XPK Start: Thu Apr 23 22:38:34 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.
2026-04-23 22:38:59.724937: 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)
I0423 22:38:59.935961 139170030212928 max_utils.py:273] Attempting to initialize the jax distributed system...
I0423 22:39:08.977003 139170030212928 distributed.py:149] Starting JAX distributed service on [::]:8482
I0423 22:39:08.979371 139170030212928 distributed.py:172] Connecting to JAX distributed service on mt-16-zarr3-false-sv-dor83-slice-job-0-0.mt-16-zarr3-false-sv-dor83:8482
I0423 22:39:09.949151 139170030212928 max_utils.py:284] Jax distributed system initialized!
I0423 22:39:15.970059 139170030212928 max_utils.py:800] System Information: Jax Version: 0.9.2
I0423 22:39:15.970185 139170030212928 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0423 22:39:15.970226 139170030212928 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
I0423 22:39:15.970263 139170030212928 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0423 22:39:16.667376 139170030212928 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0423 22:39:16.667661 139170030212928 checkpointing.py:677] Setting up checkpoint logger...
I0423 22:39:16.667715 139170030212928 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=False
I0423 22:39:16.667758 139170030212928 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0423 22:39:16.668115 139170030212928 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 0x7e925493f230>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0423 22:39:20.097965 139170030212928 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0423 22:39:20.098221 139170030212928 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 0x7e7dfc47b710>}, handler_registry=None
I0423 22:39:20.098468 139170030212928 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e7dfc47b710>` 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`.
I0423 22:39:20.098518 139170030212928 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e7dfc47f8f0>` 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`.
I0423 22:39:20.098554 139170030212928 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 0x7e7dfc47b710>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e7dfc47b710>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e7dfc47f8f0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e7dfc47f8f0>}).
I0423 22:39:20.098877 139170030212928 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0423 22:39:20.098949 139170030212928 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7e7dfc297560> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0423 22:39:20.790459 139170030212928 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints
I0423 22:39:21.222136 139170030212928 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_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7e7dfc47df10>
I0423 22:39:21.222316 139170030212928 checkpointing.py:301] Checkpoint manager created!
I0423 22:39:22.149722 139170030212928 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0423 22:39:22.149840 139170030212928 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 22:39:22.550270 139170030212928 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0423 22:39:22.550366 139170030212928 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 22:39:22.567018 139170030212928 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0423 22:39:22.567078 139170030212928 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 22:39:22.591324 139170030212928 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 22:39:22.591392 139170030212928 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 22:39:22.608030 139170030212928 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 22:39:22.608104 139170030212928 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 22:39:22.624706 139170030212928 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 22:39:22.624770 139170030212928 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 22:39:22.649718 139170030212928 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0423 22:39:22.649808 139170030212928 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 22:39:22.670941 139170030212928 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0423 22:39:22.671017 139170030212928 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0423 22:39:22.885369 139170030212928 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0423 22:39:22.885487 139170030212928 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0423 22:39:24.311149 139170030212928 maxtext_utils.py:1668] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0423 22:39:24.311279 139170030212928 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)
I0423 22:39:24.311334 139170030212928 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)
I0423 22:39:24.311393 139170030212928 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')
I0423 22:39:24.311448 139170030212928 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)
I0423 22:39:24.311488 139170030212928 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)
I0423 22:39:24.311541 139170030212928 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)
I0423 22:39:24.311598 139170030212928 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')
I0423 22:39:24.311639 139170030212928 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)
I0423 22:39:24.311676 139170030212928 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)
I0423 22:39:24.311726 139170030212928 maxtext_utils.py:1668] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0423 22:39:24.311774 139170030212928 maxtext_utils.py:1668] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0423 22:39:24.809177 139170030212928 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0423 22:39:24.809274 139170030212928 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0423 22:39:24.824835 139170030212928 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0423 22:39:24.824897 139170030212928 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0423 22:39:35.781287 139170030212928 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.
I0423 22:39:35.782065 139170030212928 metric_logger.py:301] number parameters: 1.104 billion
I0423 22:39:47.790482 139170030212928 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0423 22:39:47.944610 139170030212928 checkpointing.py:776] Waited 0.15410685539245605 seconds for step 0 to finish before starting checkpointing.
I0423 22:39:47.947420 139170030212928 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 22:39:47.949211 139170030212928 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0423 22:39:47.950584 139170030212928 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints/0.
I0423 22:39:48.696928 139170030212928 signaling_client.py:364] Using JaxDistributedSignalingClient
I0423 22:39:48.697938 139170030212928 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0423 22:39:48.697994 139170030212928 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
I0423 22:39:48.970346 139170030212928 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.273491s
I0423 22:39:48.970526 139170030212928 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.524 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.27922487258911133 s) (per-host)
I0423 22:39:48.970589 139170030212928 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.279295s (batch_requests_ready=0.002384s, total_serialization_initiated=0.276832s, others=0.000079s)
I0423 22:39:48.970708 139170030212928 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.283524s (all_items=0.000018s, per_item={'items': '0.00001812'}, temp_paths=0.283505)
I0423 22:39:48.971552 139170030212928 event_tracking.py:125] [process=6] [async] Finished blocking save in 1.02 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints/0.
I0423 22:39:48.971864 139040976271104 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 22:59:48.971826
I0423 22:39:48.986732 139170030212928 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 22:39:48.987057 139036681303808 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0423 22:39:48.987244 139170030212928 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776983987.9473994, 'wait_for_prev_duration_secs': 6.961822509765625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776983987.9492507, 'checkpointer_blocking_duration_secs': 1.0227689743041992, 'get_old_steps_start_time': 1776983988.972047, 'get_old_steps_duration_secs': 3.075599670410156e-05, 'checkpoint_manager_blocking_start_time': 1776983987.9451563, 'checkpoint_manager_blocking_duration_secs': 1.0420470237731934}
I0423 22:39:48.987369 139170030212928 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0423 22:39:48.987422 139170030212928 max_utils.py:750]
Memstats: After params initialized:
I0423 22:39:48.987474 139170030212928 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0423 22:39:48.987505 139170030212928 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0423 22:39:48.987532 139170030212928 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0423 22:39:48.987557 139170030212928 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0423 22:39:49.298666 139170030212928 metric_logger.py:196] completed step: 0, seconds: 12.008, TFLOP/s/device: 1.131, Tokens/s/device: 170.549, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0423 22:39:49.485812 139170030212928 metric_logger.py:196] completed step: 1, seconds: 1.507, TFLOP/s/device: 9.019, Tokens/s/device: 1359.373, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0423 22:39:49.877298 139170030212928 metric_logger.py:196] completed step: 2, seconds: 0.040, TFLOP/s/device: 339.670, Tokens/s/device: 51198.720, total_weights: 65536, loss: 10.560, lm_loss: 10.560, perplexity: 38547.719
I0423 22:39:50.024692 139170030212928 metric_logger.py:196] completed step: 3, seconds: 0.392, TFLOP/s/device: 34.694, Tokens/s/device: 5229.399, total_weights: 65536, loss: 9.980, lm_loss: 9.980, perplexity: 21590.506
I0423 22:39:50.029144 139170030212928 checkpointing.py:772] Waiting for step 5 to finish before checkpoint...
I0423 22:39:50.318436 139170030212928 checkpointing.py:776] Waited 0.2892591953277588 seconds for step 5 to finish before starting checkpointing.
I0423 22:39:50.321571 139170030212928 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 22:39:51.919877 2826 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0423 22:39:53.954561 139038235051776 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_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints/0/items/array_metadatas/process_6
I0423 22:40:29.919499 139040976271104 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 38.314 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.22815561294556 s) (per-host)
I0423 22:40:29.919620 139040976271104 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 40.947641s.
I0423 22:40:38.650126 139040976271104 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 49.678108s.
I0423 22:40:38.650418 139036681303808 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 22:40:38.650546 139036681303808 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0423 22:40:38.650595 139036681303808 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0423 22:40:38.653165 139036681303808 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0423 22:40:38.653369 139170030212928 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.
W0423 22:40:38.653503 139170030212928 checkpoint_manager.py:1452] Waiting for previous save to complete took 48.331932 seconds. If this number is high, consider checkpointing less frequently.
I0423 22:40:38.655489 139170030212928 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 5
I0423 22:40:38.658121 139170030212928 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints/5.
I0423 22:40:38.955751 139170030212928 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0423 22:40:38.955845 139170030212928 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
I0423 22:40:38.995352 139170030212928 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.040606s
I0423 22:40:38.995532 139170030212928 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 34.880 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04422473907470703 s) (per-host)
I0423 22:40:38.995585 139170030212928 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.044288s (batch_requests_ready=0.001932s, total_serialization_initiated=0.042285s, others=0.000070s)
I0423 22:40:38.995680 139170030212928 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.049287s (all_items=0.000016s, per_item={'items': '0.00001645'}, temp_paths=0.049270)
I0423 22:40:38.996442 139170030212928 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.34 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints/5.
I0423 22:40:38.996756 139036681303808 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 23:00:38.996721
I0423 22:40:39.000033 139170030212928 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=5] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 22:40:39.000342 139040967878400 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0423 22:40:39.000507 139170030212928 standard_logger.py:34] {'step': 5, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776983990.3215408, 'wait_for_prev_duration_secs': 48.331931829452515, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776984038.6555283, 'checkpointer_blocking_duration_secs': 0.3413715362548828, 'get_old_steps_start_time': 1776984038.9969237, 'get_old_steps_duration_secs': 3.123283386230469e-05, 'checkpoint_manager_blocking_start_time': 1776983990.3190303, 'checkpoint_manager_blocking_duration_secs': 48.68144178390503}
I0423 22:40:39.000626 139170030212928 checkpointing.py:408] Started an asynchronous checkpoint save for step 5
I0423 22:40:39.001434 139170030212928 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.730, Tokens/s/device: 13374.345, total_weights: 65536, loss: 9.460, lm_loss: 9.460, perplexity: 12840.355
I0423 22:40:39.007640 139170030212928 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.304, Tokens/s/device: 13913.043, total_weights: 65536, loss: 8.959, lm_loss: 8.959, perplexity: 7776.467
I0423 22:40:44.373516 139019400754944 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_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints/5/items/array_metadatas/process_6
I0423 22:41:01.966815 139170030212928 metric_logger.py:196] completed step: 6, seconds: 48.977, TFLOP/s/device: 0.277, Tokens/s/device: 41.816, total_weights: 65536, loss: 8.469, lm_loss: 8.469, perplexity: 4762.740
I0423 22:41:02.114240 139170030212928 metric_logger.py:196] completed step: 7, seconds: 22.813, TFLOP/s/device: 0.596, Tokens/s/device: 89.773, total_weights: 65536, loss: 8.003, lm_loss: 8.003, perplexity: 2989.151
I0423 22:41:02.261772 139170030212928 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.181, Tokens/s/device: 13442.378, total_weights: 65536, loss: 7.572, lm_loss: 7.572, perplexity: 1943.098
I0423 22:41:02.408374 139170030212928 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0423 22:41:02.409037 139170030212928 checkpointing.py:776] Waited 0.0006844997406005859 seconds for step 9 to finish before starting checkpointing.
I0423 22:41:02.411154 139170030212928 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=5][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 22:41:21.176446 139036681303808 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.409 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.225120067596436 s) (per-host)
I0423 22:41:21.176552 139036681303808 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.179690s.
I0423 22:41:28.401466 139036681303808 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 49.404587s.
I0423 22:41:28.401777 139040967878400 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 22:41:28.401911 139040967878400 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0423 22:41:28.401970 139040967878400 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is syncing with other hosts...
I0423 22:41:28.403754 139040967878400 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is done on all hosts.
I0423 22:41:28.403925 139170030212928 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.
W0423 22:41:28.404056 139170030212928 checkpoint_manager.py:1452] Waiting for previous save to complete took 25.992907 seconds. If this number is high, consider checkpointing less frequently.
I0423 22:41:28.405582 139170030212928 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0423 22:41:28.407819 139170030212928 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints/9.
I0423 22:41:28.690614 139170030212928 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0423 22:41:28.690706 139170030212928 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
I0423 22:41:28.716787 139170030212928 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.026968s
I0423 22:41:28.716960 139170030212928 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 50.412 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.030599355697631836 s) (per-host)
I0423 22:41:28.717011 139170030212928 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.030661s (batch_requests_ready=0.001808s, total_serialization_initiated=0.028786s, others=0.000066s)
I0423 22:41:28.717122 139170030212928 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.034897s (all_items=0.000010s, per_item={'items': '0.00000978'}, temp_paths=0.034887)
I0423 22:41:28.717853 139170030212928 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.31 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints/9.
I0423 22:41:28.718167 139040967878400 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 23:01:28.718134
I0423 22:41:28.731751 139170030212928 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 22:41:28.732012 139019400754944 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0423 22:41:28.732172 139170030212928 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776984062.411123, 'wait_for_prev_duration_secs': 25.99290657043457, 'time_between_consecutive_saves_sec': 23.757917642593384, 'checkpointer_blocking_start_time': 1776984088.4056206, 'checkpointer_blocking_duration_secs': 0.3127005100250244, 'get_old_steps_start_time': 1776984088.7183406, 'get_old_steps_duration_secs': 2.4080276489257812e-05, 'checkpoint_manager_blocking_start_time': 1776984062.4093082, 'checkpoint_manager_blocking_duration_secs': 26.322829961776733}
I0423 22:41:28.732285 139170030212928 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0423 22:41:28.732328 139170030212928 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 22:41:34.687744 139039335102208 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_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_false/checkpoints/9/items/array_metadatas/process_6
I0423 22:42:10.815810 139040967878400 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.494 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.129403591156006 s) (per-host)
I0423 22:42:10.815929 139040967878400 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.097644s.
I0423 22:42:17.523888 139040967878400 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 48.805582s.
I0423 22:42:17.524216 139019400754944 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 22:42:17.524346 139019400754944 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0423 22:42:17.524388 139019400754944 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0423 22:42:17.525819 139019400754944 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0423 22:42:17.525999 139170030212928 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.
I0423 22:42:17.526178 139170030212928 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 22:42:17.527129 139170030212928 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.196, Tokens/s/device: 13896.711, 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: Thu Apr 23 22:42:29 UTC 2026
EXIT_CODE=0