XPK Start: Sat Apr 25 21:15:04 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 21:15:29.679500: 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 21:15:29.889027 136514901002048 max_utils.py:273] Attempting to initialize the jax distributed system...
I0425 21:15:38.931422 136514901002048 distributed.py:149] Starting JAX distributed service on [::]:8482
I0425 21:15:38.933912 136514901002048 distributed.py:172] Connecting to JAX distributed service on mt-15-ocdbt-false-sv-c5xw3-slice-job-0-0.mt-15-ocdbt-false-sv-c5xw3:8482
I0425 21:15:39.883600 136514901002048 max_utils.py:284] Jax distributed system initialized!
I0425 21:15:44.982714 136514901002048 max_utils.py:800] System Information: Jax Version: 0.9.2
I0425 21:15:44.982825 136514901002048 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0425 21:15:44.982868 136514901002048 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 21:15:44.982903 136514901002048 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0425 21:15:45.678570 136514901002048 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0425 21:15:45.678848 136514901002048 checkpointing.py:677] Setting up checkpoint logger...
I0425 21:15:45.678903 136514901002048 checkpointing.py:233] Creating checkpoint manager with ocdbt=False and zarr3=True
I0425 21:15:45.678947 136514901002048 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0425 21:15:45.679313 136514901002048 base_pytree_checkpoint_handler.py:441] Created BasePyTreeCheckpointHandler: use_ocdbt=False, use_zarr3=True, pytree_metadata_options=PyTreeMetadataOptions(support_rich_types=False), array_metadata_store=<orbax.checkpoint._src.metadata.array_metadata_store.Store object at 0x7c283e5e5280>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0425 21:15:48.876293 136514901002048 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0425 21:15:48.876527 136514901002048 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 0x7c13b8563c50>}, handler_registry=None
I0425 21:15:48.876765 136514901002048 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c13b8563c50>` 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 21:15:48.876816 136514901002048 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c13b8562d80>` 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 21:15:48.876852 136514901002048 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 0x7c13b8563c50>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c13b8563c50>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c13b8562d80>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c13b8562d80>}).
I0425 21:15:48.877421 136514901002048 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0425 21:15:48.877547 136514901002048 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7c13b8383560> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0425 21:15:49.832983 136514901002048 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_15_ocdbt_false/checkpoints
I0425 21:15:49.852484 136514901002048 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_15_ocdbt_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7c13b8565c40>
I0425 21:15:49.852611 136514901002048 checkpointing.py:301] Checkpoint manager created!
I0425 21:15:50.792632 136514901002048 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0425 21:15:50.792749 136514901002048 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 21:15:51.173627 136514901002048 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0425 21:15:51.173725 136514901002048 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 21:15:51.190255 136514901002048 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0425 21:15:51.190315 136514901002048 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 21:15:51.214028 136514901002048 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 21:15:51.214107 136514901002048 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 21:15:51.230842 136514901002048 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 21:15:51.230901 136514901002048 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 21:15:51.247512 136514901002048 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 21:15:51.247581 136514901002048 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 21:15:51.272720 136514901002048 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0425 21:15:51.272793 136514901002048 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 21:15:51.293864 136514901002048 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0425 21:15:51.293935 136514901002048 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0425 21:15:51.505923 136514901002048 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0425 21:15:51.506035 136514901002048 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0425 21:15:52.968716 136514901002048 maxtext_utils.py:1668] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0425 21:15:52.968842 136514901002048 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 21:15:52.968898 136514901002048 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 21:15:52.968956 136514901002048 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 21:15:52.969008 136514901002048 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 21:15:52.969046 136514901002048 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 21:15:52.969121 136514901002048 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 21:15:52.969180 136514901002048 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 21:15:52.969220 136514901002048 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 21:15:52.969255 136514901002048 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 21:15:52.969301 136514901002048 maxtext_utils.py:1668] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0425 21:15:52.969347 136514901002048 maxtext_utils.py:1668] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0425 21:15:53.468876 136514901002048 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0425 21:15:53.468971 136514901002048 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0425 21:15:53.484484 136514901002048 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0425 21:15:53.484544 136514901002048 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0425 21:16:04.578313 136514901002048 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 21:16:04.579169 136514901002048 metric_logger.py:301] number parameters: 1.104 billion
I0425 21:16:16.445764 136514901002048 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0425 21:16:16.600409 136514901002048 checkpointing.py:776] Waited 0.15462040901184082 seconds for step 0 to finish before starting checkpointing.
I0425 21:16:16.603203 136514901002048 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 21:16:16.604893 136514901002048 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0425 21:16:16.606258 136514901002048 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_15_ocdbt_false/checkpoints/0.
I0425 21:16:16.929149 136514901002048 signaling_client.py:364] Using JaxDistributedSignalingClient
I0425 21:16:16.930125 136514901002048 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0425 21:16:16.930183 136514901002048 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 21:16:17.208694 136514901002048 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.279644s
I0425 21:16:17.208870 136514901002048 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.409 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28516721725463867 s) (per-host)
I0425 21:16:17.208924 136514901002048 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.285231s (batch_requests_ready=0.002210s, total_serialization_initiated=0.282948s, others=0.000072s)
I0425 21:16:17.209018 136514901002048 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.289470s (all_items=0.000019s, per_item={'items': '0.00001884'}, temp_paths=0.289452)
I0425 21:16:17.209908 136514901002048 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.60 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_15_ocdbt_false/checkpoints/0.
I0425 21:16:17.210249 136387192063744 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 21:36:17.210210
I0425 21:16:17.632989 136514901002048 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 21:16:17.633388 136385562384128 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 21:16:17.633566 136514901002048 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_15_ocdbt_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777151776.603184, 'wait_for_prev_duration_secs': 6.103515625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777151776.604931, 'checkpointer_blocking_duration_secs': 0.6054871082305908, 'get_old_steps_start_time': 1777151777.210444, 'get_old_steps_duration_secs': 3.0517578125e-05, 'checkpoint_manager_blocking_start_time': 1777151776.600965, 'checkpoint_manager_blocking_duration_secs': 1.0325515270233154}
I0425 21:16:17.633680 136514901002048 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0425 21:16:17.633730 136514901002048 max_utils.py:750]
Memstats: After params initialized:
I0425 21:16:17.633787 136514901002048 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0425 21:16:17.633820 136514901002048 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0425 21:16:17.633847 136514901002048 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0425 21:16:17.633873 136514901002048 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0425 21:16:17.946969 136514901002048 metric_logger.py:196] completed step: 0, seconds: 11.866, TFLOP/s/device: 1.145, Tokens/s/device: 172.587, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0425 21:16:18.131391 136514901002048 metric_logger.py:196] completed step: 1, seconds: 1.500, TFLOP/s/device: 9.060, Tokens/s/device: 1365.580, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0425 21:16:18.543820 136514901002048 metric_logger.py:196] completed step: 2, seconds: 0.037, TFLOP/s/device: 364.491, Tokens/s/device: 54940.043, total_weights: 65536, loss: 10.560, lm_loss: 10.560, perplexity: 38547.719
I0425 21:16:18.691293 136514901002048 metric_logger.py:196] completed step: 3, seconds: 0.413, TFLOP/s/device: 32.937, Tokens/s/device: 4964.692, total_weights: 65536, loss: 9.980, lm_loss: 9.980, perplexity: 21590.506
I0425 21:16:18.695762 136514901002048 checkpointing.py:772] Waiting for step 5 to finish before checkpoint...
I0425 21:16:18.985581 136514901002048 checkpointing.py:776] Waited 0.2897775173187256 seconds for step 5 to finish before starting checkpointing.
I0425 21:16:18.988617 136514901002048 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 21:16:20.779419 2767 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0425 21:16:22.776423 136385570776832 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_15_ocdbt_false/checkpoints/0/items/array_metadatas/process_6
I0425 21:16:28.086271 136387192063744 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 141.509 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 11.162530899047852 s) (per-host)
I0425 21:16:28.086390 136387192063744 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 10.876013s.
I0425 21:16:34.930401 136387192063744 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 17.720002s.
I0425 21:16:34.930723 136385562384128 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 21:16:34.930835 136385562384128 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 21:16:34.930887 136385562384128 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0425 21:16:34.932936 136385562384128 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0425 21:16:34.933153 136514901002048 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 21:16:34.933292 136514901002048 checkpoint_manager.py:1452] Waiting for previous save to complete took 15.944673 seconds. If this number is high, consider checkpointing less frequently.
I0425 21:16:34.935591 136514901002048 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 5
I0425 21:16:34.938361 136514901002048 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_15_ocdbt_false/checkpoints/5.
I0425 21:16:35.429058 136514901002048 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0425 21:16:35.429169 136514901002048 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 21:16:35.463154 136514901002048 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.035369s
I0425 21:16:35.463330 136514901002048 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 39.723 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.0388331413269043 s) (per-host)
I0425 21:16:35.463380 136514901002048 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.038894s (batch_requests_ready=0.001791s, total_serialization_initiated=0.037034s, others=0.000068s)
I0425 21:16:35.463474 136514901002048 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.044042s (all_items=0.000017s, per_item={'items': '0.00001669'}, temp_paths=0.044025)
I0425 21:16:35.464229 136514901002048 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.53 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_15_ocdbt_false/checkpoints/5.
I0425 21:16:35.464534 136385562384128 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 21:36:35.464501
I0425 21:16:35.473432 136514901002048 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=5] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 21:16:35.473735 136387089135360 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 21:16:35.473902 136514901002048 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_15_ocdbt_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777151778.9885864, 'wait_for_prev_duration_secs': 15.94467306137085, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777151794.9356294, 'checkpointer_blocking_duration_secs': 0.5290482044219971, 'get_old_steps_start_time': 1777151795.4647012, 'get_old_steps_duration_secs': 3.123283386230469e-05, 'checkpoint_manager_blocking_start_time': 1777151778.9861858, 'checkpoint_manager_blocking_duration_secs': 16.487680196762085}
I0425 21:16:35.474007 136514901002048 checkpointing.py:408] Started an asynchronous checkpoint save for step 5
I0425 21:16:35.474817 136514901002048 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.737, Tokens/s/device: 13375.393, total_weights: 65536, loss: 9.460, lm_loss: 9.460, perplexity: 12840.355
I0425 21:16:35.480896 136514901002048 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.269, Tokens/s/device: 13907.847, total_weights: 65536, loss: 8.959, lm_loss: 8.959, perplexity: 7776.467
I0425 21:16:40.540903 136385570776832 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_15_ocdbt_false/checkpoints/5/items/array_metadatas/process_6
I0425 21:16:46.214550 136385562384128 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 146.394 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 10.790013790130615 s) (per-host)
I0425 21:16:46.214677 136385562384128 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 10.750032s.
I0425 21:16:52.592292 136385562384128 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 17.127621s.
I0425 21:16:52.592545 136387089135360 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 21:16:52.592662 136387089135360 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 21:16:52.592711 136387089135360 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is syncing with other hosts...
I0425 21:16:52.594338 136387089135360 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is done on all hosts.
I0425 21:16:58.608587 136514901002048 metric_logger.py:196] completed step: 6, seconds: 16.784, TFLOP/s/device: 0.810, Tokens/s/device: 122.024, total_weights: 65536, loss: 8.469, lm_loss: 8.469, perplexity: 4762.740
I0425 21:16:58.756074 136514901002048 metric_logger.py:196] completed step: 7, seconds: 22.982, TFLOP/s/device: 0.591, Tokens/s/device: 89.115, total_weights: 65536, loss: 8.003, lm_loss: 8.003, perplexity: 2989.151
I0425 21:16:58.903558 136514901002048 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.155, Tokens/s/device: 13438.408, total_weights: 65536, loss: 7.572, lm_loss: 7.572, perplexity: 1943.098
I0425 21:16:59.050033 136514901002048 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0425 21:16:59.050773 136514901002048 checkpointing.py:776] Waited 0.0007646083831787109 seconds for step 9 to finish before starting checkpointing.
I0425 21:16:59.052915 136514901002048 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 21:16:59.054825 136514901002048 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0425 21:16:59.056254 136514901002048 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_15_ocdbt_false/checkpoints/9.
I0425 21:16:59.554264 136514901002048 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0425 21:16:59.554362 136514901002048 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 21:16:59.588704 136514901002048 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.035360s
I0425 21:16:59.588868 136514901002048 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 39.706 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.038849830627441406 s) (per-host)
I0425 21:16:59.588917 136514901002048 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.038908s (batch_requests_ready=0.001661s, total_serialization_initiated=0.037184s, others=0.000062s)
I0425 21:16:59.589016 136514901002048 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.043115s (all_items=0.000011s, per_item={'items': '0.00001121'}, temp_paths=0.043104)
I0425 21:16:59.589774 136514901002048 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.53 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260425_201227/linen_xpk_test_pipeline_scan_nnx_20260425_201227_15_ocdbt_false/checkpoints/9.
I0425 21:16:59.590115 136361520658176 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 21:36:59.590063
I0425 21:16:59.592135 136514901002048 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 21:16:59.592437 136361378047744 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 21:16:59.592603 136514901002048 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_15_ocdbt_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777151819.0528848, 'wait_for_prev_duration_secs': 7.367134094238281e-05, 'time_between_consecutive_saves_sec': 6.45850396156311, 'checkpointer_blocking_start_time': 1777151819.054865, 'checkpointer_blocking_duration_secs': 0.5353999137878418, 'get_old_steps_start_time': 1777151819.5902839, 'get_old_steps_duration_secs': 2.8848648071289062e-05, 'checkpoint_manager_blocking_start_time': 1777151819.051011, 'checkpoint_manager_blocking_duration_secs': 0.5415570735931396}
I0425 21:16:59.592730 136514901002048 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0425 21:16:59.592783 136514901002048 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 21:17:04.717293 136385612740352 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_15_ocdbt_false/checkpoints/9/items/array_metadatas/process_6
I0425 21:17:10.031537 136361520658176 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 150.704 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 10.481471061706543 s) (per-host)
I0425 21:17:10.031660 136361520658176 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 10.441427s.
I0425 21:17:16.095312 136361520658176 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 16.505060s.
I0425 21:17:16.095629 136361378047744 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 21:17:16.095748 136361378047744 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 21:17:16.095797 136361378047744 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0425 21:17:16.097321 136361378047744 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0425 21:17:16.097503 136514901002048 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 21:17:16.097650 136514901002048 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 21:17:16.098665 136514901002048 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.186, Tokens/s/device: 13895.297, 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 21:17:29 UTC 2026
EXIT_CODE=0