MaxView

XPK Start: Mon Apr 20 08:45:02 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-20 08:45:26.818243: 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)
I0420 08:45:26.995489 139315833874240 max_utils.py:273] Attempting to initialize the jax distributed system...
I0420 08:45:36.036991 139315833874240 distributed.py:149] Starting JAX distributed service on [::]:8482
I0420 08:45:36.039286 139315833874240 distributed.py:172] Connecting to JAX distributed service on mt-10-shardy-true-zhdu8-slice-job-0-0.mt-10-shardy-true-zhdu8:8482
I0420 08:45:37.344895 139315833874240 max_utils.py:284] Jax distributed system initialized!
I0420 08:45:43.535645 139315833874240 max_utils.py:800] System Information: Jax Version: 0.9.2
I0420 08:45:43.535770 139315833874240 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0420 08:45:43.535813 139315833874240 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
I0420 08:45:43.535848 139315833874240 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0420 08:45:44.603593 139315833874240 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 08:45:44.604198 139315833874240 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 08:45:44.604383 139315833874240 checkpointing.py:688] Setting up checkpoint logger...
I0420 08:45:44.604432 139315833874240 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0420 08:45:44.604476 139315833874240 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0420 08:45:44.604845 139315833874240 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 0x7eb45e514e00>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0420 08:45:47.251889 139315833874240 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0420 08:45:47.252170 139315833874240 checkpoint_manager.py:708] [process=1][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7eb33b1d8b00>}, handler_registry=None
I0420 08:45:47.252408 139315833874240 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7eb33b1d8b00>` 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`.
I0420 08:45:47.252456 139315833874240 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7eb33b2c7830>` 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`.
I0420 08:45:47.252492 139315833874240 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 0x7eb33b1d8b00>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7eb33b1d8b00>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7eb33b2c7830>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7eb33b2c7830>}).
I0420 08:45:47.252821 139315833874240 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0420 08:45:47.252896 139315833874240 async_checkpointer.py:192] [process=1][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7ea02c200360> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0420 08:45:47.987044 139315833874240 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533/linen_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533_10_shardy_true/checkpoints
I0420 08:45:48.416537 139315833874240 checkpoint_manager.py:929] [process=1][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_trainstate_and_training_loop_20260420_081533/linen_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533_10_shardy_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7eb33b2e1370>
I0420 08:45:48.416735 139315833874240 checkpointing.py:302] Checkpoint manager created!
I0420 08:45:49.345804 139315833874240 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0420 08:45:49.345906 139315833874240 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 08:45:49.728880 139315833874240 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 08:45:49.728972 139315833874240 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 08:45:49.750611 139315833874240 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 08:45:49.750718 139315833874240 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 08:45:49.774682 139315833874240 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 08:45:49.774766 139315833874240 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 08:45:49.791309 139315833874240 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 08:45:49.791369 139315833874240 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 08:45:49.807649 139315833874240 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 08:45:49.807721 139315833874240 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 08:45:49.832190 139315833874240 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0420 08:45:49.832262 139315833874240 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 08:45:49.853022 139315833874240 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0420 08:45:49.853082 139315833874240 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0420 08:45:50.066746 139315833874240 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0420 08:45:50.066848 139315833874240 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0420 08:45:51.489669 139315833874240 maxtext_utils.py:1821]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0420 08:45:51.489829 139315833874240 maxtext_utils.py:1821]  params/params/decoder/layers/mlp/wi_0/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0420 08:45:51.489904 139315833874240 maxtext_utils.py:1821]  params/params/decoder/layers/mlp/wi_1/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0420 08:45:51.489993 139315833874240 maxtext_utils.py:1821]  params/params/decoder/layers/mlp/wo/kernel
    Shape:     float32[7168,16,2048]
    Logical:   P('mlp', 'layers', 'embed')
    Physical:  (None, None, 'fsdp')
I0420 08:45:51.490074 139315833874240 maxtext_utils.py:1821]  params/params/decoder/layers/post_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0420 08:45:51.490134 139315833874240 maxtext_utils.py:1821]  params/params/decoder/layers/pre_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0420 08:45:51.490211 139315833874240 maxtext_utils.py:1821]  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)
I0420 08:45:51.490300 139315833874240 maxtext_utils.py:1821]  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')
I0420 08:45:51.490364 139315833874240 maxtext_utils.py:1821]  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)
I0420 08:45:51.490420 139315833874240 maxtext_utils.py:1821]  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)
I0420 08:45:51.490493 139315833874240 maxtext_utils.py:1821]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0420 08:45:51.490564 139315833874240 maxtext_utils.py:1821]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0420 08:45:51.978356 139315833874240 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 08:45:51.978450 139315833874240 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 08:45:51.993954 139315833874240 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 08:45:51.994022 139315833874240 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 08:46:02.884549 139315833874240 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.
I0420 08:46:02.885382 139315833874240 metric_logger.py:301] number parameters: 1.104 billion
I0420 08:46:14.105639 139315833874240 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0420 08:46:14.259462 139315833874240 checkpointing.py:798] Waited 0.1538090705871582 seconds for step 0 to finish before starting checkpointing.
I0420 08:46:14.261907 139315833874240 checkpoint_manager.py:2009] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 08:46:14.263474 139315833874240 checkpoint_manager.py:1512] [process=1] Saving checkpoint at step 0
I0420 08:46:14.264948 139315833874240 event_tracking.py:70] [process=1] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533/linen_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533_10_shardy_true/checkpoints/0.
I0420 08:46:14.552371 139315833874240 signaling_client.py:364] Using JaxDistributedSignalingClient
I0420 08:46:14.553368 139315833874240 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0420 08:46:14.553423 139315833874240 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
I0420 08:46:14.832055 139315833874240 base_pytree_checkpoint_handler.py:154] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.279761s
I0420 08:46:14.832227 139315833874240 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/blocking_gbytes_per_sec: 5.405 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2853987216949463 s) (per-host)
I0420 08:46:14.832280 139315833874240 base_pytree_checkpoint_handler.py:768] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.285460s (batch_requests_ready=0.002450s, total_serialization_initiated=0.282940s, others=0.000070s)
I0420 08:46:14.832370 139315833874240 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.289637s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.289619)
I0420 08:46:14.833200 139315833874240 event_tracking.py:125] [process=1] [async] Finished blocking save in 0.57 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533/linen_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533_10_shardy_true/checkpoints/0.
I0420 08:46:14.833544 139187374200576 async_checkpointer.py:76] [process=1][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 09:06:14.833506
I0420 08:46:14.835622 139315833874240 checkpoint_manager.py:1560] [process=1][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 08:46:14.835911 139184685700864 async_checkpointer.py:280] [process=1][thread=save_finalize] Waiting for background save thread=async_save.
I0420 08:46:14.836091 139315833874240 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533/linen_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776674774.2618885, 'wait_for_prev_duration_secs': 6.246566772460938e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776674774.2635121, 'checkpointer_blocking_duration_secs': 0.5701892375946045, 'get_old_steps_start_time': 1776674774.8337405, 'get_old_steps_duration_secs': 3.075599670410156e-05, 'checkpoint_manager_blocking_start_time': 1776674774.2599695, 'checkpoint_manager_blocking_duration_secs': 0.5760824680328369}
I0420 08:46:14.836200 139315833874240 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0420 08:46:14.836251 139315833874240 max_utils.py:750] 
Memstats: After params initialized:
I0420 08:46:14.836301 139315833874240 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_2(process=1,(2,0,0,0))
I0420 08:46:14.836335 139315833874240 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_3(process=1,(3,0,0,0))
I0420 08:46:14.836359 139315833874240 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_6(process=1,(2,1,0,0))
I0420 08:46:14.836382 139315833874240 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_7(process=1,(3,1,0,0))
I0420 08:46:15.146533 139315833874240 metric_logger.py:196] completed step: 0, seconds: 11.220, TFLOP/s/device: 1.211, Tokens/s/device: 182.529, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0420 08:46:15.326993 139315833874240 metric_logger.py:196] completed step: 1, seconds: 1.039, TFLOP/s/device: 13.074, Tokens/s/device: 1970.692, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0420 08:46:15.757732 139315833874240 metric_logger.py:196] completed step: 2, seconds: 0.033, TFLOP/s/device: 406.532, Tokens/s/device: 61277.003, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0420 08:46:15.905124 139315833874240 metric_logger.py:196] completed step: 3, seconds: 0.431, TFLOP/s/device: 31.519, Tokens/s/device: 4750.847, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0420 08:46:16.200004 139315833874240 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.902, Tokens/s/device: 13400.335, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0420 08:46:16.206602 139315833874240 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.250, Tokens/s/device: 13904.920, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0420 08:46:18.034333    2800 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0420 08:46:20.090733 139184694093568 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533/linen_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533_10_shardy_true/checkpoints/0/items/array_metadatas/process_1
I0420 08:46:37.766188 139315833874240 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.916, Tokens/s/device: 6920.930, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0420 08:46:37.913682 139315833874240 metric_logger.py:196] completed step: 7, seconds: 21.413, TFLOP/s/device: 0.635, Tokens/s/device: 95.643, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0420 08:46:38.061244 139315833874240 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.187, Tokens/s/device: 13443.172, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0420 08:46:38.207909 139315833874240 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0420 08:46:38.208553 139315833874240 checkpointing.py:798] Waited 0.0006656646728515625 seconds for step 9 to finish before starting checkpointing.
I0420 08:46:38.210645 139315833874240 checkpoint_manager.py:2020] [process=1][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 08:46:51.463894 139187374200576 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/gbytes_per_sec: 42.788 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.91702675819397 s) (per-host)
I0420 08:46:51.464017 139187374200576 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.630362s.
I0420 08:47:00.361483 139187374200576 async_checkpointer.py:160] [process=1][thread=async_save] Background save thread done. Time taken: 45.527811s.
I0420 08:47:00.361732 139184685700864 async_checkpointer.py:288] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 08:47:00.361797 139184685700864 async_checkpointer.py:298] [process=1][thread=save_finalize] No errors found in background save thread=async_save.
I0420 08:47:00.361849 139184685700864 checkpoint_manager.py:2137] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0420 08:47:00.363794 139184685700864 checkpoint_manager.py:2146] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0420 08:47:00.363984 139315833874240 checkpoint_manager.py:2032] [process=1][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0420 08:47:00.364129 139315833874240 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.153486 seconds. If this number is high, consider checkpointing less frequently.
I0420 08:47:00.366050 139315833874240 checkpoint_manager.py:1512] [process=1] Saving checkpoint at step 9
I0420 08:47:00.368173 139315833874240 event_tracking.py:70] [process=1] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533/linen_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533_10_shardy_true/checkpoints/9.
I0420 08:47:00.654177 139315833874240 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0420 08:47:00.654263 139315833874240 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
I0420 08:47:00.689987 139315833874240 base_pytree_checkpoint_handler.py:154] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.036990s
I0420 08:47:00.690151 139315833874240 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/blocking_gbytes_per_sec: 38.233 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.040346384048461914 s) (per-host)
I0420 08:47:00.690202 139315833874240 base_pytree_checkpoint_handler.py:768] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.040406s (batch_requests_ready=0.001729s, total_serialization_initiated=0.038610s, others=0.000067s)
I0420 08:47:00.690298 139315833874240 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.044459s (all_items=0.000015s, per_item={'items': '0.00001454'}, temp_paths=0.044444)
I0420 08:47:00.691026 139315833874240 event_tracking.py:125] [process=1] [async] Finished blocking save in 0.32 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533/linen_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533_10_shardy_true/checkpoints/9.
I0420 08:47:00.691298 139184685700864 async_checkpointer.py:76] [process=1][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 09:07:00.691277
I0420 08:47:00.697373 139315833874240 checkpoint_manager.py:1560] [process=1][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 08:47:00.697660 139165253494528 async_checkpointer.py:280] [process=1][thread=save_finalize] Waiting for background save thread=async_save.
I0420 08:47:00.697779 139315833874240 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533/linen_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776674798.2106144, 'wait_for_prev_duration_secs': 22.153486013412476, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776674820.3660908, 'checkpointer_blocking_duration_secs': 0.3253037929534912, 'get_old_steps_start_time': 1776674820.6914182, 'get_old_steps_duration_secs': 2.9087066650390625e-05, 'checkpoint_manager_blocking_start_time': 1776674798.2087932, 'checkpoint_manager_blocking_duration_secs': 22.48895025253296}
I0420 08:47:00.697941 139315833874240 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0420 08:47:00.697990 139315833874240 checkpoint_manager.py:2020] [process=1][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 08:47:06.494268 139184694093568 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533/linen_xpk_feat_nnx_trainstate_and_training_loop_20260420_081533_10_shardy_true/checkpoints/9/items/array_metadatas/process_1
I0420 08:47:42.845099 139184685700864 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/gbytes_per_sec: 37.435 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.19525861740112 s) (per-host)
I0420 08:47:42.845209 139184685700864 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.153840s.
I0420 08:47:52.262941 139184685700864 async_checkpointer.py:160] [process=1][thread=async_save] Background save thread done. Time taken: 51.571552s.
I0420 08:47:52.263177 139165253494528 async_checkpointer.py:288] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 08:47:52.263237 139165253494528 async_checkpointer.py:298] [process=1][thread=save_finalize] No errors found in background save thread=async_save.
I0420 08:47:52.263300 139165253494528 checkpoint_manager.py:2137] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0420 08:47:52.264910 139165253494528 checkpoint_manager.py:2146] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0420 08:47:52.265092 139315833874240 checkpoint_manager.py:2032] [process=1][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0420 08:47:52.265244 139315833874240 checkpoint_manager.py:2009] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 08:47:52.266299 139315833874240 metric_logger.py:196] completed step: 9, seconds: 0.148, TFLOP/s/device: 92.059, Tokens/s/device: 13876.091, 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: Mon Apr 20 08:48:03 UTC 2026
EXIT_CODE=0