XPK Start: Mon Apr 20 08:16:14 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 08:16:38.577218: 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 08:16:38.758350 140063469889344 max_utils.py:273] Attempting to initialize the jax distributed system...
I0420 08:16:47.799218 140063469889344 distributed.py:149] Starting JAX distributed service on [::]:8482
I0420 08:16:47.801502 140063469889344 distributed.py:172] Connecting to JAX distributed service on mt-01-base-24ais-slice-job-0-0.mt-01-base-24ais:8482
I0420 08:16:49.944478 140063469889344 max_utils.py:284] Jax distributed system initialized!
I0420 08:16:56.031857 140063469889344 max_utils.py:800] System Information: Jax Version: 0.9.2
I0420 08:16:56.031959 140063469889344 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0420 08:16:56.032001 140063469889344 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 08:16:56.032056 140063469889344 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0420 08:16:56.728852 140063469889344 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 08:16:56.729459 140063469889344 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 08:16:56.729639 140063469889344 checkpointing.py:688] Setting up checkpoint logger...
I0420 08:16:56.729690 140063469889344 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0420 08:16:56.729733 140063469889344 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0420 08:16:56.730076 140063469889344 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 0x7f6278967920>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0420 08:17:00.422601 140063469889344 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0420 08:17:00.422815 140063469889344 checkpoint_manager.py:708] [process=2][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7f4e4458eab0>}, handler_registry=None
I0420 08:17:00.423063 140063469889344 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7f4e4458eab0>` 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 08:17:00.423114 140063469889344 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f4e44590890>` 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 08:17:00.423150 140063469889344 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 0x7f4e4458eab0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7f4e4458eab0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f4e44590890>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7f4e44590890>}).
I0420 08:17:00.423482 140063469889344 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0420 08:17:00.423551 140063469889344 async_checkpointer.py:192] [process=2][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7f4e0c1042c0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0420 08:17:01.607253 140063469889344 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_01_base/checkpoints
I0420 08:17:01.899984 140063469889344 checkpoint_manager.py:929] [process=2][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_01_base/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7f4e4458fef0>
I0420 08:17:01.900207 140063469889344 checkpointing.py:302] Checkpoint manager created!
I0420 08:17:02.834614 140063469889344 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0420 08:17:02.834721 140063469889344 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 08:17:03.219830 140063469889344 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 08:17:03.219932 140063469889344 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 08:17:03.237009 140063469889344 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 08:17:03.237083 140063469889344 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 08:17:03.261292 140063469889344 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 08:17:03.261360 140063469889344 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 08:17:03.278809 140063469889344 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 08:17:03.278885 140063469889344 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 08:17:03.296045 140063469889344 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 08:17:03.296104 140063469889344 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 08:17:03.321603 140063469889344 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0420 08:17:03.321672 140063469889344 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 08:17:03.342911 140063469889344 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0420 08:17:03.342975 140063469889344 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0420 08:17:03.567368 140063469889344 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0420 08:17:03.567487 140063469889344 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0420 08:17:05.007499 140063469889344 maxtext_utils.py:1821] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0420 08:17:05.007626 140063469889344 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 08:17:05.007679 140063469889344 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 08:17:05.007738 140063469889344 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 08:17:05.007793 140063469889344 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 08:17:05.007832 140063469889344 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 08:17:05.007884 140063469889344 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 08:17:05.007935 140063469889344 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 08:17:05.007976 140063469889344 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 08:17:05.008011 140063469889344 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 08:17:05.008074 140063469889344 maxtext_utils.py:1821] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0420 08:17:05.008123 140063469889344 maxtext_utils.py:1821] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0420 08:17:05.504979 140063469889344 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 08:17:05.505082 140063469889344 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 08:17:05.520758 140063469889344 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 08:17:05.520821 140063469889344 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 08:17:16.447851 140063469889344 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 08:17:16.448734 140063469889344 metric_logger.py:301] number parameters: 1.104 billion
I0420 08:17:27.970645 140063469889344 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0420 08:17:28.125506 140063469889344 checkpointing.py:798] Waited 0.15483975410461426 seconds for step 0 to finish before starting checkpointing.
I0420 08:17:28.127835 140063469889344 checkpoint_manager.py:2009] [process=2][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 08:17:28.129601 140063469889344 checkpoint_manager.py:1512] [process=2] Saving checkpoint at step 0
I0420 08:17:28.130996 140063469889344 event_tracking.py:70] [process=2] [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_01_base/checkpoints/0.
I0420 08:17:28.888595 140063469889344 signaling_client.py:364] Using JaxDistributedSignalingClient
I0420 08:17:28.889554 140063469889344 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0420 08:17:28.889612 140063469889344 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 08:17:29.170296 140063469889344 base_pytree_checkpoint_handler.py:154] [process=2][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.281774s
I0420 08:17:29.170475 140063469889344 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/blocking_gbytes_per_sec: 5.369 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28729963302612305 s) (per-host)
I0420 08:17:29.170528 140063469889344 base_pytree_checkpoint_handler.py:768] [process=2][thread=MainThread] Initiated Pytree async_save. Time taken: 0.287363s (batch_requests_ready=0.002233s, total_serialization_initiated=0.285058s, others=0.000072s)
I0420 08:17:29.170615 140063469889344 composite_checkpoint_handler.py:715] [process=2][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.291596s (all_items=0.000018s, per_item={'items': '0.00001764'}, temp_paths=0.291578)
I0420 08:17:29.171402 140063469889344 event_tracking.py:125] [process=2] [async] Finished blocking save in 1.04 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_01_base/checkpoints/0.
I0420 08:17:29.171736 139935886591744 async_checkpointer.py:76] [process=2][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 08:37:29.171697
I0420 08:17:29.176785 140063469889344 checkpoint_manager.py:1560] [process=2][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 08:17:29.177087 139934279112448 async_checkpointer.py:280] [process=2][thread=save_finalize] Waiting for background save thread=async_save.
I0420 08:17:29.177242 140063469889344 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_01_base/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776673048.1278162, 'wait_for_prev_duration_secs': 6.175041198730469e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776673048.1296458, 'checkpointer_blocking_duration_secs': 1.0422415733337402, 'get_old_steps_start_time': 1776673049.1719124, 'get_old_steps_duration_secs': 2.956390380859375e-05, 'checkpoint_manager_blocking_start_time': 1776673048.1259534, 'checkpoint_manager_blocking_duration_secs': 1.0512502193450928}
I0420 08:17:29.177353 140063469889344 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0420 08:17:29.177407 140063469889344 max_utils.py:750]
Memstats: After params initialized:
I0420 08:17:29.177456 140063469889344 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_8(process=2,(0,2,0,0))
I0420 08:17:29.177490 140063469889344 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_9(process=2,(1,2,0,0))
I0420 08:17:29.177518 140063469889344 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_12(process=2,(0,3,0,0))
I0420 08:17:29.177543 140063469889344 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_13(process=2,(1,3,0,0))
I0420 08:17:29.486388 140063469889344 metric_logger.py:196] completed step: 0, seconds: 11.522, TFLOP/s/device: 1.179, Tokens/s/device: 177.750, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0420 08:17:29.667855 140063469889344 metric_logger.py:196] completed step: 1, seconds: 1.514, TFLOP/s/device: 8.973, Tokens/s/device: 1352.486, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0420 08:17:30.073916 140063469889344 metric_logger.py:196] completed step: 2, seconds: 0.034, TFLOP/s/device: 394.745, Tokens/s/device: 59500.291, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0420 08:17:30.221229 140063469889344 metric_logger.py:196] completed step: 3, seconds: 0.406, TFLOP/s/device: 33.449, Tokens/s/device: 5041.789, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0420 08:17:30.516310 140063469889344 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.819, Tokens/s/device: 13387.808, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0420 08:17:30.522187 140063469889344 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.294, Tokens/s/device: 13911.626, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0420 08:17:32.956984 2813 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0420 08:17:34.994460 139934287505152 array_metadata_store.py:203] [process=2][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_01_base/checkpoints/0/items/array_metadatas/process_2
I0420 08:17:52.429199 140063469889344 metric_logger.py:196] completed step: 6, seconds: 0.295, TFLOP/s/device: 45.985, Tokens/s/device: 6931.330, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0420 08:17:52.576585 140063469889344 metric_logger.py:196] completed step: 7, seconds: 21.760, TFLOP/s/device: 0.624, Tokens/s/device: 94.117, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0420 08:17:52.723927 140063469889344 metric_logger.py:196] completed step: 8, seconds: 0.153, TFLOP/s/device: 89.088, Tokens/s/device: 13428.275, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0420 08:17:52.870836 140063469889344 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0420 08:17:52.871516 140063469889344 checkpointing.py:798] Waited 0.0006992816925048828 seconds for step 9 to finish before starting checkpointing.
I0420 08:17:52.873572 140063469889344 checkpoint_manager.py:2020] [process=2][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 08:18:07.125854 139935886591744 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/gbytes_per_sec: 41.305 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 38.24263095855713 s) (per-host)
I0420 08:18:07.125977 139935886591744 async_checkpointer.py:90] [process=2][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.954130s.
I0420 08:18:15.543880 139935886591744 async_checkpointer.py:160] [process=2][thread=async_save] Background save thread done. Time taken: 46.372017s.
I0420 08:18:15.544273 139934279112448 async_checkpointer.py:288] [process=2][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 08:18:15.544417 139934279112448 async_checkpointer.py:298] [process=2][thread=save_finalize] No errors found in background save thread=async_save.
I0420 08:18:15.544477 139934279112448 checkpoint_manager.py:2137] [process=2][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0420 08:18:15.546506 139934279112448 checkpoint_manager.py:2146] [process=2][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0420 08:18:15.546702 140063469889344 checkpoint_manager.py:2032] [process=2][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0420 08:18:15.546835 140063469889344 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.673267 seconds. If this number is high, consider checkpointing less frequently.
I0420 08:18:15.548678 140063469889344 checkpoint_manager.py:1512] [process=2] Saving checkpoint at step 9
I0420 08:18:15.550616 140063469889344 event_tracking.py:70] [process=2] [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_01_base/checkpoints/9.
I0420 08:18:16.280001 140063469889344 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0420 08:18:16.280109 140063469889344 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 08:18:16.320433 140063469889344 base_pytree_checkpoint_handler.py:154] [process=2][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.041513s
I0420 08:18:16.320610 140063469889344 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/blocking_gbytes_per_sec: 34.258 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.045027732849121094 s) (per-host)
I0420 08:18:16.320663 140063469889344 base_pytree_checkpoint_handler.py:768] [process=2][thread=MainThread] Initiated Pytree async_save. Time taken: 0.045091s (batch_requests_ready=0.001838s, total_serialization_initiated=0.043181s, others=0.000072s)
I0420 08:18:16.320757 140063469889344 composite_checkpoint_handler.py:715] [process=2][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.049220s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=0.049205)
I0420 08:18:16.321498 140063469889344 event_tracking.py:125] [process=2] [async] Finished blocking save in 0.77 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_01_base/checkpoints/9.
I0420 08:18:16.321797 139934279112448 async_checkpointer.py:76] [process=2][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 08:38:16.321762
I0420 08:18:16.323794 140063469889344 checkpoint_manager.py:1560] [process=2][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 08:18:16.324027 139930514507520 async_checkpointer.py:280] [process=2][thread=save_finalize] Waiting for background save thread=async_save.
I0420 08:18:16.324128 140063469889344 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_01_base/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776673072.873539, 'wait_for_prev_duration_secs': 22.673267126083374, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776673095.5487165, 'checkpointer_blocking_duration_secs': 0.7732281684875488, 'get_old_steps_start_time': 1776673096.3219693, 'get_old_steps_duration_secs': 3.147125244140625e-05, 'checkpoint_manager_blocking_start_time': 1776673072.8717597, 'checkpoint_manager_blocking_duration_secs': 23.45233392715454}
I0420 08:18:16.324295 140063469889344 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0420 08:18:16.324342 140063469889344 checkpoint_manager.py:2020] [process=2][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 08:18:21.764243 139934287505152 array_metadata_store.py:203] [process=2][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_01_base/checkpoints/9/items/array_metadatas/process_2
I0420 08:18:59.389000 139934279112448 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/gbytes_per_sec: 36.638 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.113378286361694 s) (per-host)
I0420 08:18:59.389142 139934279112448 async_checkpointer.py:90] [process=2][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.067229s.
I0420 08:19:07.397037 139934279112448 async_checkpointer.py:160] [process=2][thread=async_save] Background save thread done. Time taken: 51.075090s.
I0420 08:19:07.397287 139930514507520 async_checkpointer.py:288] [process=2][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 08:19:07.397344 139930514507520 async_checkpointer.py:298] [process=2][thread=save_finalize] No errors found in background save thread=async_save.
I0420 08:19:07.397397 139930514507520 checkpoint_manager.py:2137] [process=2][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0420 08:19:07.399237 139930514507520 checkpoint_manager.py:2146] [process=2][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0420 08:19:07.399426 140063469889344 checkpoint_manager.py:2032] [process=2][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0420 08:19:07.399582 140063469889344 checkpoint_manager.py:2009] [process=2][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 08:19:07.400518 140063469889344 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.276, Tokens/s/device: 13908.886, 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 08:19:19 UTC 2026
EXIT_CODE=0