:301] number parameters: 1.104 billion
I0425 21:33:08.823246 132778434664256 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0425 21:33:08.976902 132778434664256 checkpointing.py:776] Waited 0.15362763404846191 seconds for step 0 to finish before starting checkpointing.
I0425 21:33:08.980022 132778434664256 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 21:33:08.981846 132778434664256 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0425 21:33:08.983488 132778434664256 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_false/checkpoints/0.
I0425 21:33:09.297705 132778434664256 signaling_client.py:364] Using JaxDistributedSignalingClient
I0425 21:33:09.298628 132778434664256 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0425 21:33:09.298687 132778434664256 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 21:33:09.572191 132778434664256 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.274559s
I0425 21:33:09.572375 132778434664256 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.508 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2800731658935547 s) (per-host)
I0425 21:33:09.572433 132778434664256 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.280141s (batch_requests_ready=0.002205s, total_serialization_initiated=0.277861s, others=0.000076s)
I0425 21:33:09.572532 132778434664256 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.284266s (all_items=0.000018s, per_item={'items': '0.00001836'}, temp_paths=0.284248)
I0425 21:33:09.573341 132778434664256 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.59 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_false/checkpoints/0.
I0425 21:33:09.573678 132649698834176 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 21:53:09.573639
I0425 21:33:09.584471 132778434664256 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 21:33:09.584699 132646454355712 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 21:33:09.584817 132778434664256 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_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777152788.9800036, 'wait_for_prev_duration_secs': 6.29425048828125e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777152788.9818854, 'checkpointer_blocking_duration_secs': 0.5919406414031982, 'get_old_steps_start_time': 1777152789.573851, 'get_old_steps_duration_secs': 2.86102294921875e-05, 'checkpoint_manager_blocking_start_time': 1777152788.9775226, 'checkpoint_manager_blocking_duration_secs': 0.6072573661804199}
I0425 21:33:09.584922 132778434664256 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0425 21:33:09.584969 132778434664256 max_utils.py:750]
Memstats: After params initialized:
I0425 21:33:09.585016 132778434664256 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0425 21:33:09.585047 132778434664256 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0425 21:33:09.585074 132778434664256 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0425 21:33:09.585111 132778434664256 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0425 21:33:09.897882 132778434664256 metric_logger.py:196] completed step: 0, seconds: 11.660, TFLOP/s/device: 1.165, Tokens/s/device: 175.644, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0425 21:33:10.073641 132778434664256 metric_logger.py:196] completed step: 1, seconds: 1.073, TFLOP/s/device: 12.662, Tokens/s/device: 1908.569, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0425 21:33:10.494045 132778434664256 metric_logger.py:196] completed step: 2, seconds: 0.028, TFLOP/s/device: 477.311, Tokens/s/device: 71945.479, total_weights: 65536, loss: 10.560, lm_loss: 10.560, perplexity: 38547.719
I0425 21:33:10.641502 132778434664256 metric_logger.py:196] completed step: 3, seconds: 0.421, TFLOP/s/device: 32.298, Tokens/s/device: 4868.320, total_weights: 65536, loss: 9.980, lm_loss: 9.980, perplexity: 21590.506
I0425 21:33:10.646022 132778434664256 checkpointing.py:772] Waiting for step 5 to finish before checkpoint...
I0425 21:33:10.935363 132778434664256 checkpointing.py:776] Waited 0.2893030643463135 seconds for step 5 to finish before starting checkpointing.
I0425 21:33:10.938621 132778434664256 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 21:33:12.533849 2854 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0425 21:33:14.560930 132647000631040 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_false/checkpoints/0/items/array_metadatas/process_6
I0425 21:33:51.672377 132649698834176 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.272 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.380038261413574 s) (per-host)
I0425 21:33:51.672501 132649698834176 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.098711s.
I0425 21:33:58.693464 132649698834176 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 49.119657s.
I0425 21:33:58.693685 132646454355712 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 21:33:58.693750 132646454355712 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 21:33:58.693799 132646454355712 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0425 21:33:58.695986 132646454355712 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0425 21:33:58.696171 132778434664256 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 21:33:58.696320 132778434664256 checkpoint_manager.py:1452] Waiting for previous save to complete took 47.757696 seconds. If this number is high, consider checkpointing less frequently.
I0425 21:33:58.698316 132778434664256 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 5
I0425 21:33:58.700773 132778434664256 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_false/checkpoints/5.
I0425 21:33:58.982120 132778434664256 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0425 21:33:58.982210 132778434664256 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 21:33:59.020471 132778434664256 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.039519s
I0425 21:33:59.020643 132778434664256 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 35.897 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04297208786010742 s) (per-host)
I0425 21:33:59.020696 132778434664256 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.043034s (batch_requests_ready=0.001774s, total_serialization_initiated=0.041191s, others=0.000069s)
I0425 21:33:59.020792 132778434664256 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.048057s (all_items=0.000017s, per_item={'items': '0.00001693'}, temp_paths=0.048040)
I0425 21:33:59.021553 132778434664256 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.32 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_false/checkpoints/5.
I0425 21:33:59.021907 132646454355712 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 21:53:59.021869
I0425 21:33:59.025446 132778434664256 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=5] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 21:33:59.025717 132649690441472 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 21:33:59.025879 132778434664256 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_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777152790.9385912, 'wait_for_prev_duration_secs': 47.75769639015198, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777152838.6983545, 'checkpointer_blocking_duration_secs': 0.32369446754455566, 'get_old_steps_start_time': 1777152839.022072, 'get_old_steps_duration_secs': 4.291534423828125e-05, 'checkpoint_manager_blocking_start_time': 1777152790.9359453, 'checkpoint_manager_blocking_duration_secs': 48.0898973941803}
I0425 21:33:59.025995 132778434664256 checkpointing.py:408] Started an asynchronous checkpoint save for step 5
I0425 21:33:59.026813 132778434664256 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.685, Tokens/s/device: 13367.535, total_weights: 65536, loss: 9.460, lm_loss: 9.460, perplexity: 12840.355
I0425 21:33:59.032968 132778434664256 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.265, Tokens/s/device: 13907.186, total_weights: 65536, loss: 8.959, lm_loss: 8.959, perplexity: 7776.467
I0425 21:34:04.164740 132642837157632 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_false/checkpoints/5/items/array_metadatas/process_6
I0425 21:34:21.868456 132778434664256 metric_logger.py:196] completed step: 6, seconds: 48.385, TFLOP/s/device: 0.281, Tokens/s/device: 42.327, total_weights: 65536, loss: 8.469, lm_loss: 8.469, perplexity: 4762.740
I0425 21:34:22.015899 132778434664256 metric_logger.py:196] completed step: 7, seconds: 22.689, TFLOP/s/device: 0.599, Tokens/s/device: 90.263, total_weights: 65536, loss: 8.003, lm_loss: 8.003, perplexity: 2989.151
I0425 21:34:22.163080 132778434664256 metric_logger.py:196] completed step: 8, seconds: 0.153, TFLOP/s/device: 89.016, Tokens/s/device: 13417.454, total_weights: 65536, loss: 7.572, lm_loss: 7.572, perplexity: 1943.098
I0425 21:34:22.309511 132778434664256 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0425 21:34:22.310223 132778434664256 checkpointing.py:776] Waited 0.000743865966796875 seconds for step 9 to finish before starting checkpointing.
I0425 21:34:22.312182 132778434664256 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=5][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 21:34:40.364327 132646454355712 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 38.167 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.38661551475525 s) (per-host)
I0425 21:34:40.364456 132646454355712 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.342436s.
I0425 21:34:49.206967 132646454355712 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 50.184930s.
I0425 21:34:49.207276 132649690441472 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 21:34:49.207403 132649690441472 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 21:34:49.207453 132649690441472 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is syncing with other hosts...
I0425 21:34:49.209991 132649690441472 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is done on all hosts.
I0425 21:34:49.210164 132778434664256 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 21:34:49.210290 132778434664256 checkpoint_manager.py:1452] Waiting for previous save to complete took 26.898115 seconds. If this number is high, consider checkpointing less frequently.
I0425 21:34:49.211919 132778434664256 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0425 21:34:49.213949 132778434664256 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_false/checkpoints/9.
I0425 21:34:49.500995 132778434664256 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0425 21:34:49.501091 132778434664256 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 21:34:49.537024 132778434664256 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.036892s
I0425 21:34:49.537209 132778434664256 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 38.056 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.040534257888793945 s) (per-host)
I0425 21:34:49.537262 132778434664256 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.040597s (batch_requests_ready=0.001722s, total_serialization_initiated=0.038808s, others=0.000068s)
I0425 21:34:49.537368 132778434664256 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.044726s (all_items=0.000015s, per_item={'items': '0.00001478'}, temp_paths=0.044711)
I0425 21:34:49.537945 132778434664256 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_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_16_zarr3_false/checkpoints/9.
I0425 21:34:49.538260 132642837157632 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 21:54:49.538222
I0425 21:34:49.539920 132778434664256 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 21:34:49.540203 132649690441472 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 21:34:49.540331 132778434664256 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_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777152862.3121521, 'wait_for_prev_duration_secs': 26.898114681243896, 'time_between_consecutive_saves_sec': 23.61612629890442, 'checkpointer_blocking_start_time': 1777152889.2119575, 'checkpointer_blocking_duration_secs': 0.3264484405517578, 'get_old_steps_start_time': 1777152889.5384226, 'get_old_steps_duration_secs': 2.2649765014648438e-05, 'checkpoint_manager_blocking_start_time': 1777152862.3104532, 'checkpoint_manager_blocking_duration_secs': 27.229848861694336}
I0425 21:34:49.540475 132778434664256 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0425 21:34:49.540523 132778434664256 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 21:34:54.992080 132648647255808 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_false/checkpoints/9/items/array_metadatas/process_6
I0425 21:35:31.396082 132642837157632 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.700 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.89936709403992 s) (per-host)
I0425 21:35:31.396216 132642837157632 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.857839s.
I0425 21:35:41.683754 132642837157632 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 52.145362s.
I0425 21:35:41.684033 132649690441472 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 21:35:41.684193 132649690441472 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 21:35:41.684237 132649690441472 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0425 21:35:41.685750 132649690441472 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0425 21:35:41.685935 132778434664256 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 21:35:41.686089 132778434664256 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 21:35:41.687118 132778434664256 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.269, Tokens/s/device: 13907.847, 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 21:35:50 UTC 2026
EXIT_CODE=0