MaxView

XPK Start: Wed Apr 22 12:44:08 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-22 12:44:32.752442: 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 12:44:32.932676 138207656257344 max_utils.py:273] Attempting to initialize the jax distributed system...
I0422 12:44:41.975043 138207656257344 distributed.py:149] Starting JAX distributed service on [::]:8482
I0422 12:44:41.977391 138207656257344 distributed.py:172] Connecting to JAX distributed service on mt-02-synthetic-fpyy1-slice-job-0-0.mt-02-synthetic-fpyy1:8482
I0422 12:44:44.015648 138207656257344 max_utils.py:284] Jax distributed system initialized!
I0422 12:44:50.209705 138207656257344 max_utils.py:800] System Information: Jax Version: 0.9.2
I0422 12:44:50.209810 138207656257344 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0422 12:44:50.209849 138207656257344 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 12:44:50.209884 138207656257344 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0422 12:44:50.915158 138207656257344 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0422 12:44:50.915763 138207656257344 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0422 12:44:50.915946 138207656257344 checkpointing.py:688] Setting up checkpoint logger...
I0422 12:44:50.915995 138207656257344 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0422 12:44:50.916038 138207656257344 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0422 12:44:50.916383 138207656257344 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 0x7db2859c26c0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0422 12:44:53.789525 138207656257344 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0422 12:44:53.789764 138207656257344 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 0x7d9e503faf30>}, handler_registry=None
I0422 12:44:53.789999 138207656257344 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d9e503faf30>` 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 12:44:53.790046 138207656257344 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d9e503ea780>` 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 12:44:53.790082 138207656257344 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 0x7d9e503faf30>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d9e503faf30>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d9e503ea780>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d9e503ea780>}).
I0422 12:44:53.790417 138207656257344 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0422 12:44:53.790487 138207656257344 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d9e5020c540> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0422 12:44:54.930510 138207656257344 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_02_synthetic/checkpoints
I0422 12:44:54.954890 138207656257344 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_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d9e503ef620>
I0422 12:44:54.955014 138207656257344 checkpointing.py:302] Checkpoint manager created!
I0422 12:44:55.907040 138207656257344 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0422 12:44:55.907150 138207656257344 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 12:44:56.290153 138207656257344 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 12:44:56.290259 138207656257344 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 12:44:56.308142 138207656257344 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 12:44:56.308201 138207656257344 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 12:44:56.332834 138207656257344 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 12:44:56.332917 138207656257344 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 12:44:56.349917 138207656257344 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 12:44:56.349990 138207656257344 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 12:44:56.366795 138207656257344 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 12:44:56.366855 138207656257344 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 12:44:56.391931 138207656257344 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0422 12:44:56.392001 138207656257344 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 12:44:56.413673 138207656257344 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0422 12:44:56.413740 138207656257344 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0422 12:44:56.632746 138207656257344 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0422 12:44:56.632854 138207656257344 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0422 12:44:58.071432 138207656257344 maxtext_utils.py:1827]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0422 12:44:58.071563 138207656257344 maxtext_utils.py:1827]  params/params/decoder/layers/mlp/wi_0/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0422 12:44:58.071617 138207656257344 maxtext_utils.py:1827]  params/params/decoder/layers/mlp/wi_1/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0422 12:44:58.071675 138207656257344 maxtext_utils.py:1827]  params/params/decoder/layers/mlp/wo/kernel
    Shape:     float32[7168,16,2048]
    Logical:   P('mlp', 'layers', 'embed')
    Physical:  (None, None, 'fsdp')
I0422 12:44:58.071728 138207656257344 maxtext_utils.py:1827]  params/params/decoder/layers/post_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0422 12:44:58.071765 138207656257344 maxtext_utils.py:1827]  params/params/decoder/layers/pre_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0422 12:44:58.071816 138207656257344 maxtext_utils.py:1827]  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 12:44:58.071868 138207656257344 maxtext_utils.py:1827]  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 12:44:58.071908 138207656257344 maxtext_utils.py:1827]  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 12:44:58.071945 138207656257344 maxtext_utils.py:1827]  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 12:44:58.071992 138207656257344 maxtext_utils.py:1827]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0422 12:44:58.072041 138207656257344 maxtext_utils.py:1827]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0422 12:44:58.570185 138207656257344 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0422 12:44:58.570288 138207656257344 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0422 12:44:58.586279 138207656257344 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0422 12:44:58.586353 138207656257344 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0422 12:45:09.531562 138207656257344 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 12:45:09.532438 138207656257344 metric_logger.py:301] number parameters: 1.104 billion
I0422 12:45:20.871882 138207656257344 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0422 12:45:21.026411 138207656257344 checkpointing.py:798] Waited 0.15451502799987793 seconds for step 0 to finish before starting checkpointing.
I0422 12:45:21.029035 138207656257344 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 12:45:21.030796 138207656257344 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0422 12:45:21.032237 138207656257344 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_02_synthetic/checkpoints/0.
I0422 12:45:22.209067 138207656257344 signaling_client.py:364] Using JaxDistributedSignalingClient
I0422 12:45:22.209971 138207656257344 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0422 12:45:22.210028 138207656257344 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 12:45:22.491225 138207656257344 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.282232s
I0422 12:45:22.491407 138207656257344 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.357 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2879455089569092 s) (per-host)
I0422 12:45:22.491462 138207656257344 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.288010s (batch_requests_ready=0.002411s, total_serialization_initiated=0.285527s, others=0.000073s)
I0422 12:45:22.491559 138207656257344 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.292110s (all_items=0.000020s, per_item={'items': '0.00002003'}, temp_paths=0.292090)
I0422 12:45:22.492319 138207656257344 event_tracking.py:125] [process=6] [async] Finished blocking save in 1.46 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_02_synthetic/checkpoints/0.
I0422 12:45:22.492668 138080764311296 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 13:05:22.492631
I0422 12:45:22.505084 138207656257344 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 12:45:22.505393 138080193869568 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0422 12:45:22.505544 138207656257344 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776861921.0290174, 'wait_for_prev_duration_secs': 6.270408630371094e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776861921.0308347, 'checkpointer_blocking_duration_secs': 1.4620025157928467, 'get_old_steps_start_time': 1776861922.4928706, 'get_old_steps_duration_secs': 3.62396240234375e-05, 'checkpoint_manager_blocking_start_time': 1776861921.026884, 'checkpoint_manager_blocking_duration_secs': 1.478621244430542}
I0422 12:45:22.505649 138207656257344 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0422 12:45:22.505699 138207656257344 max_utils.py:750] 
Memstats: After params initialized:
I0422 12:45:22.505750 138207656257344 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0422 12:45:22.505783 138207656257344 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0422 12:45:22.505810 138207656257344 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0422 12:45:22.505835 138207656257344 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0422 12:45:22.817769 138207656257344 metric_logger.py:196] completed step: 0, seconds: 11.339, TFLOP/s/device: 1.198, Tokens/s/device: 180.610, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0422 12:45:22.990075 138207656257344 metric_logger.py:196] completed step: 1, seconds: 1.944, TFLOP/s/device: 6.988, Tokens/s/device: 1053.295, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0422 12:45:23.404079 138207656257344 metric_logger.py:196] completed step: 2, seconds: 0.025, TFLOP/s/device: 543.463, Tokens/s/device: 81916.723, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0422 12:45:23.551358 138207656257344 metric_logger.py:196] completed step: 3, seconds: 0.414, TFLOP/s/device: 32.791, Tokens/s/device: 4942.586, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0422 12:45:23.846580 138207656257344 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.914, Tokens/s/device: 13402.089, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0422 12:45:23.852176 138207656257344 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.223, Tokens/s/device: 13900.861, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0422 12:45:25.485074    2789 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0422 12:45:27.994809 138080202262272 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_02_synthetic/checkpoints/0/items/array_metadatas/process_6
I0422 12:45:46.097084 138207656257344 metric_logger.py:196] completed step: 6, seconds: 0.295, TFLOP/s/device: 45.982, Tokens/s/device: 6930.954, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0422 12:45:46.244202 138207656257344 metric_logger.py:196] completed step: 7, seconds: 22.098, TFLOP/s/device: 0.615, Tokens/s/device: 92.679, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0422 12:45:46.391827 138207656257344 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.175, Tokens/s/device: 13441.407, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0422 12:45:46.538511 138207656257344 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0422 12:45:46.539181 138207656257344 checkpointing.py:798] Waited 0.0007011890411376953 seconds for step 9 to finish before starting checkpointing.
I0422 12:45:46.541157 138207656257344 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 12:45:59.741059 138080764311296 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 42.080 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.537559509277344 s) (per-host)
I0422 12:45:59.741194 138080764311296 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.248405s.
I0422 12:46:08.657713 138080764311296 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 46.164908s.
I0422 12:46:08.657921 138080193869568 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 12:46:08.657987 138080193869568 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0422 12:46:08.658035 138080193869568 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0422 12:46:08.659752 138080193869568 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0422 12:46:08.659934 138207656257344 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 12:46:08.660071 138207656257344 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.118912 seconds. If this number is high, consider checkpointing less frequently.
I0422 12:46:08.662064 138207656257344 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0422 12:46:08.664162 138207656257344 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_02_synthetic/checkpoints/9.
I0422 12:46:08.950981 138207656257344 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0422 12:46:08.951075 138207656257344 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 12:46:08.992854 138207656257344 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.043012s
I0422 12:46:08.993017 138207656257344 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 33.032 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.046698808670043945 s) (per-host)
I0422 12:46:08.993068 138207656257344 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.046758s (batch_requests_ready=0.001977s, total_serialization_initiated=0.044713s, others=0.000068s)
I0422 12:46:08.993169 138207656257344 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.050756s (all_items=0.000016s, per_item={'items': '0.00001597'}, temp_paths=0.050740)
I0422 12:46:08.993824 138207656257344 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.33 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_02_synthetic/checkpoints/9.
I0422 12:46:08.994170 138080193869568 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 13:06:08.994134
I0422 12:46:08.996170 138207656257344 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 12:46:08.996459 138058250827520 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0422 12:46:08.996625 138207656257344 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776861946.5411277, 'wait_for_prev_duration_secs': 22.1189124584198, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776861968.6621165, 'checkpointer_blocking_duration_secs': 0.33215880393981934, 'get_old_steps_start_time': 1776861968.9943023, 'get_old_steps_duration_secs': 3.0040740966796875e-05, 'checkpoint_manager_blocking_start_time': 1776861946.53941, 'checkpoint_manager_blocking_duration_secs': 22.45718002319336}
I0422 12:46:08.996739 138207656257344 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0422 12:46:08.996783 138207656257344 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 12:46:14.829173 138080202262272 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_02_synthetic/checkpoints/9/items/array_metadatas/process_6
I0422 12:46:51.118924 138080193869568 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.456 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.17256474494934 s) (per-host)
I0422 12:46:51.119051 138080193869568 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.124258s.
I0422 12:46:59.501393 138080193869568 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 50.506583s.
I0422 12:46:59.501689 138058250827520 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 12:46:59.501814 138058250827520 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0422 12:46:59.501861 138058250827520 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0422 12:46:59.503521 138058250827520 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0422 12:46:59.503690 138207656257344 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 12:46:59.503835 138207656257344 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 12:46:59.504802 138207656257344 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.319, Tokens/s/device: 13915.312, 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 12:47:08 UTC 2026
EXIT_CODE=0