XPK Start: Fri Apr 24 20:12:21 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.
2026-04-24 20:12:45.805715: 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)
I0424 20:12:46.016353 137631943055168 max_utils.py:273] Attempting to initialize the jax distributed system...
I0424 20:12:55.057459 137631943055168 distributed.py:149] Starting JAX distributed service on [::]:8482
I0424 20:12:55.059745 137631943055168 distributed.py:172] Connecting to JAX distributed service on mt-02-synthetic-ob3e2-slice-job-0-0.mt-02-synthetic-ob3e2:8482
I0424 20:12:56.560842 137631943055168 max_utils.py:284] Jax distributed system initialized!
I0424 20:13:02.760979 137631943055168 max_utils.py:800] System Information: Jax Version: 0.9.2
I0424 20:13:02.761083 137631943055168 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0424 20:13:02.761124 137631943055168 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
I0424 20:13:02.761159 137631943055168 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0424 20:13:03.454133 137631943055168 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0424 20:13:03.454406 137631943055168 checkpointing.py:677] Setting up checkpoint logger...
I0424 20:13:03.454459 137631943055168 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0424 20:13:03.454503 137631943055168 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0424 20:13:03.454853 137631943055168 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 0x7d2c4e31ff20>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0424 20:13:06.298116 137631943055168 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0424 20:13:06.298353 137631943055168 checkpoint_manager.py:708] [process=5][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d17fc0c7b00>}, handler_registry=None
I0424 20:13:06.298581 137631943055168 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d17fc0c7b00>` 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`.
I0424 20:13:06.298631 137631943055168 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d17fc0c8c80>` 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`.
I0424 20:13:06.298679 137631943055168 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 0x7d17fc0c7b00>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d17fc0c7b00>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d17fc0c8c80>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d17fc0c8c80>}).
I0424 20:13:06.298992 137631943055168 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0424 20:13:06.299061 137631943055168 async_checkpointer.py:192] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d17244f32e0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0424 20:13:07.652524 137631943055168 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_02_synthetic/checkpoints
I0424 20:13:07.745362 137631943055168 checkpoint_manager.py:929] [process=5][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d17fc0c9580>
I0424 20:13:07.745532 137631943055168 checkpointing.py:301] Checkpoint manager created!
I0424 20:13:08.671424 137631943055168 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0424 20:13:08.671540 137631943055168 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 20:13:09.055703 137631943055168 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0424 20:13:09.055797 137631943055168 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 20:13:09.072284 137631943055168 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0424 20:13:09.072351 137631943055168 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 20:13:09.095939 137631943055168 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 20:13:09.096007 137631943055168 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 20:13:09.112429 137631943055168 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 20:13:09.112485 137631943055168 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 20:13:09.128829 137631943055168 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 20:13:09.128889 137631943055168 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 20:13:09.153554 137631943055168 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0424 20:13:09.153625 137631943055168 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 20:13:09.174671 137631943055168 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0424 20:13:09.174738 137631943055168 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0424 20:13:09.390486 137631943055168 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0424 20:13:09.390598 137631943055168 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0424 20:13:10.807153 137631943055168 maxtext_utils.py:1668] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0424 20:13:10.807280 137631943055168 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)
I0424 20:13:10.807334 137631943055168 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)
I0424 20:13:10.807391 137631943055168 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')
I0424 20:13:10.807443 137631943055168 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)
I0424 20:13:10.807482 137631943055168 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)
I0424 20:13:10.807533 137631943055168 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)
I0424 20:13:10.807585 137631943055168 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')
I0424 20:13:10.807624 137631943055168 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)
I0424 20:13:10.807675 137631943055168 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)
I0424 20:13:10.807727 137631943055168 maxtext_utils.py:1668] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0424 20:13:10.807776 137631943055168 maxtext_utils.py:1668] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0424 20:13:11.301974 137631943055168 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0424 20:13:11.302065 137631943055168 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0424 20:13:11.317683 137631943055168 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0424 20:13:11.317741 137631943055168 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0424 20:13:22.310386 137631943055168 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.
I0424 20:13:22.311251 137631943055168 metric_logger.py:301] number parameters: 1.104 billion
I0424 20:13:33.961592 137631943055168 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0424 20:13:34.115703 137631943055168 checkpointing.py:776] Waited 0.15407752990722656 seconds for step 0 to finish before starting checkpointing.
I0424 20:13:34.118166 137631943055168 checkpoint_manager.py:2009] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0424 20:13:34.119677 137631943055168 checkpoint_manager.py:1512] [process=5] Saving checkpoint at step 0
I0424 20:13:34.121063 137631943055168 event_tracking.py:70] [process=5] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_02_synthetic/checkpoints/0.
I0424 20:13:34.844926 137631943055168 signaling_client.py:364] Using JaxDistributedSignalingClient
I0424 20:13:34.845956 137631943055168 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0424 20:13:34.846015 137631943055168 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
I0424 20:13:35.122713 137631943055168 base_pytree_checkpoint_handler.py:154] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.277860s
I0424 20:13:35.122884 137631943055168 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/blocking_gbytes_per_sec: 5.444 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28337812423706055 s) (per-host)
I0424 20:13:35.122938 137631943055168 base_pytree_checkpoint_handler.py:768] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.283442s (batch_requests_ready=0.002337s, total_serialization_initiated=0.281034s, others=0.000071s)
I0424 20:13:35.123036 137631943055168 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.287493s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.287475)
I0424 20:13:35.123845 137631943055168 event_tracking.py:125] [process=5] [async] Finished blocking save in 1.00 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_02_synthetic/checkpoints/0.
I0424 20:13:35.124192 137503631136512 async_checkpointer.py:76] [process=5][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 20:33:35.124154
I0424 20:13:35.132348 137631943055168 checkpoint_manager.py:1560] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 20:13:35.132630 137503052101376 async_checkpointer.py:280] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0424 20:13:35.132807 137631943055168 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777061614.1181493, 'wait_for_prev_duration_secs': 5.817413330078125e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777061614.1197145, 'checkpointer_blocking_duration_secs': 1.004636287689209, 'get_old_steps_start_time': 1777061615.124376, 'get_old_steps_duration_secs': 2.8848648071289062e-05, 'checkpoint_manager_blocking_start_time': 1777061614.1162531, 'checkpoint_manager_blocking_duration_secs': 1.0165166854858398}
I0424 20:13:35.132913 137631943055168 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0424 20:13:35.132963 137631943055168 max_utils.py:750]
Memstats: After params initialized:
I0424 20:13:35.133013 137631943055168 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_18(process=5,(2,4,0,0))
I0424 20:13:35.133045 137631943055168 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_19(process=5,(3,4,0,0))
I0424 20:13:35.133072 137631943055168 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_22(process=5,(2,5,0,0))
I0424 20:13:35.133099 137631943055168 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_23(process=5,(3,5,0,0))
I0424 20:13:35.447300 137631943055168 metric_logger.py:196] completed step: 0, seconds: 11.650, TFLOP/s/device: 1.166, Tokens/s/device: 175.791, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0424 20:13:35.619549 137631943055168 metric_logger.py:196] completed step: 1, seconds: 1.484, TFLOP/s/device: 9.155, Tokens/s/device: 1379.869, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0424 20:13:36.022877 137631943055168 metric_logger.py:196] completed step: 2, seconds: 0.025, TFLOP/s/device: 544.400, Tokens/s/device: 82057.857, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0424 20:13:36.170197 137631943055168 metric_logger.py:196] completed step: 3, seconds: 0.404, TFLOP/s/device: 33.654, Tokens/s/device: 5072.747, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0424 20:13:36.465084 137631943055168 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.892, Tokens/s/device: 13398.757, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0424 20:13:36.470950 137631943055168 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.277, Tokens/s/device: 13908.980, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0424 20:13:38.110148 2804 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0424 20:13:40.974980 137503060494080 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_02_synthetic/checkpoints/0/items/array_metadatas/process_5
I0424 20:13:57.863860 137631943055168 metric_logger.py:196] completed step: 6, seconds: 0.295, TFLOP/s/device: 45.999, Tokens/s/device: 6933.489, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0424 20:13:58.011018 137631943055168 metric_logger.py:196] completed step: 7, seconds: 21.246, TFLOP/s/device: 0.640, Tokens/s/device: 96.394, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0424 20:13:58.158608 137631943055168 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.180, Tokens/s/device: 13442.113, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0424 20:13:58.305295 137631943055168 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0424 20:13:58.305982 137631943055168 checkpointing.py:776] Waited 0.0007064342498779297 seconds for step 9 to finish before starting checkpointing.
I0424 20:13:58.308026 137631943055168 checkpoint_manager.py:2020] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 20:14:10.952435 137503631136512 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/gbytes_per_sec: 43.741 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.112892627716064 s) (per-host)
I0424 20:14:10.952559 137503631136512 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 35.828252s.
I0424 20:14:21.060597 137503631136512 async_checkpointer.py:160] [process=5][thread=async_save] Background save thread done. Time taken: 45.936275s.
I0424 20:14:21.060857 137503052101376 async_checkpointer.py:288] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 20:14:21.060915 137503052101376 async_checkpointer.py:298] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0424 20:14:21.060971 137503052101376 checkpoint_manager.py:2137] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0424 20:14:21.062633 137503052101376 checkpoint_manager.py:2146] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0424 20:14:21.062805 137631943055168 checkpoint_manager.py:2032] [process=5][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0424 20:14:21.062934 137631943055168 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.754909 seconds. If this number is high, consider checkpointing less frequently.
I0424 20:14:21.064796 137631943055168 checkpoint_manager.py:1512] [process=5] Saving checkpoint at step 9
I0424 20:14:21.066802 137631943055168 event_tracking.py:70] [process=5] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_02_synthetic/checkpoints/9.
I0424 20:14:21.353697 137631943055168 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0424 20:14:21.353786 137631943055168 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
I0424 20:14:21.391406 137631943055168 base_pytree_checkpoint_handler.py:154] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.038781s
I0424 20:14:21.391574 137631943055168 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/blocking_gbytes_per_sec: 35.925 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.042939186096191406 s) (per-host)
I0424 20:14:21.391626 137631943055168 base_pytree_checkpoint_handler.py:768] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.043000s (batch_requests_ready=0.002530s, total_serialization_initiated=0.040404s, others=0.000067s)
I0424 20:14:21.391727 137631943055168 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.046971s (all_items=0.000015s, per_item={'items': '0.00001526'}, temp_paths=0.046956)
I0424 20:14:21.392364 137631943055168 event_tracking.py:125] [process=5] [async] Finished blocking save in 0.33 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_02_synthetic/checkpoints/9.
I0424 20:14:21.392706 137503052101376 async_checkpointer.py:76] [process=5][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 20:34:21.392670
I0424 20:14:21.395504 137631943055168 checkpoint_manager.py:1560] [process=5][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 20:14:21.395797 137503060494080 async_checkpointer.py:280] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0424 20:14:21.395957 137631943055168 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777061638.3079958, 'wait_for_prev_duration_secs': 22.75490927696228, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777061661.0648372, 'checkpointer_blocking_duration_secs': 0.32800889015197754, 'get_old_steps_start_time': 1777061661.39287, 'get_old_steps_duration_secs': 2.8848648071289062e-05, 'checkpoint_manager_blocking_start_time': 1777061638.3062174, 'checkpoint_manager_blocking_duration_secs': 23.089706420898438}
I0424 20:14:21.396075 137631943055168 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0424 20:14:21.396121 137631943055168 checkpoint_manager.py:2020] [process=5][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 20:14:26.598932 137478304409344 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_02_synthetic/checkpoints/9/items/array_metadatas/process_5
I0424 20:15:03.776438 137503052101376 base_pytree_checkpoint_handler.py:130] [process=5] /jax/orbax/write/gbytes_per_sec: 37.230 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.427764654159546 s) (per-host)
I0424 20:15:03.776561 137503052101376 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.383744s.
I0424 20:15:10.763493 137503052101376 async_checkpointer.py:160] [process=5][thread=async_save] Background save thread done. Time taken: 49.370663s.
I0424 20:15:10.763792 137503060494080 async_checkpointer.py:288] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 20:15:10.763910 137503060494080 async_checkpointer.py:298] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0424 20:15:10.763958 137503060494080 checkpoint_manager.py:2137] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0424 20:15:10.765418 137503060494080 checkpoint_manager.py:2146] [process=5][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0424 20:15:10.765601 137631943055168 checkpoint_manager.py:2032] [process=5][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0424 20:15:10.765768 137631943055168 checkpoint_manager.py:2009] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0424 20:15:10.766812 137631943055168 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.223, Tokens/s/device: 13900.861, total_weights: 65536, loss: 8.179, lm_loss: 8.179, perplexity: 3564.635
Per train step:
Total TFLOPs: 13.59
split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Fri Apr 24 20:15:21 UTC 2026
EXIT_CODE=0