XPK Start: Sat Apr 25 15:10:42 UTC 2026
2026-04-25 15:10:46.176859: 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 00:00:1777129846.189739 10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1777129846.193474 10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
W0000 00:00:1777129846.205302 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1777129846.205319 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1777129846.205322 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1777129846.205324 10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
2026-04-25 15:11:05.371827: 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)
I0425 15:11:05.898372 133026795149120 max_utils.py:273] Attempting to initialize the jax distributed system...
INFO:2026-04-25 15:11:14,938:jax._src.distributed:140: Starting JAX distributed service on [::]:8482
I0425 15:11:14.938657 133026795149120 distributed.py:140] Starting JAX distributed service on [::]:8482
INFO:2026-04-25 15:11:14,940:jax._src.distributed:157: Connecting to JAX distributed service on mt-04-int8-so9xn-slice-job-0-0.mt-04-int8-so9xn:8482
I0425 15:11:14.940986 133026795149120 distributed.py:157] Connecting to JAX distributed service on mt-04-int8-so9xn-slice-job-0-0.mt-04-int8-so9xn:8482
I0425 15:11:15.995849 133026795149120 max_utils.py:284] Jax distributed system initialized!
I0425 15:11:23.410891 133026795149120 max_utils.py:800] System Information: Jax Version: 0.8.1
I0425 15:11:23.410997 133026795149120 max_utils.py:801] System Information: Jaxlib Version: 0.8.1
I0425 15:11:23.411037 133026795149120 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
I0425 15:11:23.411071 133026795149120 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0425 15:11:24.080007 133026795149120 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0425 15:11:24.080614 133026795149120 maxtext_utils.py:1631] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0425 15:11:24.080913 133026795149120 checkpointing.py:688] Setting up checkpoint logger...
I0425 15:11:24.080967 133026795149120 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0425 15:11:24.081009 133026795149120 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None
I0425 15:11:24.081374 133026795149120 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 0x78fc377790a0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0425 15:11:26.954517 133026795149120 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0425 15:11:26.954756 133026795149120 checkpoint_manager.py:709] [process=6][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78e76c7e5b80>}, handler_registry=None
I0425 15:11:26.954990 133026795149120 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78e76c7e5b80>` 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`.
I0425 15:11:26.955038 133026795149120 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78e76c7ea690>` 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`.
I0425 15:11:26.955074 133026795149120 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 0x78e76c7e5b80>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x78e76c7e5b80>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78e76c7ea690>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x78e76c7ea690>}).
I0425 15:11:26.955600 133026795149120 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33
I0425 15:11:26.955679 133026795149120 async_checkpointer.py:177] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x78e76c59c180> timeout: 600 secs and primary_host=0 for async checkpoint writes
I0425 15:11:28.503248 133026795149120 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_04_int8/checkpoints
I0425 15:11:28.505443 133026795149120 checkpoint_manager.py:929] [process=6][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_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_04_int8/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x78e7c414fd70>
I0425 15:11:28.505555 133026795149120 checkpointing.py:302] Checkpoint manager created!
I0425 15:11:29.693828 133026795149120 nnx_wrappers.py:455] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0425 15:11:29.693938 133026795149120 nnx_wrappers.py:455] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 15:11:30.067147 133026795149120 attentions.py:1084] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0425 15:11:30.067245 133026795149120 attentions.py:1084] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 15:11:30.082992 133026795149120 attentions.py:1085] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0425 15:11:30.083055 133026795149120 attentions.py:1085] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 15:11:30.164500 133026795149120 attentions.py:1150] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0425 15:11:30.164606 133026795149120 attentions.py:1150] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 15:11:30.180585 133026795149120 attentions.py:1151] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0425 15:11:30.180647 133026795149120 attentions.py:1151] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 15:11:30.196474 133026795149120 attentions.py:1152] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0425 15:11:30.196543 133026795149120 attentions.py:1152] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 15:11:30.220705 133026795149120 attentions.py:1193] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0425 15:11:30.220775 133026795149120 attentions.py:1193] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 15:11:30.301578 133026795149120 linears.py:541] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0425 15:11:30.301667 133026795149120 linears.py:541] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0425 15:11:30.900312 133026795149120 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0425 15:11:30.900425 133026795149120 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0425 15:11:33.586261 133026795149120 maxtext_utils.py:1740] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: PartitionSpec('norm',)
Physical: (None,)
I0425 15:11:33.586403 133026795149120 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)
I0425 15:11:33.586458 133026795149120 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)
I0425 15:11:33.586517 133026795149120 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')
I0425 15:11:33.586574 133026795149120 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)
I0425 15:11:33.586614 133026795149120 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)
I0425 15:11:33.586667 133026795149120 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)
I0425 15:11:33.586719 133026795149120 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')
I0425 15:11:33.586760 133026795149120 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)
I0425 15:11:33.586795 133026795149120 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)
I0425 15:11:33.586843 133026795149120 maxtext_utils.py:1740] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: PartitionSpec('embed', 'vocab')
Physical: ('fsdp', None)
I0425 15:11:33.586892 133026795149120 maxtext_utils.py:1740] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: PartitionSpec('vocab', 'embed')
Physical: (None, 'fsdp')
I0425 15:11:35.285433 133026795149120 train.py:158] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0425 15:11:35.285530 133026795149120 train.py:158] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0425 15:11:35.300353 133026795149120 train.py:165] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0425 15:11:35.300411 133026795149120 train.py:165] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0425 15:11:49.318693 133026795149120 max_utils.py:791] Total memory size: 1.8 GB, Output size: 0.4 GB, Temp size: 1.4 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0425 15:11:49.319538 133026795149120 metric_logger.py:289] number parameters: 1.104 billion
I0425 15:12:04.414469 133026795149120 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0425 15:12:04.654965 133026795149120 checkpointing.py:798] Waited 0.24047613143920898 seconds for step 0 to finish before starting checkpointing.
I0425 15:12:04.657473 133026795149120 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 15:12:04.659364 133026795149120 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 0
I0425 15:12:04.662212 133026795149120 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_04_int8/checkpoints/0.
I0425 15:12:05.246073 133026795149120 signaling_client.py:364] Using JaxDistributedSignalingClient
I0425 15:12:05.247203 133026795149120 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array.
I0425 15:12:05.247266 133026795149120 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
I0425 15:12:05.497224 133026795149120 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.251230s
I0425 15:12:05.497398 133026795149120 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.049 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.5058794021606445 s) (per-host)
I0425 15:12:05.497451 133026795149120 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.505943s (batch_requests_ready=0.250934s, total_serialization_initiated=0.254938s, others=0.000071s)
I0425 15:12:05.497547 133026795149120 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.510167s (all_items=0.000018s, per_item={'items': '0.00001836'}, temp_paths=0.510149)
I0425 15:12:05.498522 132894957508352 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started.
I0425 15:12:05.498687 133026795149120 async_checkpointer.py:561] Finished blocking save. Time taken: 0.839250s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_04_int8/checkpoints/0.
I0425 15:12:05.828752 133026795149120 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 15:12:05.829173 132893938546432 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 15:12:05.829349 133026795149120 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_04_int8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777129924.6574538, 'wait_for_prev_duration_secs': 6.437301635742188e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777129924.659405, 'checkpointer_blocking_duration_secs': 0.8393917083740234, 'get_old_steps_start_time': 1777129925.498817, 'get_old_steps_duration_secs': 2.4557113647460938e-05, 'checkpoint_manager_blocking_start_time': 1777129924.655622, 'checkpoint_manager_blocking_duration_secs': 1.1736869812011719}
I0425 15:12:05.829456 133026795149120 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0425 15:12:05.829506 133026795149120 max_utils.py:750]
Memstats: After params initialized:
I0425 15:12:05.829559 133026795149120 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0425 15:12:05.829594 133026795149120 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0425 15:12:05.829620 133026795149120 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0425 15:12:05.829643 133026795149120 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0425 15:12:06.102990 133026795149120 metric_logger.py:185] completed step: 0, seconds: 15.095, TFLOP/s/device: 0.900, Tokens/s/device: 135.676, total_weights: 65536, loss: 10.874
I0425 15:12:06.358536 133026795149120 metric_logger.py:185] completed step: 1, seconds: 1.687, TFLOP/s/device: 8.053, Tokens/s/device: 1213.830, total_weights: 65536, loss: 10.874
I0425 15:12:06.767338 133026795149120 metric_logger.py:185] completed step: 2, seconds: 0.027, TFLOP/s/device: 508.805, Tokens/s/device: 76692.630, total_weights: 65536, loss: 10.262
I0425 15:12:06.997161 133026795149120 metric_logger.py:185] completed step: 3, seconds: 0.409, TFLOP/s/device: 33.182, Tokens/s/device: 5001.624, total_weights: 65536, loss: 9.731
I0425 15:12:07.457653 133026795149120 metric_logger.py:185] completed step: 4, seconds: 0.234, TFLOP/s/device: 58.097, Tokens/s/device: 8757.077, total_weights: 65536, loss: 9.272
I0425 15:12:07.463137 133026795149120 metric_logger.py:185] completed step: 5, seconds: 0.230, TFLOP/s/device: 59.126, Tokens/s/device: 8912.059, total_weights: 65536, loss: 8.887
I0425 15:12:29.773379 2774 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0425 15:12:30.029262 133026795149120 metric_logger.py:185] completed step: 6, seconds: 0.461, TFLOP/s/device: 29.483, Tokens/s/device: 4443.962, total_weights: 65536, loss: 8.588
I0425 15:12:30.259987 133026795149120 metric_logger.py:185] completed step: 7, seconds: 22.334, TFLOP/s/device: 0.608, Tokens/s/device: 91.699, total_weights: 65536, loss: 8.380
I0425 15:12:30.490385 133026795149120 metric_logger.py:185] completed step: 8, seconds: 0.238, TFLOP/s/device: 57.037, Tokens/s/device: 8597.239, total_weights: 65536, loss: 8.251
I0425 15:12:30.720488 133026795149120 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0425 15:12:30.721177 133026795149120 checkpointing.py:798] Waited 0.0007073879241943359 seconds for step 9 to finish before starting checkpointing.
I0425 15:12:30.723152 133026795149120 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 15:12:31.679818 132894965901056 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_04_int8/checkpoints/0/items/array_metadatas/process_6
I0425 15:12:58.781397 132894957508352 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 29.366 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 53.78983473777771 s) (per-host)
I0425 15:12:58.781518 132894957508352 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 53.282883s.
I0425 15:13:07.731206 132894957508352 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 62.232557s.
I0425 15:13:07.731488 132893938546432 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 15:13:07.731632 132893938546432 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 15:13:07.731682 132893938546432 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0425 15:13:07.733212 132893938546432 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0425 15:13:07.733418 133026795149120 checkpoint_manager.py:2036] [process=6][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0425 15:13:07.733569 133026795149120 checkpoint_manager.py:1458] Waiting for previous save to complete took 37.010413 seconds. If this number is high, consider checkpointing less frequently.
I0425 15:13:07.735532 133026795149120 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 9
I0425 15:13:07.739009 133026795149120 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_04_int8/checkpoints/9.
I0425 15:13:08.394412 133026795149120 jax_array_handlers.py:358] Scheduling D2H of 39 prioritized jax.Array.
I0425 15:13:08.394516 133026795149120 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
I0425 15:13:08.435962 133026795149120 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.042922s
I0425 15:13:08.436146 133026795149120 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 3.743 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.41211962699890137 s) (per-host)
I0425 15:13:08.436201 133026795149120 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.412185s (batch_requests_ready=0.366981s, total_serialization_initiated=0.045131s, others=0.000073s)
I0425 15:13:08.436296 133026795149120 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.416601s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.416586)
I0425 15:13:08.437247 132897151379200 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started.
I0425 15:13:08.437408 133026795149120 async_checkpointer.py:561] Finished blocking save. Time taken: 0.701795s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_04_int8/checkpoints/9.
I0425 15:13:08.439346 133026795149120 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 15:13:08.439590 132893938546432 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 15:13:08.439728 133026795149120 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_04_int8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777129950.72312, 'wait_for_prev_duration_secs': 37.01041316986084, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777129987.735578, 'checkpointer_blocking_duration_secs': 0.7019379138946533, 'get_old_steps_start_time': 1777129988.437535, 'get_old_steps_duration_secs': 2.7418136596679688e-05, 'checkpoint_manager_blocking_start_time': 1777129950.721407, 'checkpoint_manager_blocking_duration_secs': 37.7182879447937}
I0425 15:13:08.439886 133026795149120 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0425 15:13:08.439932 133026795149120 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 15:13:13.092123 132897159771904 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_set_defaults_true_20260425_150412/linen_xpk_feat_nnx_set_defaults_true_20260425_150412_04_int8/checkpoints/9/items/array_metadatas/process_6
I0425 15:13:49.031697 132897151379200 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 38.520 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.00762867927551 s) (per-host)
I0425 15:13:49.031820 132897151379200 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 40.594460s.
I0425 15:13:56.290200 132897151379200 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 47.852826s.
I0425 15:13:56.290527 132893938546432 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 15:13:56.290644 132893938546432 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 15:13:56.290693 132893938546432 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0425 15:13:56.291914 132893938546432 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0425 15:13:56.292111 133026795149120 checkpoint_manager.py:2036] [process=6][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0425 15:13:56.292256 133026795149120 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 15:13:56.293032 133026795149120 metric_logger.py:185] completed step: 9, seconds: 0.230, TFLOP/s/device: 59.050, Tokens/s/device: 8900.633, total_weights: 65536, loss: 8.175
Per train step:
Total TFLOPs: 13.59
split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Sat Apr 25 15:14:07 UTC 2026
EXIT_CODE=0