XPK Start: Sat Apr 18 13:17:57 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-18 13:18:21.021439: 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)
I0418 13:18:21.558916 139629234177856 max_utils.py:273] Attempting to initialize the jax distributed system...
I0418 13:18:30.602874 139629234177856 distributed.py:149] Starting JAX distributed service on [::]:8482
I0418 13:18:30.605217 139629234177856 distributed.py:172] Connecting to JAX distributed service on mt-07-eval-py68q-slice-job-0-0.mt-07-eval-py68q:8482
I0418 13:18:32.426449 139629234177856 max_utils.py:284] Jax distributed system initialized!
I0418 13:18:38.454791 139629234177856 max_utils.py:800] System Information: Jax Version: 0.9.2
I0418 13:18:38.454895 139629234177856 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0418 13:18:38.454934 139629234177856 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
I0418 13:18:38.454967 139629234177856 train_utils.py:348] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0418 13:18:39.145864 139629234177856 maxtext_utils.py:1551] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0418 13:18:39.146146 139629234177856 checkpointing.py:677] Setting up checkpoint logger...
I0418 13:18:39.146202 139629234177856 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0418 13:18:39.146246 139629234177856 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0418 13:18:39.146584 139629234177856 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 0x7efd5b114200>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0418 13:18:42.186839 139629234177856 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0418 13:18:42.187077 139629234177856 checkpoint_manager.py:708] [process=3][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ee9d8631370>}, handler_registry=None
I0418 13:18:42.187324 139629234177856 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ee9d8631370>` 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`.
I0418 13:18:42.187371 139629234177856 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ee9d830eb10>` 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`.
I0418 13:18:42.187405 139629234177856 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 0x7ee9d8631370>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ee9d8631370>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ee9d830eb10>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ee9d830eb10>}).
I0418 13:18:42.187725 139629234177856 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0418 13:18:42.187795 139629234177856 async_checkpointer.py:192] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7ee9b87abba0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0418 13:18:43.763381 139629234177856 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_07_eval/checkpoints
I0418 13:18:44.203818 139629234177856 checkpoint_manager.py:929] [process=3][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7ee9d829a000>
I0418 13:18:44.203997 139629234177856 checkpointing.py:301] Checkpoint manager created!
I0418 13:18:45.134675 139629234177856 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0418 13:18:45.134788 139629234177856 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 13:18:45.514159 139629234177856 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 13:18:45.514259 139629234177856 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 13:18:45.530498 139629234177856 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 13:18:45.530559 139629234177856 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 13:18:45.554140 139629234177856 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 13:18:45.554205 139629234177856 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 13:18:45.570597 139629234177856 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 13:18:45.570664 139629234177856 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 13:18:45.586909 139629234177856 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 13:18:45.586966 139629234177856 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 13:18:45.612429 139629234177856 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0418 13:18:45.612498 139629234177856 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 13:18:45.633302 139629234177856 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0418 13:18:45.633363 139629234177856 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0418 13:18:45.850428 139629234177856 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0418 13:18:45.850536 139629234177856 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0418 13:18:47.275524 139629234177856 maxtext_utils.py:1654] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0418 13:18:47.275661 139629234177856 maxtext_utils.py:1654] params/params/decoder/layers/mlp/wi_0/kernel
Shape: float32[2048,16,7168]
Logical: P('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0418 13:18:47.275714 139629234177856 maxtext_utils.py:1654] params/params/decoder/layers/mlp/wi_1/kernel
Shape: float32[2048,16,7168]
Logical: P('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0418 13:18:47.275771 139629234177856 maxtext_utils.py:1654] params/params/decoder/layers/mlp/wo/kernel
Shape: float32[7168,16,2048]
Logical: P('mlp', 'layers', 'embed')
Physical: (None, None, 'fsdp')
I0418 13:18:47.275847 139629234177856 maxtext_utils.py:1654] params/params/decoder/layers/post_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: P('norm', 'layers')
Physical: (None, None)
I0418 13:18:47.275888 139629234177856 maxtext_utils.py:1654] params/params/decoder/layers/pre_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: P('norm', 'layers')
Physical: (None, None)
I0418 13:18:47.275942 139629234177856 maxtext_utils.py:1654] 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)
I0418 13:18:47.276026 139629234177856 maxtext_utils.py:1654] 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')
I0418 13:18:47.276080 139629234177856 maxtext_utils.py:1654] 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)
I0418 13:18:47.276121 139629234177856 maxtext_utils.py:1654] 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)
I0418 13:18:47.276171 139629234177856 maxtext_utils.py:1654] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0418 13:18:47.276222 139629234177856 maxtext_utils.py:1654] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0418 13:18:47.772101 139629234177856 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0418 13:18:47.772211 139629234177856 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0418 13:18:47.789487 139629234177856 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0418 13:18:47.789550 139629234177856 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0418 13:18:58.709382 139629234177856 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.
I0418 13:18:58.710321 139629234177856 metric_logger.py:301] number parameters: 1.104 billion
I0418 13:19:09.860888 139629234177856 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0418 13:19:10.068333 139629234177856 checkpointing.py:776] Waited 0.20742368698120117 seconds for step 0 to finish before starting checkpointing.
I0418 13:19:10.070995 139629234177856 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 13:19:10.072585 139629234177856 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 0
I0418 13:19:10.074007 139629234177856 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_07_eval/checkpoints/0.
I0418 13:19:10.805634 139629234177856 signaling_client.py:364] Using JaxDistributedSignalingClient
I0418 13:19:10.806717 139629234177856 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 13:19:10.806772 139629234177856 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
I0418 13:19:11.085134 139629234177856 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.279590s
I0418 13:19:11.085310 139629234177856 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 5.413 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2849845886230469 s) (per-host)
I0418 13:19:11.085363 139629234177856 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.285048s (batch_requests_ready=0.002187s, total_serialization_initiated=0.282789s, others=0.000072s)
I0418 13:19:11.085452 139629234177856 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.289275s (all_items=0.000027s, per_item={'items': '0.00002694'}, temp_paths=0.289248)
I0418 13:19:11.086285 139629234177856 event_tracking.py:125] [process=3] [async] Finished blocking save in 1.01 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_07_eval/checkpoints/0.
I0418 13:19:11.086544 139505114793728 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-18 13:39:11.086524
I0418 13:19:11.094004 139629234177856 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 13:19:11.094257 139503557670656 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0418 13:19:11.094409 139629234177856 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776518350.0709772, 'wait_for_prev_duration_secs': 5.936622619628906e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776518350.072636, 'checkpointer_blocking_duration_secs': 1.0140116214752197, 'get_old_steps_start_time': 1776518351.086669, 'get_old_steps_duration_secs': 2.956390380859375e-05, 'checkpoint_manager_blocking_start_time': 1776518350.0689034, 'checkpoint_manager_blocking_duration_secs': 1.0254690647125244}
I0418 13:19:11.094516 139629234177856 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0418 13:19:11.094565 139629234177856 max_utils.py:750]
Memstats: After params initialized:
I0418 13:19:11.094624 139629234177856 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_10(process=3,(2,2,0,0))
I0418 13:19:11.094661 139629234177856 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_11(process=3,(3,2,0,0))
I0418 13:19:11.094690 139629234177856 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_14(process=3,(2,3,0,0))
I0418 13:19:11.094714 139629234177856 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_15(process=3,(3,3,0,0))
I0418 13:19:11.410652 139629234177856 metric_logger.py:196] completed step: 0, seconds: 11.150, TFLOP/s/device: 1.219, Tokens/s/device: 183.670, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0418 13:19:11.589683 139629234177856 metric_logger.py:196] completed step: 1, seconds: 1.548, TFLOP/s/device: 8.776, Tokens/s/device: 1322.831, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0418 13:19:12.012861 139629234177856 metric_logger.py:196] completed step: 2, seconds: 0.032, TFLOP/s/device: 426.959, Tokens/s/device: 64355.969, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0418 13:19:12.160236 139629234177856 metric_logger.py:196] completed step: 3, seconds: 0.423, TFLOP/s/device: 32.084, Tokens/s/device: 4835.994, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0418 13:19:12.455389 139629234177856 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.735, Tokens/s/device: 13375.043, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0418 13:19:12.461599 139629234177856 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.394, Tokens/s/device: 13926.667, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0418 13:19:14.420482 2714 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0418 13:19:18.292925 139504576870144 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_07_eval/checkpoints/0/items/array_metadatas/process_3
I0418 13:19:35.064023 139629234177856 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.906, Tokens/s/device: 6919.503, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0418 13:19:35.211449 139629234177856 metric_logger.py:196] completed step: 7, seconds: 22.456, TFLOP/s/device: 0.605, Tokens/s/device: 91.202, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0418 13:19:35.358860 139629234177856 metric_logger.py:196] completed step: 8, seconds: 0.153, TFLOP/s/device: 89.038, Tokens/s/device: 13420.708, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0418 13:19:35.505468 139629234177856 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0418 13:19:35.506129 139629234177856 checkpointing.py:776] Waited 0.0006797313690185547 seconds for step 9 to finish before starting checkpointing.
I0418 13:19:35.508246 139629234177856 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 13:19:47.062243 139505114793728 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 43.561 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.26186680793762 s) (per-host)
I0418 13:19:47.062366 139505114793728 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 35.975763s.
I0418 13:19:58.140119 139505114793728 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 47.053499s.
I0418 13:19:58.140404 139503557670656 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 13:19:58.140532 139503557670656 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0418 13:19:58.140582 139503557670656 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0418 13:19:58.143660 139503557670656 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0418 13:19:58.143796 139629234177856 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0418 13:19:58.143888 139629234177856 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.635646 seconds. If this number is high, consider checkpointing less frequently.
I0418 13:19:58.145784 139629234177856 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 9
I0418 13:19:58.147975 139629234177856 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_07_eval/checkpoints/9.
I0418 13:19:59.280555 139629234177856 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 13:19:59.280660 139629234177856 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
I0418 13:19:59.316112 139629234177856 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.036817s
I0418 13:19:59.316284 139629234177856 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 38.408 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.040163278579711914 s) (per-host)
I0418 13:19:59.316337 139629234177856 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.040226s (batch_requests_ready=0.001722s, total_serialization_initiated=0.038434s, others=0.000070s)
I0418 13:19:59.316426 139629234177856 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.044367s (all_items=0.000016s, per_item={'items': '0.00001597'}, temp_paths=0.044351)
I0418 13:19:59.317113 139629234177856 event_tracking.py:125] [process=3] [async] Finished blocking save in 1.17 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_07_eval/checkpoints/9.
I0418 13:19:59.317456 139480967128832 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-18 13:39:59.317419
I0418 13:19:59.319402 139629234177856 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 13:19:59.319687 139503557670656 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0418 13:19:59.319845 139629234177856 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776518375.5082161, 'wait_for_prev_duration_secs': 22.6356463432312, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776518398.145822, 'checkpointer_blocking_duration_secs': 1.1717841625213623, 'get_old_steps_start_time': 1776518399.3176403, 'get_old_steps_duration_secs': 3.170967102050781e-05, 'checkpoint_manager_blocking_start_time': 1776518375.5063486, 'checkpoint_manager_blocking_duration_secs': 23.81346344947815}
I0418 13:19:59.319952 139629234177856 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0418 13:19:59.320001 139629234177856 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 13:20:05.197962 139504576870144 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260418_125719/linen_xpk_main_20260418_125719_07_eval/checkpoints/9/items/array_metadatas/process_3
I0418 13:20:41.637135 139480967128832 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 37.289 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.36097455024719 s) (per-host)
I0418 13:20:41.637270 139480967128832 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.319694s.
I0418 13:20:52.646466 139480967128832 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 53.328875s.
I0418 13:20:52.646748 139503557670656 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 13:20:52.646864 139503557670656 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0418 13:20:52.646912 139503557670656 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0418 13:20:52.649690 139503557670656 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0418 13:20:52.649848 139629234177856 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0418 13:20:52.649989 139629234177856 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 13:20:52.650935 139629234177856 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.379, Tokens/s/device: 13924.300, 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: Sat Apr 18 13:21:02 UTC 2026
EXIT_CODE=0