XPK Start: Wed Apr 22 13:13:34 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 13:13:59.191425: 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 13:13:59.372208 133127924594496 max_utils.py:273] Attempting to initialize the jax distributed system...
I0422 13:14:08.412675 133127924594496 distributed.py:149] Starting JAX distributed service on [::]:8482
I0422 13:14:08.414978 133127924594496 distributed.py:172] Connecting to JAX distributed service on mt-10-shardy-true-eodfn-slice-job-0-0.mt-10-shardy-true-eodfn:8482
I0422 13:14:10.183913 133127924594496 max_utils.py:284] Jax distributed system initialized!
I0422 13:14:16.425072 133127924594496 max_utils.py:800] System Information: Jax Version: 0.9.2
I0422 13:14:16.425176 133127924594496 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0422 13:14:16.425217 133127924594496 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 13:14:16.425254 133127924594496 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0422 13:14:17.125069 133127924594496 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0422 13:14:17.125644 133127924594496 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0422 13:14:17.125820 133127924594496 checkpointing.py:688] Setting up checkpoint logger...
I0422 13:14:17.125869 133127924594496 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0422 13:14:17.125911 133127924594496 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0422 13:14:17.126260 133127924594496 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 0x7913a1b1b7a0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0422 13:14:20.081327 133127924594496 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0422 13:14:20.081632 133127924594496 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 0x78ff58219670>}, handler_registry=None
I0422 13:14:20.081871 133127924594496 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78ff58219670>` 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 13:14:20.081918 133127924594496 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78ff5821ecf0>` 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 13:14:20.081969 133127924594496 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 0x78ff58219670>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78ff58219670>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78ff5821ecf0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78ff5821ecf0>}).
I0422 13:14:20.082279 133127924594496 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0422 13:14:20.082350 133127924594496 async_checkpointer.py:192] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x78ff5803c540> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0422 13:14:21.126995 133127924594496 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_10_shardy_true/checkpoints
I0422 13:14:21.155186 133127924594496 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_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_10_shardy_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x78fe50570ad0>
I0422 13:14:21.155317 133127924594496 checkpointing.py:302] Checkpoint manager created!
I0422 13:14:22.089469 133127924594496 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0422 13:14:22.089573 133127924594496 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 13:14:22.471009 133127924594496 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 13:14:22.471094 133127924594496 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 13:14:22.487673 133127924594496 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 13:14:22.487734 133127924594496 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 13:14:22.512000 133127924594496 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 13:14:22.512067 133127924594496 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 13:14:22.528831 133127924594496 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 13:14:22.528893 133127924594496 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 13:14:22.545396 133127924594496 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 13:14:22.545455 133127924594496 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 13:14:22.570142 133127924594496 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0422 13:14:22.570208 133127924594496 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 13:14:22.591588 133127924594496 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0422 13:14:22.591652 133127924594496 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0422 13:14:22.810683 133127924594496 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0422 13:14:22.810789 133127924594496 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0422 13:14:24.239894 133127924594496 maxtext_utils.py:1827] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0422 13:14:24.240041 133127924594496 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 13:14:24.240096 133127924594496 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 13:14:24.240178 133127924594496 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 13:14:24.240234 133127924594496 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 13:14:24.240291 133127924594496 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 13:14:24.240364 133127924594496 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 13:14:24.240422 133127924594496 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 13:14:24.240464 133127924594496 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 13:14:24.240504 133127924594496 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 13:14:24.240551 133127924594496 maxtext_utils.py:1827] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0422 13:14:24.240598 133127924594496 maxtext_utils.py:1827] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0422 13:14:24.734084 133127924594496 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0422 13:14:24.734176 133127924594496 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0422 13:14:24.749712 133127924594496 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0422 13:14:24.749773 133127924594496 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0422 13:14:35.697101 133127924594496 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 13:14:35.697910 133127924594496 metric_logger.py:301] number parameters: 1.104 billion
I0422 13:14:47.380977 133127924594496 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0422 13:14:47.534380 133127924594496 checkpointing.py:798] Waited 0.1533830165863037 seconds for step 0 to finish before starting checkpointing.
I0422 13:14:47.536918 133127924594496 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 13:14:47.538459 133127924594496 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 0
I0422 13:14:47.539837 133127924594496 event_tracking.py:70] [process=3] [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_10_shardy_true/checkpoints/0.
I0422 13:14:47.878256 133127924594496 signaling_client.py:364] Using JaxDistributedSignalingClient
I0422 13:14:47.879232 133127924594496 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0422 13:14:47.879288 133127924594496 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 13:14:48.160273 133127924594496 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.282092s
I0422 13:14:48.160442 133127924594496 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 5.363 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2876124382019043 s) (per-host)
I0422 13:14:48.160500 133127924594496 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.287680s (batch_requests_ready=0.002312s, total_serialization_initiated=0.285292s, others=0.000075s)
I0422 13:14:48.160597 133127924594496 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.291792s (all_items=0.000018s, per_item={'items': '0.00001764'}, temp_paths=0.291775)
I0422 13:14:48.161407 133127924594496 event_tracking.py:125] [process=3] [async] Finished blocking save in 0.62 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_10_shardy_true/checkpoints/0.
I0422 13:14:48.161767 132999375664896 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 13:34:48.161729
I0422 13:14:48.164906 133127924594496 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 13:14:48.165226 132997252585216 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0422 13:14:48.165380 133127924594496 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_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776863687.5368996, 'wait_for_prev_duration_secs': 7.295608520507812e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776863687.5385032, 'checkpointer_blocking_duration_secs': 0.6233694553375244, 'get_old_steps_start_time': 1776863688.1618986, 'get_old_steps_duration_secs': 4.172325134277344e-05, 'checkpoint_manager_blocking_start_time': 1776863687.5349226, 'checkpoint_manager_blocking_duration_secs': 0.630420446395874}
I0422 13:14:48.165481 133127924594496 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0422 13:14:48.165531 133127924594496 max_utils.py:750]
Memstats: After params initialized:
I0422 13:14:48.165579 133127924594496 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_10(process=3,(2,2,0,0))
I0422 13:14:48.165612 133127924594496 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_11(process=3,(3,2,0,0))
I0422 13:14:48.165640 133127924594496 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_14(process=3,(2,3,0,0))
I0422 13:14:48.165666 133127924594496 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_15(process=3,(3,3,0,0))
I0422 13:14:48.481591 133127924594496 metric_logger.py:196] completed step: 0, seconds: 11.683, TFLOP/s/device: 1.163, Tokens/s/device: 175.299, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0422 13:14:48.665348 133127924594496 metric_logger.py:196] completed step: 1, seconds: 1.099, TFLOP/s/device: 12.364, Tokens/s/device: 1863.573, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0422 13:14:49.068020 133127924594496 metric_logger.py:196] completed step: 2, seconds: 0.037, TFLOP/s/device: 368.745, Tokens/s/device: 55581.187, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0422 13:14:49.215473 133127924594496 metric_logger.py:196] completed step: 3, seconds: 0.403, TFLOP/s/device: 33.707, Tokens/s/device: 5080.676, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0422 13:14:49.510413 133127924594496 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.855, Tokens/s/device: 13393.237, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0422 13:14:49.516288 133127924594496 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.411, Tokens/s/device: 13929.225, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0422 13:14:51.170359 2807 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0422 13:14:53.237302 132998846027520 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_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_10_shardy_true/checkpoints/0/items/array_metadatas/process_3
I0422 13:15:11.389200 133127924594496 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.979, Tokens/s/device: 6930.438, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0422 13:15:11.536530 133127924594496 metric_logger.py:196] completed step: 7, seconds: 21.726, TFLOP/s/device: 0.625, Tokens/s/device: 94.265, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0422 13:15:11.683832 133127924594496 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.197, Tokens/s/device: 13444.672, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0422 13:15:11.830515 133127924594496 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0422 13:15:11.831194 133127924594496 checkpointing.py:798] Waited 0.0007116794586181641 seconds for step 9 to finish before starting checkpointing.
I0422 13:15:11.833233 133127924594496 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 13:15:24.426289 132999375664896 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 43.213 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.553425788879395 s) (per-host)
I0422 13:15:24.426396 132999375664896 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.263959s.
I0422 13:15:33.610598 132999375664896 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 45.448143s.
I0422 13:15:33.610912 132997252585216 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 13:15:33.611045 132997252585216 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0422 13:15:33.611095 132997252585216 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0422 13:15:33.612818 132997252585216 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0422 13:15:33.613003 133127924594496 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.
W0422 13:15:33.613152 133127924594496 checkpoint_manager.py:1452] Waiting for previous save to complete took 21.779917 seconds. If this number is high, consider checkpointing less frequently.
I0422 13:15:33.615015 133127924594496 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 9
I0422 13:15:33.617005 133127924594496 event_tracking.py:70] [process=3] [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_10_shardy_true/checkpoints/9.
I0422 13:15:33.911317 133127924594496 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0422 13:15:33.911411 133127924594496 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 13:15:33.949334 133127924594496 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.039173s
I0422 13:15:33.949506 133127924594496 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 36.100 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04273056983947754 s) (per-host)
I0422 13:15:33.949559 133127924594496 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.042793s (batch_requests_ready=0.001917s, total_serialization_initiated=0.040806s, others=0.000070s)
I0422 13:15:33.949652 133127924594496 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.046836s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.046821)
I0422 13:15:33.950363 133127924594496 event_tracking.py:125] [process=3] [async] Finished blocking save in 0.34 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_10_shardy_true/checkpoints/9.
I0422 13:15:33.950665 132997252585216 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 13:35:33.950632
I0422 13:15:33.952661 133127924594496 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 13:15:33.952985 132974058137344 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0422 13:15:33.953129 133127924594496 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_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776863711.8332045, 'wait_for_prev_duration_secs': 21.779917001724243, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776863733.6150515, 'checkpointer_blocking_duration_secs': 0.3357529640197754, 'get_old_steps_start_time': 1776863733.9508278, 'get_old_steps_duration_secs': 2.956390380859375e-05, 'checkpoint_manager_blocking_start_time': 1776863711.8314145, 'checkpoint_manager_blocking_duration_secs': 22.121678590774536}
I0422 13:15:33.953303 133127924594496 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0422 13:15:33.953350 133127924594496 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 13:15:39.448180 132998846027520 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_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_10_shardy_true/checkpoints/9/items/array_metadatas/process_3
I0422 13:16:16.600510 132997252585216 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 36.998 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.693692684173584 s) (per-host)
I0422 13:16:16.600633 132997252585216 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.649859s.
I0422 13:16:25.103344 132997252585216 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 51.152556s.
I0422 13:16:25.103611 132974058137344 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 13:16:25.103748 132974058137344 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0422 13:16:25.103807 132974058137344 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0422 13:16:25.105812 132974058137344 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0422 13:16:25.106018 133127924594496 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.
I0422 13:16:25.106168 133127924594496 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 13:16:25.107136 133127924594496 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.317, Tokens/s/device: 13915.029, 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 13:16:35 UTC 2026
EXIT_CODE=0