XPK Start: Mon Apr 20 19:33:34 UTC 2026
2026-04-20 19:33:38.160843: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:467] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
E0000 00:00:1776713618.173359 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1776713618.177049 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
W0000 00:00:1776713618.188740 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776713618.188764 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776713618.188766 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776713618.188768 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
2026-04-20 19:33:57.540489: 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 19:33:58.064277 134646368245568 max_utils.py:273] Attempting to initialize the jax distributed system...
INFO:2026-04-20 19:34:07,104:jax._src.distributed:140: Starting JAX distributed service on [::]:8482
I0420 19:34:07.104068 134646368245568 distributed.py:140] Starting JAX distributed service on [::]:8482
INFO:2026-04-20 19:34:07,106:jax._src.distributed:157: Connecting to JAX distributed service on mt-07-eval-vtfuh-slice-job-0-0.mt-07-eval-vtfuh:8482
I0420 19:34:07.106442 134646368245568 distributed.py:157] Connecting to JAX distributed service on mt-07-eval-vtfuh-slice-job-0-0.mt-07-eval-vtfuh:8482
I0420 19:34:08.559755 134646368245568 max_utils.py:284] Jax distributed system initialized!
I0420 19:34:15.493363 134646368245568 max_utils.py:800] System Information: Jax Version: 0.8.1
I0420 19:34:15.493469 134646368245568 max_utils.py:801] System Information: Jaxlib Version: 0.8.1
I0420 19:34:15.493510 134646368245568 max_utils.py:802] System Information: Jax Backend: PJRT C API
TFRT TPU v6 lite
Built on Nov 12 2025 14:16:36 (1762985796) cl/831091709
I0420 19:34:15.493546 134646368245568 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0420 19:34:16.111702 134646368245568 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 19:34:16.112316 134646368245568 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 19:34:16.112496 134646368245568 checkpointing.py:688] Setting up checkpoint logger...
I0420 19:34:16.112544 134646368245568 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0420 19:34:16.112588 134646368245568 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None
I0420 19:34:16.112943 134646368245568 base_pytree_checkpoint_handler.py:415] 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 0x7a7548956750>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0420 19:34:19.051460 134646368245568 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0420 19:34:19.051751 134646368245568 checkpoint_manager.py:709] [process=0][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a7407b65f40>}, handler_registry=None
I0420 19:34:19.052038 134646368245568 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a7407b65f40>` 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 19:34:19.052115 134646368245568 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a60c863a8d0>` 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 19:34:19.052167 134646368245568 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 0x7a7407b65f40>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a7407b65f40>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a60c863a8d0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a60c863a8d0>}).
I0420 19:34:19.052778 134646368245568 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33
I0420 19:34:19.052879 134646368245568 async_checkpointer.py:177] [process=0][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7a6028728180> timeout: 600 secs and primary_host=0 for async checkpoint writes
I0420 19:34:19.778242 134646368245568 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints
I0420 19:34:20.121521 134646368245568 checkpoint_manager.py:929] [process=0][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_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7a60c8636c90>
I0420 19:34:20.121694 134646368245568 checkpointing.py:302] Checkpoint manager created!
I0420 19:34:21.008546 134646368245568 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0420 19:34:21.008650 134646368245568 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 19:34:21.378128 134646368245568 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 19:34:21.378220 134646368245568 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 19:34:21.394116 134646368245568 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 19:34:21.394173 134646368245568 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 19:34:21.417061 134646368245568 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 19:34:21.417142 134646368245568 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 19:34:21.433051 134646368245568 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 19:34:21.433132 134646368245568 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 19:34:21.448994 134646368245568 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 19:34:21.449058 134646368245568 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 19:34:21.473763 134646368245568 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0420 19:34:21.473832 134646368245568 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 19:34:21.493418 134646368245568 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0420 19:34:21.493486 134646368245568 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0420 19:34:21.681589 134646368245568 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0420 19:34:21.681697 134646368245568 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0420 19:34:23.115361 134646368245568 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: PartitionSpec('norm',)
Physical: (None,)
I0420 19:34:23.115492 134646368245568 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wi_0/kernel
Shape: float32[2048,16,7168]
Logical: PartitionSpec('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0420 19:34:23.115549 134646368245568 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wi_1/kernel
Shape: float32[2048,16,7168]
Logical: PartitionSpec('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0420 19:34:23.115615 134646368245568 maxtext_utils.py:1740] params/params/decoder/layers/mlp/wo/kernel
Shape: float32[7168,16,2048]
Logical: PartitionSpec('mlp', 'layers', 'embed')
Physical: (None, None, 'fsdp')
I0420 19:34:23.115672 134646368245568 maxtext_utils.py:1740] params/params/decoder/layers/post_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: PartitionSpec('norm', 'layers')
Physical: (None, None)
I0420 19:34:23.115712 134646368245568 maxtext_utils.py:1740] params/params/decoder/layers/pre_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: PartitionSpec('norm', 'layers')
Physical: (None, None)
I0420 19:34:23.115767 134646368245568 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/key/kernel
Shape: float32[2048,16,16,128]
Logical: PartitionSpec('embed', 'layers', 'kv_heads', 'kv_head_dim')
Physical: ('fsdp', None, None, None)
I0420 19:34:23.115823 134646368245568 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/out/kernel
Shape: float32[16,16,128,2048]
Logical: PartitionSpec('heads', 'layers', 'kv', 'embed')
Physical: (None, None, None, 'fsdp')
I0420 19:34:23.115866 134646368245568 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/query/kernel
Shape: float32[2048,16,16,128]
Logical: PartitionSpec('embed', 'layers', 'q_heads', 'kv')
Physical: ('fsdp', None, None, None)
I0420 19:34:23.115905 134646368245568 maxtext_utils.py:1740] params/params/decoder/layers/self_attention/value/kernel
Shape: float32[2048,16,16,128]
Logical: PartitionSpec('embed', 'layers', 'kv_heads', 'kv_head_dim')
Physical: ('fsdp', None, None, None)
I0420 19:34:23.115956 134646368245568 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: PartitionSpec('embed', 'vocab')
Physical: ('fsdp', None)
I0420 19:34:23.116006 134646368245568 maxtext_utils.py:1740] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: PartitionSpec('vocab', 'embed')
Physical: (None, 'fsdp')
I0420 19:34:23.619385 134646368245568 train.py:158] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 19:34:23.619482 134646368245568 train.py:158] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 19:34:23.634421 134646368245568 train.py:165] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 19:34:23.634481 134646368245568 train.py:165] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 19:34:34.205836 134646368245568 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 19:34:34.211909 134646368245568 metric_logger.py:289] number parameters: 1.104 billion
I0420 19:34:45.865996 134646368245568 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0420 19:34:46.019502 134646368245568 checkpointing.py:798] Waited 0.1534900665283203 seconds for step 0 to finish before starting checkpointing.
I0420 19:34:46.021974 134646368245568 checkpoint_manager.py:2013] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 19:34:46.023782 134646368245568 checkpoint_manager.py:1518] [process=0] Saving checkpoint at step 0
I0420 19:34:46.026582 134646368245568 async_checkpointer.py:452] [process=0] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0.
I0420 19:34:46.300323 134646368245568 signaling_client.py:364] Using JaxDistributedSignalingClient
I0420 19:34:46.551647 134519382341376 atomicity.py:137] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0
I0420 19:34:46.569707 134646368245568 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array.
I0420 19:34:46.569802 134646368245568 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 19:34:46.809726 134646368245568 base_pytree_checkpoint_handler.py:153] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.240998s
I0420 19:34:46.810641 134646368245568 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.069 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5026779174804688 s) (per-host)
I0420 19:34:46.810708 134646368245568 base_pytree_checkpoint_handler.py:737] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.502763s (batch_requests_ready=0.256960s, total_serialization_initiated=0.244991s, others=0.000812s)
I0420 19:34:46.810785 134646368245568 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.506718s (all_items=0.000019s, per_item={'items': '0.00001907'}, temp_paths=0.506699)
I0420 19:34:46.811675 134519902426880 async_checkpointer.py:79] [process=0][thread=async_save] Background save thread started.
I0420 19:34:46.811782 134646368245568 async_checkpointer.py:561] Finished blocking save. Time taken: 0.787929s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0.
I0420 19:34:46.841617 134646368245568 checkpoint_manager.py:1566] [process=0][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 19:34:46.841912 134519373948672 async_checkpointer.py:265] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0420 19:34:46.842069 134646368245568 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776713686.0219567, 'wait_for_prev_duration_secs': 6.318092346191406e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776713686.0238206, 'checkpointer_blocking_duration_secs': 0.7880687713623047, 'get_old_steps_start_time': 1776713686.81191, 'get_old_steps_duration_secs': 2.5272369384765625e-05, 'checkpoint_manager_blocking_start_time': 1776713686.02014, 'checkpoint_manager_blocking_duration_secs': 0.8218910694122314}
I0420 19:34:46.842182 134646368245568 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0420 19:34:46.842234 134646368245568 max_utils.py:750]
Memstats: After params initialized:
I0420 19:34:46.842287 134646368245568 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_0(process=0,(0,0,0,0))
I0420 19:34:46.842324 134646368245568 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_1(process=0,(1,0,0,0))
I0420 19:34:46.842350 134646368245568 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_4(process=0,(0,1,0,0))
I0420 19:34:46.842374 134646368245568 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_5(process=0,(1,1,0,0))
I0420 19:34:47.154838 134646368245568 metric_logger.py:185] completed step: 0, seconds: 11.287, TFLOP/s/device: 1.204, Tokens/s/device: 181.447, total_weights: 65536, loss: 10.874
I0420 19:34:47.157592 134646368245568 metric_logger.py:269] To see full metrics 'tensorboard --logdir=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/tensorboard/'
I0420 19:34:47.564099 134646368245568 metric_logger.py:185] completed step: 1, seconds: 1.287, TFLOP/s/device: 10.555, Tokens/s/device: 1591.023, total_weights: 65536, loss: 10.874
I0420 19:34:47.710157 134646368245568 metric_logger.py:185] completed step: 2, seconds: 0.410, TFLOP/s/device: 33.138, Tokens/s/device: 4994.915, total_weights: 65536, loss: 10.263
I0420 19:34:47.857554 134646368245568 metric_logger.py:185] completed step: 3, seconds: 0.008, TFLOP/s/device: 1607.374, Tokens/s/device: 242280.847, total_weights: 65536, loss: 9.734
I0420 19:35:06.632045 134646368245568 metric_logger.py:185] completed step: 4, seconds: 0.147, TFLOP/s/device: 92.294, Tokens/s/device: 13911.626, total_weights: 65536, loss: 9.278
I0420 19:35:06.640708 134646368245568 metric_logger.py:185] completed step: 5, seconds: 0.148, TFLOP/s/device: 92.104, Tokens/s/device: 13882.863, total_weights: 65536, loss: 8.892
I0420 19:35:06.786926 134646368245568 metric_logger.py:185] completed step: 6, seconds: 18.773, TFLOP/s/device: 0.724, Tokens/s/device: 109.090, total_weights: 65536, loss: 8.592
I0420 19:35:06.934209 134646368245568 metric_logger.py:185] completed step: 7, seconds: 0.007, TFLOP/s/device: 1997.520, Tokens/s/device: 301087.915, total_weights: 65536, loss: 8.384
I0420 19:35:07.081697 134646368245568 metric_logger.py:185] completed step: 8, seconds: 0.147, TFLOP/s/device: 92.446, Tokens/s/device: 13934.437, total_weights: 65536, loss: 8.255
I0420 19:35:07.135908 134520976168704 atomicity.py:137] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0/items
I0420 19:35:07.228772 134646368245568 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0420 19:35:07.229483 134646368245568 checkpointing.py:798] Waited 0.000732421875 seconds for step 9 to finish before starting checkpointing.
I0420 19:35:07.231423 134646368245568 checkpoint_manager.py:2024] [process=0][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 19:35:07.295654 134519894034176 checkpoint.py:188] Wrote Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776713706887076355, 'commit_timestamp_nsecs': None, 'custom_metadata': {}}, json={"item_handlers": null, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1776713706887076355, "commit_timestamp_nsecs": null, "custom_metadata": {}} to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0/_CHECKPOINT_METADATA
I0420 19:35:08.278739 2638 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0420 19:35:09.950720 134519919212288 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0/items/array_metadatas/process_0
I0420 19:35:44.745417 134519910819584 base_pytree_checkpoint_handler.py:1232] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 36.473758s (commit=35.502856s, array_metadata_write=0.970901s)
I0420 19:35:44.746530 134519902426880 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/gbytes_per_sec: 27.030 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 58.43853402137756 s) (per-host)
I0420 19:35:44.746656 134519902426880 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 57.934916s.
I0420 19:35:45.471830 134519902426880 checkpoint.py:228] Read Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776713706887076355, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} from gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0/_CHECKPOINT_METADATA
I0420 19:35:46.219955 134519894034176 checkpoint.py:247] Updated Metadata={'item_handlers': {'items': 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler'}, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776713706887076355, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0/_CHECKPOINT_METADATA
I0420 19:35:46.574269 134519902426880 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000247s.
I0420 19:35:46.988803 134519902426880 ocdbt_utils.py:49] Param validation support for Zarr3 will be added later (b/362328389).
I0420 19:35:47.928656 134519902426880 base_pytree_checkpoint_handler.py:1356] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 2.327386s. use_zarr3=True, enable_post_merge_validation=True, directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0/items
I0420 19:35:47.930660 134519902426880 atomicity.py:608] Finalizing gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0/items
I0420 19:35:48.345053 134519902426880 atomicity.py:608] Finalizing gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0
I0420 19:35:49.744885 134519902426880 atomicity.py:790] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0`.
I0420 19:35:49.745610 134519902426880 async_checkpointer.py:420] Finished async_save (blocking + background). Time taken: 63.721776s. directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/0
I0420 19:35:49.746908 134519902426880 async_checkpointer.py:144] [process=0][thread=async_save] Background save thread done. Time taken: 62.935171s.
I0420 19:35:49.747157 134519373948672 async_checkpointer.py:273] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 19:35:49.747319 134519373948672 async_checkpointer.py:283] [process=0][thread=save_finalize] No errors found in background save thread=async_save.
I0420 19:35:49.747386 134519373948672 checkpoint_manager.py:2133] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0420 19:35:49.749011 134519373948672 checkpoint_manager.py:2142] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0420 19:35:49.749167 134646368245568 checkpoint_manager.py:2036] [process=0][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0420 19:35:49.749302 134646368245568 checkpoint_manager.py:1458] Waiting for previous save to complete took 42.517884 seconds. If this number is high, consider checkpointing less frequently.
I0420 19:35:49.751289 134646368245568 checkpoint_manager.py:1518] [process=0] Saving checkpoint at step 9
I0420 19:35:49.754800 134646368245568 async_checkpointer.py:452] [process=0] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/9.
I0420 19:35:50.703099 134519902426880 atomicity.py:137] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/9
I0420 19:35:50.706438 134646368245568 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array.
I0420 19:35:50.706501 134646368245568 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 19:35:50.732730 134646368245568 base_pytree_checkpoint_handler.py:153] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.027273s
I0420 19:35:50.733487 134646368245568 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.510 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.279970645904541 s) (per-host)
I0420 19:35:50.733558 134646368245568 base_pytree_checkpoint_handler.py:737] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.280058s (batch_requests_ready=0.249833s, total_serialization_initiated=0.029553s, others=0.000672s)
I0420 19:35:50.733637 134646368245568 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.283952s (all_items=0.000016s, per_item={'items': '0.00001574'}, temp_paths=0.283937)
I0420 19:35:50.734581 134519862855424 async_checkpointer.py:79] [process=0][thread=async_save] Background save thread started.
I0420 19:35:50.734746 134646368245568 async_checkpointer.py:561] Finished blocking save. Time taken: 0.983383s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/9.
I0420 19:35:50.770789 134646368245568 checkpoint_manager.py:1566] [process=0][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 19:35:50.771106 134519373948672 async_checkpointer.py:265] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0420 19:35:50.771274 134646368245568 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776713707.231392, 'wait_for_prev_duration_secs': 42.517884492874146, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776713749.7513301, 'checkpointer_blocking_duration_secs': 0.9835238456726074, 'get_old_steps_start_time': 1776713750.7348726, 'get_old_steps_duration_secs': 2.7179718017578125e-05, 'checkpoint_manager_blocking_start_time': 1776713707.2297132, 'checkpoint_manager_blocking_duration_secs': 43.54152798652649}
I0420 19:35:50.771378 134646368245568 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0420 19:35:50.771421 134646368245568 checkpoint_manager.py:2024] [process=0][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 19:35:52.919154 134518828685056 atomicity.py:137] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/9/items
I0420 19:35:55.728845 134519919212288 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/9/items/array_metadatas/process_0
I0420 19:36:34.964225 134519910819584 base_pytree_checkpoint_handler.py:1232] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 40.901630s (commit=39.799254s, array_metadata_write=1.102376s)
I0420 19:36:34.965363 134519862855424 base_pytree_checkpoint_handler.py:129] [process=0] /jax/checkpoint/write/gbytes_per_sec: 35.487 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 44.51181769371033 s) (per-host)
I0420 19:36:34.965471 134519862855424 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 44.230769s.
I0420 19:36:36.631956 134519862855424 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000254s.
I0420 19:36:36.639531 134519862855424 ocdbt_utils.py:49] Param validation support for Zarr3 will be added later (b/362328389).
I0420 19:36:37.727965 134519862855424 base_pytree_checkpoint_handler.py:1356] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 1.943529s. use_zarr3=True, enable_post_merge_validation=True, directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/9/items
I0420 19:36:37.729658 134519862855424 atomicity.py:608] Finalizing gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/9/items
I0420 19:36:38.144950 134519862855424 atomicity.py:608] Finalizing gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/9
I0420 19:36:39.441531 134519862855424 atomicity.py:790] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/9`.
I0420 19:36:39.442429 134519862855424 async_checkpointer.py:420] Finished async_save (blocking + background). Time taken: 49.691072s. directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260420_190404/linen_xpk_feat_nnx_set_defaults_true_20260420_190404_07_eval/checkpoints/9
I0420 19:36:39.443972 134519862855424 async_checkpointer.py:144] [process=0][thread=async_save] Background save thread done. Time taken: 48.709268s.
I0420 19:36:39.444145 134519373948672 async_checkpointer.py:273] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 19:36:39.444221 134519373948672 async_checkpointer.py:283] [process=0][thread=save_finalize] No errors found in background save thread=async_save.
I0420 19:36:39.444283 134519373948672 checkpoint_manager.py:2133] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0420 19:36:39.446051 134519373948672 checkpoint_manager.py:2142] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0420 19:36:39.446215 134646368245568 checkpoint_manager.py:2036] [process=0][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0420 19:36:39.446355 134646368245568 checkpoint_manager.py:2013] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 19:36:39.447072 134646368245568 metric_logger.py:185] completed step: 9, seconds: 0.148, TFLOP/s/device: 92.064, Tokens/s/device: 13876.843, total_weights: 65536, loss: 8.179
Per train step:
Total TFLOPs: 13.59
split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Mon Apr 20 19:36:49 UTC 2026
EXIT_CODE=0