MaxView

XPK Start: Wed Apr 22 22:01:58 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
`rope_parameters`'s factor field must be a float >= 1, got 40
`rope_parameters`'s beta_fast field must be a float, got 32
`rope_parameters`'s beta_slow field must be a float, got 1
DeepseekV32Config got `key=rope_scaling` in kwargs but hasn't set it as attribute. For RoPE standardization you need to set `self.rope_parameters` in model's config. 
2026-04-22 22:02:22.984339: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303)
I0422 22:02:23.193365 139510284412736 max_utils.py:273] Attempting to initialize the jax distributed system...
I0422 22:02:32.235611 139510284412736 distributed.py:149] Starting JAX distributed service on [::]:8482
I0422 22:02:32.238020 139510284412736 distributed.py:172] Connecting to JAX distributed service on mt-10-shardy-false-ptycj-slice-job-0-0.mt-10-shardy-false-ptycj:8482
I0422 22:02:34.016199 139510284412736 max_utils.py:284] Jax distributed system initialized!
I0422 22:02:40.194585 139510284412736 max_utils.py:800] System Information: Jax Version: 0.9.2
I0422 22:02:40.194690 139510284412736 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0422 22:02:40.194730 139510284412736 max_utils.py:802] System Information: Jax Backend: PJRT C API
TFRT TPU v6 lite
Built on Mar 4 2026 11:32:08 (1772652728) cl/878335365
I0422 22:02:40.194765 139510284412736 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0422 22:02:40.920046 139510284412736 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0422 22:02:40.920348 139510284412736 checkpointing.py:677] Setting up checkpoint logger...
I0422 22:02:40.920402 139510284412736 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0422 22:02:40.920446 139510284412736 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0422 22:02:40.920779 139510284412736 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 0x7ee18d248260>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0422 22:02:43.759654 139510284412736 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0422 22:02:43.759886 139510284412736 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 0x7ecd686eb860>}, handler_registry=None
I0422 22:02:43.760135 139510284412736 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ecd686eb860>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`.
I0422 22:02:43.760185 139510284412736 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ecd686ed7c0>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`.
I0422 22:02:43.760217 139510284412736 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 0x7ecd686eb860>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ecd686eb860>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ecd686ed7c0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ecd686ed7c0>}).
I0422 22:02:43.760547 139510284412736 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0422 22:02:43.760616 139510284412736 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7ecd6823b560> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0422 22:02:45.065587 139510284412736 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_10_shardy_false/checkpoints
I0422 22:02:45.686406 139510284412736 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=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_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_10_shardy_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7ecd686ece90>
I0422 22:02:45.686593 139510284412736 checkpointing.py:301] Checkpoint manager created!
I0422 22:02:46.636214 139510284412736 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0422 22:02:46.636331 139510284412736 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 22:02:47.015256 139510284412736 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 22:02:47.015350 139510284412736 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 22:02:47.031979 139510284412736 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 22:02:47.032041 139510284412736 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 22:02:47.055782 139510284412736 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0422 22:02:47.055853 139510284412736 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 22:02:47.072537 139510284412736 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0422 22:02:47.072607 139510284412736 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 22:02:47.089147 139510284412736 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0422 22:02:47.089205 139510284412736 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 22:02:47.114225 139510284412736 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0422 22:02:47.114294 139510284412736 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 22:02:47.135475 139510284412736 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0422 22:02:47.135541 139510284412736 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0422 22:02:47.347576 139510284412736 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0422 22:02:47.347682 139510284412736 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0422 22:02:48.767506 139510284412736 maxtext_utils.py:1668]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0422 22:02:48.767641 139510284412736 maxtext_utils.py:1668]  params/params/decoder/layers/mlp/wi_0/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0422 22:02:48.767694 139510284412736 maxtext_utils.py:1668]  params/params/decoder/layers/mlp/wi_1/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0422 22:02:48.767752 139510284412736 maxtext_utils.py:1668]  params/params/decoder/layers/mlp/wo/kernel
    Shape:     float32[7168,16,2048]
    Logical:   P('mlp', 'layers', 'embed')
    Physical:  (None, None, 'fsdp')
I0422 22:02:48.767808 139510284412736 maxtext_utils.py:1668]  params/params/decoder/layers/post_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0422 22:02:48.767846 139510284412736 maxtext_utils.py:1668]  params/params/decoder/layers/pre_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0422 22:02:48.767898 139510284412736 maxtext_utils.py:1668]  params/params/decoder/layers/self_attention/key/kernel
    Shape:     float32[2048,16,16,128]
    Logical:   P('embed', 'layers', 'kv_heads', 'kv_head_dim')
    Physical:  ('fsdp', None, None, None)
I0422 22:02:48.767950 139510284412736 maxtext_utils.py:1668]  params/params/decoder/layers/self_attention/out/kernel
    Shape:     float32[16,16,128,2048]
    Logical:   P('heads', 'layers', 'kv', 'embed')
    Physical:  (None, None, None, 'fsdp')
I0422 22:02:48.767989 139510284412736 maxtext_utils.py:1668]  params/params/decoder/layers/self_attention/query/kernel
    Shape:     float32[2048,16,16,128]
    Logical:   P('embed', 'layers', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None, None)
I0422 22:02:48.768024 139510284412736 maxtext_utils.py:1668]  params/params/decoder/layers/self_attention/value/kernel
    Shape:     float32[2048,16,16,128]
    Logical:   P('embed', 'layers', 'kv_heads', 'kv_head_dim')
    Physical:  ('fsdp', None, None, None)
I0422 22:02:48.768071 139510284412736 maxtext_utils.py:1668]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)
I0422 22:02:48.768134 139510284412736 maxtext_utils.py:1668]  params/params/token_embedder/embedding

    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0422 22:02:49.257935 139510284412736 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0422 22:02:49.258030 139510284412736 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0422 22:02:49.273645 139510284412736 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0422 22:02:49.273706 139510284412736 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0422 22:03:00.235808 139510284412736 max_utils.py:791] Total memory size: 1.5 GB, Output size: 0.4 GB, Temp size: 1.1 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0422 22:03:00.236672 139510284412736 metric_logger.py:301] number parameters: 1.104 billion
I0422 22:03:12.319242 139510284412736 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0422 22:03:12.476719 139510284412736 checkpointing.py:776] Waited 0.1574571132659912 seconds for step 0 to finish before starting checkpointing.
I0422 22:03:12.479209 139510284412736 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 22:03:12.482563 139510284412736 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0422 22:03:12.483937 139510284412736 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_10_shardy_false/checkpoints/0.
I0422 22:03:13.193830 139510284412736 signaling_client.py:364] Using JaxDistributedSignalingClient
I0422 22:03:13.194904 139510284412736 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0422 22:03:13.194963 139510284412736 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False
I0422 22:03:13.472455 139510284412736 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.278701s
I0422 22:03:13.472623 139510284412736 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.426 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28430676460266113 s) (per-host)
I0422 22:03:13.472676 139510284412736 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.284369s (batch_requests_ready=0.002319s, total_serialization_initiated=0.281980s, others=0.000069s)
I0422 22:03:13.472766 139510284412736 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.288787s (all_items=0.000018s, per_item={'items': '0.00001764'}, temp_paths=0.288769)
I0422 22:03:13.475420 139510284412736 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.99 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_10_shardy_false/checkpoints/0.
I0422 22:03:13.475775 139381678806784 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 22:23:13.475737
I0422 22:03:13.478375 139510284412736 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 22:03:13.478675 139380068968192 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0422 22:03:13.478829 139510284412736 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_10_shardy_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776895392.4791908, 'wait_for_prev_duration_secs': 6.008148193359375e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776895392.4826035, 'checkpointer_blocking_duration_secs': 0.9933242797851562, 'get_old_steps_start_time': 1776895393.4759533, 'get_old_steps_duration_secs': 3.0040740966796875e-05, 'checkpoint_manager_blocking_start_time': 1776895392.4772508, 'checkpoint_manager_blocking_duration_secs': 1.0015408992767334}
I0422 22:03:13.478938 139510284412736 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0422 22:03:13.478987 139510284412736 max_utils.py:750] 
Memstats: After params initialized:
I0422 22:03:13.479032 139510284412736 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0422 22:03:13.479072 139510284412736 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0422 22:03:13.479122 139510284412736 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0422 22:03:13.479151 139510284412736 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0422 22:03:13.792004 139510284412736 metric_logger.py:196] completed step: 0, seconds: 12.082, TFLOP/s/device: 1.125, Tokens/s/device: 169.502, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0422 22:03:13.965622 139510284412736 metric_logger.py:196] completed step: 1, seconds: 1.471, TFLOP/s/device: 9.235, Tokens/s/device: 1392.005, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0422 22:03:14.380462 139510284412736 metric_logger.py:196] completed step: 2, seconds: 0.026, TFLOP/s/device: 517.822, Tokens/s/device: 78051.755, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0422 22:03:14.527685 139510284412736 metric_logger.py:196] completed step: 3, seconds: 0.415, TFLOP/s/device: 32.734, Tokens/s/device: 4934.012, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0422 22:03:14.823065 139510284412736 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.721, Tokens/s/device: 13372.947, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0422 22:03:14.828939 139510284412736 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.446, Tokens/s/device: 13934.532, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0422 22:03:17.033381    2794 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0422 22:03:19.943652 139380077360896 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_10_shardy_false/checkpoints/0/items/array_metadatas/process_6
I0422 22:03:36.569276 139510284412736 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.903, Tokens/s/device: 6919.012, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0422 22:03:36.716508 139510284412736 metric_logger.py:196] completed step: 7, seconds: 21.593, TFLOP/s/device: 0.629, Tokens/s/device: 94.844, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0422 22:03:36.863724 139510284412736 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.126, Tokens/s/device: 13434.001, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0422 22:03:37.010571 139510284412736 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0422 22:03:37.011244 139510284412736 checkpointing.py:776] Waited 0.0006954669952392578 seconds for step 9 to finish before starting checkpointing.
I0422 22:03:37.013241 139510284412736 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 22:03:50.569707 139381678806784 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 42.256 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.38136696815491 s) (per-host)
I0422 22:03:50.569812 139381678806784 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.093924s.
I0422 22:03:58.988834 139381678806784 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 45.512930s.
I0422 22:03:58.989141 139380068968192 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 22:03:58.989278 139380068968192 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0422 22:03:58.989329 139380068968192 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0422 22:03:58.991581 139380068968192 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0422 22:03:58.991762 139510284412736 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0422 22:03:58.991894 139510284412736 checkpoint_manager.py:1452] Waiting for previous save to complete took 21.978653 seconds. If this number is high, consider checkpointing less frequently.
I0422 22:03:58.993744 139510284412736 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0422 22:03:58.995902 139510284412736 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_10_shardy_false/checkpoints/9.
I0422 22:03:59.737067 139510284412736 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0422 22:03:59.737171 139510284412736 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False
I0422 22:03:59.782930 139510284412736 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.046984s
I0422 22:03:59.783108 139510284412736 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 30.568 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.05046343803405762 s) (per-host)
I0422 22:03:59.783157 139510284412736 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.050533s (batch_requests_ready=0.001820s, total_serialization_initiated=0.048637s, others=0.000075s)
I0422 22:03:59.783246 139510284412736 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.054651s (all_items=0.000016s, per_item={'items': '0.00001574'}, temp_paths=0.054635)
I0422 22:03:59.783896 139510284412736 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.79 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_10_shardy_false/checkpoints/9.
I0422 22:03:59.784266 139380068968192 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 22:23:59.784224
I0422 22:03:59.786280 139510284412736 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 22:03:59.786584 139359894345472 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0422 22:03:59.786731 139510284412736 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_10_shardy_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776895417.0132115, 'wait_for_prev_duration_secs': 21.97865343093872, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776895438.9937844, 'checkpointer_blocking_duration_secs': 0.7906270027160645, 'get_old_steps_start_time': 1776895439.784434, 'get_old_steps_duration_secs': 3.0040740966796875e-05, 'checkpoint_manager_blocking_start_time': 1776895417.011473, 'checkpoint_manager_blocking_duration_secs': 22.77522373199463}
I0422 22:03:59.786900 139510284412736 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0422 22:03:59.786947 139510284412736 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 22:04:05.840784 139380077360896 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260422_212603/linen_xpk_test_pipeline_scan_nnx_20260422_212603_10_shardy_false/checkpoints/9/items/array_metadatas/process_6
I0422 22:04:43.096200 139380068968192 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 36.427 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.363526821136475 s) (per-host)
I0422 22:04:43.096338 139380068968192 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.311954s.
I0422 22:04:50.867959 139380068968192 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 51.083560s.
I0422 22:04:50.868265 139359894345472 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 22:04:50.868393 139359894345472 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0422 22:04:50.868443 139359894345472 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0422 22:04:50.870522 139359894345472 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0422 22:04:50.870701 139510284412736 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0422 22:04:50.870851 139510284412736 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 22:04:50.871873 139510284412736 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.309, Tokens/s/device: 13913.800, 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: Wed Apr 22 22:05:03 UTC 2026
EXIT_CODE=0