MaxView

XPK Start: Wed Apr 22 07:58:09 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 07:58:34.262423: 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 07:58:34.473255 135017575241536 max_utils.py:273] Attempting to initialize the jax distributed system...
I0422 07:58:43.514561 135017575241536 distributed.py:149] Starting JAX distributed service on [::]:8482
I0422 07:58:43.516982 135017575241536 distributed.py:172] Connecting to JAX distributed service on mt-13-scan-layers-true-vqgrn-slice-job-0-0.mt-13-scan-layers-true-vqgrn:8482
I0422 07:58:44.428102 135017575241536 max_utils.py:284] Jax distributed system initialized!
I0422 07:58:50.693799 135017575241536 max_utils.py:800] System Information: Jax Version: 0.9.2
I0422 07:58:50.693910 135017575241536 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0422 07:58:50.693950 135017575241536 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 07:58:50.693985 135017575241536 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0422 07:58:51.383319 135017575241536 maxtext_utils.py:1604] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0422 07:58:51.383604 135017575241536 checkpointing.py:677] Setting up checkpoint logger...
I0422 07:58:51.383657 135017575241536 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0422 07:58:51.383702 135017575241536 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0422 07:58:51.384041 135017575241536 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 0x7acb9f11c200>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0422 07:58:54.687336 135017575241536 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0422 07:58:54.687573 135017575241536 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 0x7aca76ee2270>}, handler_registry=None
I0422 07:58:54.687810 135017575241536 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7aca76ee2270>` 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 07:58:54.687860 135017575241536 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ab7084f21e0>` 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 07:58:54.687897 135017575241536 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 0x7aca76ee2270>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7aca76ee2270>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ab7084f21e0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ab7084f21e0>}).
I0422 07:58:54.688225 135017575241536 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0422 07:58:54.688323 135017575241536 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7ab6f817d800> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0422 07:58:56.285082 135017575241536 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260422_071422/linen_xpk_main_20260422_071422_13_scan_layers_true/checkpoints
I0422 07:58:56.709449 135017575241536 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_main_20260422_071422/linen_xpk_main_20260422_071422_13_scan_layers_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7ab70864f530>
I0422 07:58:56.709622 135017575241536 checkpointing.py:301] Checkpoint manager created!
I0422 07:58:57.660301 135017575241536 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0422 07:58:57.660423 135017575241536 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 07:58:58.041100 135017575241536 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 07:58:58.041197 135017575241536 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 07:58:58.057484 135017575241536 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 07:58:58.057545 135017575241536 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 07:58:58.081001 135017575241536 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 07:58:58.081074 135017575241536 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 07:58:58.097333 135017575241536 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 07:58:58.097398 135017575241536 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 07:58:58.113590 135017575241536 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 07:58:58.113656 135017575241536 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 07:58:58.138278 135017575241536 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0422 07:58:58.138358 135017575241536 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 07:58:58.159331 135017575241536 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0422 07:58:58.159401 135017575241536 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0422 07:58:58.368609 135017575241536 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0422 07:58:58.368720 135017575241536 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0422 07:58:59.812419 135017575241536 maxtext_utils.py:1707]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0422 07:58:59.812545 135017575241536 maxtext_utils.py:1707]  params/params/decoder/layers/mlp/wi_0/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0422 07:58:59.812599 135017575241536 maxtext_utils.py:1707]  params/params/decoder/layers/mlp/wi_1/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0422 07:58:59.812656 135017575241536 maxtext_utils.py:1707]  params/params/decoder/layers/mlp/wo/kernel
    Shape:     float32[7168,16,2048]
    Logical:   P('mlp', 'layers', 'embed')
    Physical:  (None, None, 'fsdp')
I0422 07:58:59.812707 135017575241536 maxtext_utils.py:1707]  params/params/decoder/layers/post_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0422 07:58:59.812750 135017575241536 maxtext_utils.py:1707]  params/params/decoder/layers/pre_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0422 07:58:59.812804 135017575241536 maxtext_utils.py:1707]  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 07:58:59.812856 135017575241536 maxtext_utils.py:1707]  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 07:58:59.812895 135017575241536 maxtext_utils.py:1707]  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 07:58:59.812931 135017575241536 maxtext_utils.py:1707]  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 07:58:59.812977 135017575241536 maxtext_utils.py:1707]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)
I0422 07:58:59.813024 135017575241536 maxtext_utils.py:1707]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')

I0422 07:59:00.297731 135017575241536 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0422 07:59:00.297825 135017575241536 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0422 07:59:00.313164 135017575241536 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0422 07:59:00.313225 135017575241536 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0422 07:59:11.238781 135017575241536 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 07:59:11.239546 135017575241536 metric_logger.py:301] number parameters: 1.104 billion
I0422 07:59:22.457111 135017575241536 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0422 07:59:22.836477 135017575241536 checkpointing.py:776] Waited 0.3793470859527588 seconds for step 0 to finish before starting checkpointing.
I0422 07:59:22.838739 135017575241536 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 07:59:22.840421 135017575241536 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0422 07:59:22.841821 135017575241536 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260422_071422/linen_xpk_main_20260422_071422_13_scan_layers_true/checkpoints/0.
I0422 07:59:23.597584 135017575241536 signaling_client.py:364] Using JaxDistributedSignalingClient
I0422 07:59:23.598593 135017575241536 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0422 07:59:23.598649 135017575241536 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 07:59:23.881140 135017575241536 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.283628s
I0422 07:59:23.881322 135017575241536 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.334 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2891702651977539 s) (per-host)
I0422 07:59:23.881373 135017575241536 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.289231s (batch_requests_ready=0.002347s, total_serialization_initiated=0.286815s, others=0.000070s)
I0422 07:59:23.881465 135017575241536 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.293833s (all_items=0.000018s, per_item={'items': '0.00001812'}, temp_paths=0.293815)
I0422 07:59:23.882217 135017575241536 event_tracking.py:125] [process=6] [async] Finished blocking save in 1.04 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260422_071422/linen_xpk_main_20260422_071422_13_scan_layers_true/checkpoints/0.
I0422 07:59:23.882590 134888682419968 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 08:19:23.882553
I0422 07:59:23.902547 135017575241536 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 07:59:23.902863 134887566714624 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0422 07:59:23.903027 135017575241536 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260422_071422/linen_xpk_main_20260422_071422_13_scan_layers_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776844762.838721, 'wait_for_prev_duration_secs': 6.175041198730469e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776844762.8404589, 'checkpointer_blocking_duration_secs': 1.042287826538086, 'get_old_steps_start_time': 1776844763.8827732, 'get_old_steps_duration_secs': 3.123283386230469e-05, 'checkpoint_manager_blocking_start_time': 1776844762.8368938, 'checkpoint_manager_blocking_duration_secs': 1.066094160079956}
I0422 07:59:23.903136 135017575241536 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0422 07:59:23.903187 135017575241536 max_utils.py:750] 
Memstats: After params initialized:
I0422 07:59:23.903238 135017575241536 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0422 07:59:23.903273 135017575241536 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0422 07:59:23.903324 135017575241536 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0422 07:59:23.903380 135017575241536 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0422 07:59:24.219166 135017575241536 metric_logger.py:196] completed step: 0, seconds: 11.217, TFLOP/s/device: 1.211, Tokens/s/device: 182.573, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0422 07:59:24.392432 135017575241536 metric_logger.py:196] completed step: 1, seconds: 1.760, TFLOP/s/device: 7.718, Tokens/s/device: 1163.319, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0422 07:59:24.820265 135017575241536 metric_logger.py:196] completed step: 2, seconds: 0.025, TFLOP/s/device: 553.492, Tokens/s/device: 83428.385, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0422 07:59:24.967674 135017575241536 metric_logger.py:196] completed step: 3, seconds: 0.430, TFLOP/s/device: 31.630, Tokens/s/device: 4767.680, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0422 07:59:25.262748 135017575241536 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.754, Tokens/s/device: 13377.926, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0422 07:59:25.268700 135017575241536 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.219, Tokens/s/device: 13900.295, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0422 07:59:27.247637    2799 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0422 07:59:29.710990 134888665634560 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260422_071422/linen_xpk_main_20260422_071422_13_scan_layers_true/checkpoints/0/items/array_metadatas/process_6
I0422 07:59:47.466121 135017575241536 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.970, Tokens/s/device: 6929.032, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0422 07:59:47.613689 135017575241536 metric_logger.py:196] completed step: 7, seconds: 22.051, TFLOP/s/device: 0.616, Tokens/s/device: 92.876, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0422 07:59:47.760972 135017575241536 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.438, Tokens/s/device: 13481.134, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0422 07:59:47.907670 135017575241536 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0422 07:59:47.908391 135017575241536 checkpointing.py:776] Waited 0.0007410049438476562 seconds for step 9 to finish before starting checkpointing.
I0422 07:59:47.911610 135017575241536 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 07:59:59.382806 134888682419968 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 44.134 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 35.790613651275635 s) (per-host)
I0422 07:59:59.382938 134888682419968 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 35.500236s.
I0422 08:00:11.318819 134888682419968 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 47.436101s.
I0422 08:00:11.319100 134887566714624 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 08:00:11.319217 134887566714624 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0422 08:00:11.319267 134887566714624 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0422 08:00:11.321297 134887566714624 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0422 08:00:11.321496 135017575241536 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 08:00:11.321630 135017575241536 checkpoint_manager.py:1452] Waiting for previous save to complete took 23.410018 seconds. If this number is high, consider checkpointing less frequently.
I0422 08:00:11.323440 135017575241536 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0422 08:00:11.325568 135017575241536 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260422_071422/linen_xpk_main_20260422_071422_13_scan_layers_true/checkpoints/9.
I0422 08:00:12.047113 135017575241536 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0422 08:00:12.047211 135017575241536 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 08:00:12.082805 135017575241536 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.036833s
I0422 08:00:12.082976 135017575241536 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 38.210 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04037070274353027 s) (per-host)
I0422 08:00:12.083034 135017575241536 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.040431s (batch_requests_ready=0.001878s, total_serialization_initiated=0.038484s, others=0.000069s)
I0422 08:00:12.083124 135017575241536 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.044638s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=0.044623)
I0422 08:00:12.083913 135017575241536 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.76 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260422_071422/linen_xpk_main_20260422_071422_13_scan_layers_true/checkpoints/9.
I0422 08:00:12.084263 134887566714624 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 08:20:12.084226
I0422 08:00:12.090362 135017575241536 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 08:00:12.090602 134888648849152 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0422 08:00:12.090762 135017575241536 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260422_071422/linen_xpk_main_20260422_071422_13_scan_layers_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776844787.9115832, 'wait_for_prev_duration_secs': 23.41001796722412, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776844811.3234813, 'checkpointer_blocking_duration_secs': 0.7609374523162842, 'get_old_steps_start_time': 1776844812.0844429, 'get_old_steps_duration_secs': 2.9802322387695312e-05, 'checkpoint_manager_blocking_start_time': 1776844787.9086103, 'checkpoint_manager_blocking_duration_secs': 24.182117223739624}
I0422 08:00:12.090865 135017575241536 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0422 08:00:12.090911 135017575241536 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 08:00:17.137159 134888668415744 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260422_071422/linen_xpk_main_20260422_071422_13_scan_layers_true/checkpoints/9/items/array_metadatas/process_6
I0422 08:00:53.518764 134887566714624 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 38.084 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.476118087768555 s) (per-host)
I0422 08:00:53.518893 134887566714624 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.434503s.
I0422 08:01:02.331372 134887566714624 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 50.246967s.
I0422 08:01:02.331669 134888648849152 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 08:01:02.331789 134888648849152 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0422 08:01:02.331846 134888648849152 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0422 08:01:02.333209 134888648849152 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0422 08:01:02.335860 135017575241536 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 08:01:02.336005 135017575241536 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 08:01:02.336983 135017575241536 metric_logger.py:196] completed step: 9, seconds: 0.148, TFLOP/s/device: 92.003, Tokens/s/device: 13867.635, 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 08:01:14 UTC 2026
EXIT_CODE=0