XPK Start: Tue Apr 21 16:34:46 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 16:35:10.882089: 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 16:35:11.413078 136528622532416 max_utils.py:273] Attempting to initialize the jax distributed system...
I0421 16:35:20.454905 136528622532416 distributed.py:149] Starting JAX distributed service on [::]:8482
I0421 16:35:20.457265 136528622532416 distributed.py:172] Connecting to JAX distributed service on mt-03-dropout-xrhz2-slice-job-0-0.mt-03-dropout-xrhz2:8482
I0421 16:35:21.728283 136528622532416 max_utils.py:284] Jax distributed system initialized!
I0421 16:35:27.992640 136528622532416 max_utils.py:800] System Information: Jax Version: 0.9.2
I0421 16:35:27.992748 136528622532416 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0421 16:35:27.992787 136528622532416 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 16:35:27.992820 136528622532416 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0421 16:35:28.685983 136528622532416 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 16:35:28.686573 136528622532416 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 16:35:28.686759 136528622532416 checkpointing.py:688] Setting up checkpoint logger...
I0421 16:35:28.686814 136528622532416 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0421 16:35:28.686857 136528622532416 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0421 16:35:28.687204 136528622532416 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 0x7c2b5474b170>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0421 16:35:31.621588 136528622532416 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0421 16:35:31.621858 136528622532416 checkpoint_manager.py:708] [process=3][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c17146505f0>}, handler_registry=None
I0421 16:35:31.622113 136528622532416 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c17146505f0>` 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 16:35:31.622164 136528622532416 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c171445aba0>` 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 16:35:31.622200 136528622532416 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 0x7c17146505f0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c17146505f0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c171445aba0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c171445aba0>}).
I0421 16:35:31.622523 136528622532416 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0421 16:35:31.622594 136528622532416 async_checkpointer.py:192] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7c15ec743ec0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0421 16:35:32.707500 136528622532416 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900_03_dropout/checkpoints
I0421 16:35:33.150696 136528622532416 checkpoint_manager.py:929] [process=3][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_161900/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900_03_dropout/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7c1714461be0>
I0421 16:35:33.150865 136528622532416 checkpointing.py:302] Checkpoint manager created!
I0421 16:35:33.503187 136528622532416 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0421 16:35:33.503290 136528622532416 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None).
I0421 16:35:33.891051 136528622532416 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 16:35:33.891143 136528622532416 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None).
I0421 16:35:33.907846 136528622532416 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 16:35:33.907906 136528622532416 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None).
I0421 16:35:33.931908 136528622532416 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 16:35:33.931989 136528622532416 attentions.py:1154] attentions/query Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0421 16:35:33.948607 136528622532416 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 16:35:33.948667 136528622532416 attentions.py:1155] attentions/key Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0421 16:35:33.965286 136528622532416 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 16:35:33.965348 136528622532416 attentions.py:1156] attentions/value Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0421 16:35:33.991370 136528622532416 attentions.py:1197] attentions/out Logical: bfloat16[32,128,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0421 16:35:33.991438 136528622532416 attentions.py:1197] attentions/out Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0421 16:35:34.017846 136528622532416 linears.py:525] linears/x Logical: bfloat16[32,128,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0421 16:35:34.017912 136528622532416 linears.py:525] linears/x Physical: bfloat16[32,128,7168]....................................... ('fsdp', None, None).
I0421 16:35:34.244614 136528622532416 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0421 16:35:34.244725 136528622532416 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0421 16:35:35.689456 136528622532416 maxtext_utils.py:1821] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0421 16:35:35.689579 136528622532416 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 16:35:35.689631 136528622532416 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 16:35:35.689716 136528622532416 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 16:35:35.689772 136528622532416 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 16:35:35.689811 136528622532416 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 16:35:35.689863 136528622532416 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 16:35:35.689915 136528622532416 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 16:35:35.689970 136528622532416 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 16:35:35.690007 136528622532416 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 16:35:35.690055 136528622532416 maxtext_utils.py:1821] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0421 16:35:35.690103 136528622532416 maxtext_utils.py:1821] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0421 16:35:36.228546 136528622532416 train.py:157] train/xent Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length').
I0421 16:35:36.228642 136528622532416 train.py:157] train/xent Physical: float32[32,128]............................................. ('fsdp', None).
I0421 16:35:36.244366 136528622532416 train.py:164] train/z_loss Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length').
I0421 16:35:36.244430 136528622532416 train.py:164] train/z_loss Physical: float32[32,128]............................................. ('fsdp', None).
I0421 16:35:39.790275 136528622532416 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 16:35:39.791014 136528622532416 metric_logger.py:301] number parameters: 1.104 billion
I0421 16:35:43.915116 136528622532416 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0421 16:35:44.306166 136528622532416 checkpointing.py:798] Waited 0.39102745056152344 seconds for step 0 to finish before starting checkpointing.
I0421 16:35:44.308687 136528622532416 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 16:35:44.310705 136528622532416 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 0
I0421 16:35:44.312074 136528622532416 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900_03_dropout/checkpoints/0.
I0421 16:35:44.604351 136528622532416 signaling_client.py:364] Using JaxDistributedSignalingClient
I0421 16:35:44.605281 136528622532416 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 16:35:44.605340 136528622532416 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 16:35:44.880546 136528622532416 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.276272s
I0421 16:35:44.880719 136528622532416 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 5.474 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2818136215209961 s) (per-host)
I0421 16:35:44.880768 136528622532416 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.281872s (batch_requests_ready=0.002262s, total_serialization_initiated=0.279543s, others=0.000067s)
I0421 16:35:44.880858 136528622532416 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.286390s (all_items=0.000018s, per_item={'items': '0.00001812'}, temp_paths=0.286371)
I0421 16:35:44.881677 136528622532416 event_tracking.py:125] [process=3] [async] Finished blocking save in 0.57 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900_03_dropout/checkpoints/0.
I0421 16:35:44.882078 136399917569792 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 16:55:44.882038
I0421 16:35:44.888706 136528622532416 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 16:35:44.889012 136398861883136 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0421 16:35:44.889174 136528622532416 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776789344.3086681, 'wait_for_prev_duration_secs': 6.818771362304688e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776789344.3107452, 'checkpointer_blocking_duration_secs': 0.5714881420135498, 'get_old_steps_start_time': 1776789344.8822594, 'get_old_steps_duration_secs': 2.9325485229492188e-05, 'checkpoint_manager_blocking_start_time': 1776789344.306714, 'checkpoint_manager_blocking_duration_secs': 0.5824205875396729}
I0421 16:35:44.889284 136528622532416 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0421 16:35:44.889333 136528622532416 max_utils.py:750]
Memstats: After params initialized:
I0421 16:35:44.889382 136528622532416 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_10(process=3,(2,2,0,0))
I0421 16:35:44.889416 136528622532416 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_11(process=3,(3,2,0,0))
I0421 16:35:44.889444 136528622532416 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_14(process=3,(2,3,0,0))
I0421 16:35:44.889475 136528622532416 max_utils.py:756] Using (GB) 0.41 / 31.25 (1.312000%) on TPU_15(process=3,(3,3,0,0))
I0421 16:35:45.212200 136528622532416 metric_logger.py:196] completed step: 0, seconds: 4.124, TFLOP/s/device: 0.194, Tokens/s/device: 31.038, total_weights: 4096, loss: 10.889, lm_loss: 10.889, perplexity: 53560.699
I0421 16:35:45.297588 136528622532416 metric_logger.py:196] completed step: 1, seconds: 1.296, TFLOP/s/device: 0.618, Tokens/s/device: 98.798, total_weights: 4096, loss: 10.902, lm_loss: 10.902, perplexity: 54298.523
I0421 16:35:45.733484 136528622532416 metric_logger.py:196] completed step: 2, seconds: 0.015, TFLOP/s/device: 53.495, Tokens/s/device: 8549.863, total_weights: 4096, loss: 9.956, lm_loss: 9.956, perplexity: 21082.051
I0421 16:35:45.804379 136528622532416 metric_logger.py:196] completed step: 3, seconds: 0.436, TFLOP/s/device: 1.836, Tokens/s/device: 293.396, total_weights: 4096, loss: 9.139, lm_loss: 9.139, perplexity: 9314.691
I0421 16:35:45.945954 136528622532416 metric_logger.py:196] completed step: 4, seconds: 0.076, TFLOP/s/device: 10.524, Tokens/s/device: 1681.997, total_weights: 4096, loss: 8.459, lm_loss: 8.459, perplexity: 4715.493
I0421 16:35:45.951761 136528622532416 metric_logger.py:196] completed step: 5, seconds: 0.071, TFLOP/s/device: 11.335, Tokens/s/device: 1811.645, total_weights: 4096, loss: 7.924, lm_loss: 7.924, perplexity: 2762.373
I0421 16:35:47.758502 2812 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0421 16:35:50.132599 136398870275840 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900_03_dropout/checkpoints/0/items/array_metadatas/process_3
I0421 16:35:55.384513 136528622532416 metric_logger.py:196] completed step: 6, seconds: 0.142, TFLOP/s/device: 5.634, Tokens/s/device: 900.457, total_weights: 4096, loss: 7.553, lm_loss: 7.553, perplexity: 1906.617
I0421 16:35:55.454942 136528622532416 metric_logger.py:196] completed step: 7, seconds: 9.362, TFLOP/s/device: 0.086, Tokens/s/device: 13.672, total_weights: 4096, loss: 7.267, lm_loss: 7.267, perplexity: 1431.653
I0421 16:35:55.526785 136528622532416 metric_logger.py:196] completed step: 8, seconds: 0.076, TFLOP/s/device: 10.567, Tokens/s/device: 1688.922, total_weights: 4096, loss: 7.111, lm_loss: 7.111, perplexity: 1225.983
I0421 16:35:55.598042 136528622532416 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0421 16:35:55.598675 136528622532416 checkpointing.py:798] Waited 0.0006499290466308594 seconds for step 9 to finish before starting checkpointing.
I0421 16:35:55.600790 136528622532416 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 16:36:16.800087 136399917569792 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 49.054 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 32.20114207267761 s) (per-host)
I0421 16:36:16.800215 136399917569792 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 31.918021s.
I0421 16:36:25.190424 136399917569792 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 40.308213s.
I0421 16:36:25.190728 136398861883136 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 16:36:25.190859 136398861883136 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0421 16:36:25.190913 136398861883136 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0421 16:36:25.193189 136398861883136 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0421 16:36:25.193394 136528622532416 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0421 16:36:25.193537 136528622532416 checkpoint_manager.py:1452] Waiting for previous save to complete took 29.592746 seconds. If this number is high, consider checkpointing less frequently.
I0421 16:36:25.195388 136528622532416 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 9
I0421 16:36:25.197532 136528622532416 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900_03_dropout/checkpoints/9.
I0421 16:36:26.345671 136528622532416 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 16:36:26.345766 136528622532416 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 16:36:26.381875 136528622532416 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.037222s
I0421 16:36:26.382075 136528622532416 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 37.839 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04076695442199707 s) (per-host)
I0421 16:36:26.382126 136528622532416 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.040834s (batch_requests_ready=0.001838s, total_serialization_initiated=0.038919s, others=0.000077s)
I0421 16:36:26.382218 136528622532416 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.045268s (all_items=0.000016s, per_item={'items': '0.00001574'}, temp_paths=0.045253)
I0421 16:36:26.382897 136528622532416 event_tracking.py:125] [process=3] [async] Finished blocking save in 1.19 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900_03_dropout/checkpoints/9.
I0421 16:36:26.383224 136398861883136 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 16:56:26.383187
I0421 16:36:26.385164 136528622532416 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 16:36:26.385447 136398870275840 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0421 16:36:26.385610 136528622532416 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776789355.6007612, 'wait_for_prev_duration_secs': 29.592745780944824, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776789385.195426, 'checkpointer_blocking_duration_secs': 1.1879377365112305, 'get_old_steps_start_time': 1776789386.3833878, 'get_old_steps_duration_secs': 3.0279159545898438e-05, 'checkpoint_manager_blocking_start_time': 1776789355.598909, 'checkpoint_manager_blocking_duration_secs': 30.786667346954346}
I0421 16:36:26.385718 136528622532416 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0421 16:36:26.385761 136528622532416 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 16:36:31.569440 136392536540928 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_161900_03_dropout/checkpoints/9/items/array_metadatas/process_3
I0421 16:37:06.345229 136398861883136 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 39.486 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 40.003886461257935 s) (per-host)
I0421 16:37:06.345359 136398861883136 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 39.962023s.
I0421 16:37:15.212365 136398861883136 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 48.829013s.
I0421 16:37:15.212654 136398870275840 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 16:37:15.212787 136398870275840 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0421 16:37:15.212842 136398870275840 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0421 16:37:15.214196 136398870275840 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0421 16:37:15.214372 136528622532416 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0421 16:37:15.214521 136528622532416 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 16:37:15.215471 136528622532416 metric_logger.py:196] completed step: 9, seconds: 0.070, TFLOP/s/device: 11.370, Tokens/s/device: 1817.253, 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 16:37:27 UTC 2026
EXIT_CODE=0