XPK Start: Tue Apr 21 08:40:16 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-21 08:40:55.631307: 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)
I0421 08:40:55.811723 133905819957056 max_utils.py:273] Attempting to initialize the jax distributed system...
I0421 08:41:04.853052 133905819957056 distributed.py:149] Starting JAX distributed service on [::]:8482
I0421 08:41:04.855356 133905819957056 distributed.py:172] Connecting to JAX distributed service on mt-03-dropout-cqspw-slice-job-0-0.mt-03-dropout-cqspw:8482
I0421 08:41:06.789904 133905819957056 max_utils.py:284] Jax distributed system initialized!
I0421 08:41:12.917796 133905819957056 max_utils.py:800] System Information: Jax Version: 0.9.2
I0421 08:41:12.917896 133905819957056 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0421 08:41:12.917936 133905819957056 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
I0421 08:41:12.917970 133905819957056 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0421 08:41:13.976401 133905819957056 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 08:41:13.976987 133905819957056 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 08:41:13.977164 133905819957056 checkpointing.py:688] Setting up checkpoint logger...
I0421 08:41:13.977213 133905819957056 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0421 08:41:13.977255 133905819957056 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0421 08:41:13.977570 133905819957056 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 0x79c8a933bad0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0421 08:41:16.453767 133905819957056 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0421 08:41:16.454040 133905819957056 checkpoint_manager.py:708] [process=7][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79c79ce910d0>}, handler_registry=None
I0421 08:41:16.454277 133905819957056 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79c79ce910d0>` 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`.
I0421 08:41:16.454325 133905819957056 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79c79cecc3e0>` 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`.
I0421 08:41:16.454361 133905819957056 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 0x79c79ce910d0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79c79ce910d0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79c79cecc3e0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79c79cecc3e0>}).
I0421 08:41:16.454673 133905819957056 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0421 08:41:16.454760 133905819957056 async_checkpointer.py:192] [process=7][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x79b45c397740> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0421 08:41:18.006910 133905819957056 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_03_dropout/checkpoints
I0421 08:41:18.037667 133905819957056 checkpoint_manager.py:929] [process=7][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_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_03_dropout/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x79c79ce91820>
I0421 08:41:18.037823 133905819957056 checkpointing.py:302] Checkpoint manager created!
I0421 08:41:18.388566 133905819957056 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0421 08:41:18.388666 133905819957056 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None).
I0421 08:41:18.770828 133905819957056 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 08:41:18.770918 133905819957056 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None).
I0421 08:41:18.787226 133905819957056 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 08:41:18.787285 133905819957056 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None).
I0421 08:41:18.810788 133905819957056 attentions.py:1154] attentions/query Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 08:41:18.810860 133905819957056 attentions.py:1154] attentions/query Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0421 08:41:18.827144 133905819957056 attentions.py:1155] attentions/key Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 08:41:18.827205 133905819957056 attentions.py:1155] attentions/key Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0421 08:41:18.843492 133905819957056 attentions.py:1156] attentions/value Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 08:41:18.843552 133905819957056 attentions.py:1156] attentions/value Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0421 08:41:18.868918 133905819957056 attentions.py:1197] attentions/out Logical: bfloat16[32,128,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0421 08:41:18.868987 133905819957056 attentions.py:1197] attentions/out Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0421 08:41:18.894687 133905819957056 linears.py:525] linears/x Logical: bfloat16[32,128,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0421 08:41:18.894773 133905819957056 linears.py:525] linears/x Physical: bfloat16[32,128,7168]....................................... ('fsdp', None, None).
I0421 08:41:19.116545 133905819957056 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0421 08:41:19.116651 133905819957056 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0421 08:41:20.555030 133905819957056 maxtext_utils.py:1821] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0421 08:41:20.555154 133905819957056 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)
I0421 08:41:20.555205 133905819957056 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)
I0421 08:41:20.555261 133905819957056 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')
I0421 08:41:20.555332 133905819957056 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)
I0421 08:41:20.555372 133905819957056 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)
I0421 08:41:20.555424 133905819957056 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)
I0421 08:41:20.555476 133905819957056 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')
I0421 08:41:20.555516 133905819957056 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)
I0421 08:41:20.555552 133905819957056 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)
I0421 08:41:20.555599 133905819957056 maxtext_utils.py:1821] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0421 08:41:20.555648 133905819957056 maxtext_utils.py:1821] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0421 08:41:21.074149 133905819957056 train.py:157] train/xent Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length').
I0421 08:41:21.074244 133905819957056 train.py:157] train/xent Physical: float32[32,128]............................................. ('fsdp', None).
I0421 08:41:21.089762 133905819957056 train.py:164] train/z_loss Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length').
I0421 08:41:21.089826 133905819957056 train.py:164] train/z_loss Physical: float32[32,128]............................................. ('fsdp', None).
I0421 08:41:24.618466 133905819957056 max_utils.py:791] Total memory size: 0.8 GB, Output size: 0.4 GB, Temp size: 0.4 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0421 08:41:24.619267 133905819957056 metric_logger.py:301] number parameters: 1.104 billion
I0421 08:41:28.868542 133905819957056 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0421 08:41:28.942449 133905819957056 checkpointing.py:798] Waited 0.07389211654663086 seconds for step 0 to finish before starting checkpointing.
I0421 08:41:28.945047 133905819957056 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 08:41:28.946810 133905819957056 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 0
I0421 08:41:28.948220 133905819957056 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_03_dropout/checkpoints/0.
I0421 08:41:30.104370 133905819957056 signaling_client.py:364] Using JaxDistributedSignalingClient
I0421 08:41:30.105321 133905819957056 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 08:41:30.105380 133905819957056 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
I0421 08:41:30.380980 133905819957056 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.276685s
I0421 08:41:30.381156 133905819957056 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 5.469 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28205299377441406 s) (per-host)
I0421 08:41:30.381209 133905819957056 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.282115s (batch_requests_ready=0.002197s, total_serialization_initiated=0.279849s, others=0.000070s)
I0421 08:41:30.381296 133905819957056 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.286331s (all_items=0.000018s, per_item={'items': '0.00001788'}, temp_paths=0.286313)
I0421 08:41:30.382135 133905819957056 event_tracking.py:125] [process=7] [async] Finished blocking save in 1.44 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_03_dropout/checkpoints/0.
I0421 08:41:30.382474 133777116342016 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 09:01:30.382436
I0421 08:41:30.396235 133905819957056 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 08:41:30.396523 133776034207488 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save.
I0421 08:41:30.396674 133905819957056 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776760888.9450293, 'wait_for_prev_duration_secs': 6.246566772460938e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776760888.9468489, 'checkpointer_blocking_duration_secs': 1.4357316493988037, 'get_old_steps_start_time': 1776760890.3826041, 'get_old_steps_duration_secs': 3.0279159545898438e-05, 'checkpoint_manager_blocking_start_time': 1776760888.9429529, 'checkpoint_manager_blocking_duration_secs': 1.4536855220794678}
I0421 08:41:30.396792 133905819957056 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0421 08:41:30.396844 133905819957056 max_utils.py:750]
Memstats: After params initialized:
I0421 08:41:30.396898 133905819957056 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_26(process=7,(2,6,0,0))
I0421 08:41:30.396928 133905819957056 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_27(process=7,(3,6,0,0))
I0421 08:41:30.396957 133905819957056 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_30(process=7,(2,7,0,0))
I0421 08:41:30.396983 133905819957056 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_31(process=7,(3,7,0,0))
I0421 08:41:30.712443 133905819957056 metric_logger.py:196] completed step: 0, seconds: 4.249, TFLOP/s/device: 0.188, Tokens/s/device: 30.124, total_weights: 4096, loss: 10.889, lm_loss: 10.889, perplexity: 53560.699
I0421 08:41:30.809458 133905819957056 metric_logger.py:196] completed step: 1, seconds: 1.842, TFLOP/s/device: 0.435, Tokens/s/device: 69.473, total_weights: 4096, loss: 10.902, lm_loss: 10.902, perplexity: 54298.523
I0421 08:41:31.211260 133905819957056 metric_logger.py:196] completed step: 2, seconds: 0.027, TFLOP/s/device: 30.078, Tokens/s/device: 4807.151, total_weights: 4096, loss: 9.956, lm_loss: 9.956, perplexity: 21082.051
I0421 08:41:31.283339 133905819957056 metric_logger.py:196] completed step: 3, seconds: 0.402, TFLOP/s/device: 1.993, Tokens/s/device: 318.589, total_weights: 4096, loss: 9.139, lm_loss: 9.139, perplexity: 9314.691
I0421 08:41:31.425138 133905819957056 metric_logger.py:196] completed step: 4, seconds: 0.077, TFLOP/s/device: 10.465, Tokens/s/device: 1672.525, total_weights: 4096, loss: 8.459, lm_loss: 8.459, perplexity: 4715.493
I0421 08:41:31.431033 133905819957056 metric_logger.py:196] completed step: 5, seconds: 0.072, TFLOP/s/device: 11.132, Tokens/s/device: 1779.236, total_weights: 4096, loss: 7.924, lm_loss: 7.924, perplexity: 2762.373
I0421 08:41:33.702154 2803 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0421 08:41:36.098454 133776042600192 array_metadata_store.py:203] [process=7][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_03_dropout/checkpoints/0/items/array_metadatas/process_7
I0421 08:41:41.270679 133905819957056 metric_logger.py:196] completed step: 6, seconds: 0.142, TFLOP/s/device: 5.631, Tokens/s/device: 899.906, total_weights: 4096, loss: 7.553, lm_loss: 7.553, perplexity: 1906.617
I0421 08:41:41.341321 133905819957056 metric_logger.py:196] completed step: 7, seconds: 9.770, TFLOP/s/device: 0.082, Tokens/s/device: 13.101, total_weights: 4096, loss: 7.267, lm_loss: 7.267, perplexity: 1431.653
I0421 08:41:41.412132 133905819957056 metric_logger.py:196] completed step: 8, seconds: 0.075, TFLOP/s/device: 10.725, Tokens/s/device: 1714.163, total_weights: 4096, loss: 7.111, lm_loss: 7.111, perplexity: 1225.983
I0421 08:41:41.482205 133905819957056 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0421 08:41:41.482763 133905819957056 checkpointing.py:798] Waited 0.0005526542663574219 seconds for step 9 to finish before starting checkpointing.
I0421 08:41:41.484908 133905819957056 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 08:42:01.873571 133777116342016 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 49.713 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 31.77442717552185 s) (per-host)
I0421 08:42:01.873695 133777116342016 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 31.490579s.
I0421 08:42:11.878255 133777116342016 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 41.495123s.
I0421 08:42:11.878469 133776034207488 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 08:42:11.878528 133776034207488 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save.
I0421 08:42:11.878576 133776034207488 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0421 08:42:11.880323 133776034207488 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0421 08:42:11.880500 133905819957056 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0421 08:42:11.880634 133905819957056 checkpoint_manager.py:1452] Waiting for previous save to complete took 30.395724 seconds. If this number is high, consider checkpointing less frequently.
I0421 08:42:11.882411 133905819957056 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 9
I0421 08:42:11.884436 133905819957056 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_03_dropout/checkpoints/9.
I0421 08:42:12.590913 133905819957056 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 08:42:12.591007 133905819957056 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
I0421 08:42:12.625262 133905819957056 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.035441s
I0421 08:42:12.625430 133905819957056 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 39.727 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.038829803466796875 s) (per-host)
I0421 08:42:12.625482 133905819957056 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.038890s (batch_requests_ready=0.001750s, total_serialization_initiated=0.037071s, others=0.000069s)
I0421 08:42:12.625569 133905819957056 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.043298s (all_items=0.000016s, per_item={'items': '0.00001574'}, temp_paths=0.043282)
I0421 08:42:12.626249 133905819957056 event_tracking.py:125] [process=7] [async] Finished blocking save in 0.74 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_03_dropout/checkpoints/9.
I0421 08:42:12.626587 133776034207488 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 09:02:12.626549
I0421 08:42:12.629117 133905819957056 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 08:42:12.629344 133776042600192 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save.
I0421 08:42:12.629469 133905819957056 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776760901.4848804, 'wait_for_prev_duration_secs': 30.395724058151245, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776760931.8824492, 'checkpointer_blocking_duration_secs': 0.7442970275878906, 'get_old_steps_start_time': 1776760932.6267693, 'get_old_steps_duration_secs': 3.0040740966796875e-05, 'checkpoint_manager_blocking_start_time': 1776760901.4829826, 'checkpoint_manager_blocking_duration_secs': 31.146450996398926}
I0421 08:42:12.629575 133905819957056 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0421 08:42:12.629619 133905819957056 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 08:42:18.033077 133753447827200 array_metadata_store.py:203] [process=7][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_03_dropout/checkpoints/9/items/array_metadatas/process_7
I0421 08:42:52.567149 133776034207488 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 39.509 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 39.98048663139343 s) (per-host)
I0421 08:42:52.567276 133776034207488 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 39.940575s.
I0421 08:43:02.061441 133776034207488 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 49.434724s.
I0421 08:43:02.061667 133776042600192 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 08:43:02.061755 133776042600192 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save.
I0421 08:43:02.061805 133776042600192 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0421 08:43:02.063679 133776042600192 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0421 08:43:02.063940 133905819957056 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0421 08:43:02.064098 133905819957056 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 08:43:02.065078 133905819957056 metric_logger.py:196] completed step: 9, seconds: 0.071, TFLOP/s/device: 11.345, Tokens/s/device: 1813.237, total_weights: 4096, loss: 6.981, lm_loss: 6.981, perplexity: 1075.867
Per train step:
Total TFLOPs: 0.80
split as 99.60% learnable weight flops and 0.40% attention flops
XPK End: Tue Apr 21 08:43:11 UTC 2026
EXIT_CODE=0