XPK Start: Mon Apr 20 06:14:39 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-20 06:15:03.720267: 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 06:15:03.899898 139347201169216 max_utils.py:273] Attempting to initialize the jax distributed system...
I0420 06:15:12.940868 139347201169216 distributed.py:149] Starting JAX distributed service on [::]:8482
I0420 06:15:12.943203 139347201169216 distributed.py:172] Connecting to JAX distributed service on mt-06-grad-accum-rwuy8-slice-job-0-0.mt-06-grad-accum-rwuy8:8482
I0420 06:15:14.697519 139347201169216 max_utils.py:284] Jax distributed system initialized!
I0420 06:15:20.783752 139347201169216 max_utils.py:800] System Information: Jax Version: 0.9.2
I0420 06:15:20.783854 139347201169216 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0420 06:15:20.783894 139347201169216 max_utils.py:802] System Information: Jax Backend: PJRT C API
TFRT TPU v6 lite
Built on Mar 4 2026 11:32:08 (1772652728) cl/878335365
I0420 06:15:20.783932 139347201169216 train_utils.py:348] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0420 06:15:21.481423 139347201169216 maxtext_utils.py:1551] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 06:15:21.481701 139347201169216 checkpointing.py:677] Setting up checkpoint logger...
I0420 06:15:21.481765 139347201169216 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0420 06:15:21.481808 139347201169216 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0420 06:15:21.482143 139347201169216 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 0x7ebb94a530e0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0420 06:15:24.417333 139347201169216 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0420 06:15:24.417563 139347201169216 checkpoint_manager.py:708] [process=1][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ea7882bee70>}, handler_registry=None
I0420 06:15:24.417811 139347201169216 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ea7882bee70>` 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 06:15:24.417860 139347201169216 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ea7882c2bd0>` 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 06:15:24.417898 139347201169216 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 0x7ea7882bee70>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ea7882bee70>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ea7882c2bd0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ea7882c2bd0>}).
I0420 06:15:24.418221 139347201169216 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0420 06:15:24.418292 139347201169216 async_checkpointer.py:192] [process=1][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7ea7807cbd80> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0420 06:15:25.253172 139347201169216 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_06_grad_accum/checkpoints
I0420 06:15:25.280597 139347201169216 checkpoint_manager.py:929] [process=1][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_20260420_060028/linen_xpk_main_20260420_060028_06_grad_accum/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7ea7882c1670>
I0420 06:15:25.280755 139347201169216 checkpointing.py:301] Checkpoint manager created!
I0420 06:15:27.031110 139347201169216 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0420 06:15:27.031222 139347201169216 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 06:15:27.422286 139347201169216 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 06:15:27.422383 139347201169216 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 06:15:27.438799 139347201169216 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 06:15:27.438859 139347201169216 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 06:15:27.462530 139347201169216 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 06:15:27.462596 139347201169216 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 06:15:27.478978 139347201169216 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 06:15:27.479037 139347201169216 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 06:15:27.495342 139347201169216 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 06:15:27.495401 139347201169216 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 06:15:27.519995 139347201169216 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0420 06:15:27.520062 139347201169216 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 06:15:27.541897 139347201169216 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0420 06:15:27.541970 139347201169216 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0420 06:15:27.759797 139347201169216 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0420 06:15:27.759904 139347201169216 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0420 06:15:29.181844 139347201169216 maxtext_utils.py:1654] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0420 06:15:29.181967 139347201169216 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)
I0420 06:15:29.182025 139347201169216 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)
I0420 06:15:29.182083 139347201169216 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')
I0420 06:15:29.182136 139347201169216 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)
I0420 06:15:29.182174 139347201169216 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)
I0420 06:15:29.182225 139347201169216 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)
I0420 06:15:29.182275 139347201169216 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')
I0420 06:15:29.182314 139347201169216 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)
I0420 06:15:29.182350 139347201169216 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)
I0420 06:15:29.182397 139347201169216 maxtext_utils.py:1654] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0420 06:15:29.182446 139347201169216 maxtext_utils.py:1654] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0420 06:15:29.207189 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Logical: float32[2048]............................................... Unknown.
I0420 06:15:29.207260 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Physical: float32[2048]............................................... (None,).
I0420 06:15:29.221975 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Logical: float32[2048,16,7168]....................................... Unknown.
I0420 06:15:29.222034 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Physical: float32[2048,16,7168]....................................... ('fsdp', None, None).
I0420 06:15:29.251162 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Logical: float32[7168,16,2048]....................................... Unknown.
I0420 06:15:29.251228 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Physical: float32[7168,16,2048]....................................... (None, None, 'fsdp').
I0420 06:15:29.265873 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Logical: float32[2048,16]............................................ Unknown.
I0420 06:15:29.265928 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Physical: float32[2048,16]............................................ (None, None).
I0420 06:15:29.295085 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Logical: float32[2048,16,16,128]..................................... Unknown.
I0420 06:15:29.295153 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Physical: float32[2048,16,16,128]..................................... ('fsdp', None, None, None).
I0420 06:15:29.309730 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Logical: float32[16,16,128,2048]..................................... Unknown.
I0420 06:15:29.309782 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Physical: float32[16,16,128,2048]..................................... (None, None, None, 'fsdp').
I0420 06:15:29.353372 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Logical: float32[2048,32000]......................................... Unknown.
I0420 06:15:29.353445 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Physical: float32[2048,32000]......................................... ('fsdp', None).
I0420 06:15:29.368094 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Logical: float32[32000,2048]......................................... Unknown.
I0420 06:15:29.368152 139347201169216 gradient_accumulation.py:68] gradient_accumulation/inputs Physical: float32[32000,2048]......................................... (None, 'fsdp').
I0420 06:15:30.031399 139347201169216 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 06:15:30.031491 139347201169216 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 06:15:30.047108 139347201169216 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 06:15:30.047171 139347201169216 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 06:15:41.408888 139347201169216 max_utils.py:791] Total memory size: 1.7 GB, Output size: 0.4 GB, Temp size: 1.3 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0420 06:15:41.409768 139347201169216 metric_logger.py:301] number parameters: 1.104 billion
I0420 06:15:43.237338 139347201169216 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0420 06:15:53.915810 139347201169216 checkpointing.py:776] Waited 10.678449630737305 seconds for step 0 to finish before starting checkpointing.
I0420 06:15:53.918154 139347201169216 checkpoint_manager.py:2009] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 06:15:53.919916 139347201169216 checkpoint_manager.py:1512] [process=1] Saving checkpoint at step 0
I0420 06:15:53.921402 139347201169216 event_tracking.py:70] [process=1] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_06_grad_accum/checkpoints/0.
I0420 06:15:54.636198 139347201169216 signaling_client.py:364] Using JaxDistributedSignalingClient
I0420 06:15:54.638503 139347201169216 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0420 06:15:54.638561 139347201169216 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 06:15:54.908959 139347201169216 base_pytree_checkpoint_handler.py:154] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.272833s
I0420 06:15:54.909134 139347201169216 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/blocking_gbytes_per_sec: 5.544 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.27823734283447266 s) (per-host)
I0420 06:15:54.909189 139347201169216 base_pytree_checkpoint_handler.py:768] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.278302s (batch_requests_ready=0.002287s, total_serialization_initiated=0.275943s, others=0.000072s)
I0420 06:15:54.909281 139347201169216 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.282465s (all_items=0.000019s, per_item={'items': '0.00001884'}, temp_paths=0.282446)
I0420 06:15:54.910089 139347201169216 event_tracking.py:125] [process=1] [async] Finished blocking save in 0.99 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_06_grad_accum/checkpoints/0.
I0420 06:15:54.910430 139219250951936 async_checkpointer.py:76] [process=1][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 06:35:54.910393
I0420 06:15:54.924022 139347201169216 checkpoint_manager.py:1560] [process=1][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 06:15:54.924269 139218739259136 async_checkpointer.py:280] [process=1][thread=save_finalize] Waiting for background save thread=async_save.
I0420 06:15:54.924400 139347201169216 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_06_grad_accum/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776665753.918135, 'wait_for_prev_duration_secs': 6.103515625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776665753.9199605, 'checkpointer_blocking_duration_secs': 0.9906277656555176, 'get_old_steps_start_time': 1776665754.9106126, 'get_old_steps_duration_secs': 2.8848648071289062e-05, 'checkpoint_manager_blocking_start_time': 1776665753.9163086, 'checkpoint_manager_blocking_duration_secs': 1.0080537796020508}
I0420 06:15:54.924505 139347201169216 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0420 06:15:54.924554 139347201169216 max_utils.py:750]
Memstats: After params initialized:
I0420 06:15:54.924601 139347201169216 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_2(process=1,(2,0,0,0))
I0420 06:15:54.924632 139347201169216 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_3(process=1,(3,0,0,0))
I0420 06:15:54.924659 139347201169216 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_6(process=1,(2,1,0,0))
I0420 06:15:54.924692 139347201169216 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_7(process=1,(3,1,0,0))
I0420 06:15:55.237213 139347201169216 metric_logger.py:196] completed step: 0, seconds: 1.827, TFLOP/s/device: 29.740, Tokens/s/device: 4482.707, total_weights: 262144, loss: 10.877, lm_loss: 10.877, perplexity: 52959.059
I0420 06:15:55.849415 139347201169216 metric_logger.py:196] completed step: 1, seconds: 11.998, TFLOP/s/device: 4.530, Tokens/s/device: 682.767, total_weights: 262144, loss: 10.877, lm_loss: 10.877, perplexity: 52959.059
I0420 06:15:56.427485 139347201169216 metric_logger.py:196] completed step: 2, seconds: 0.034, TFLOP/s/device: 1602.681, Tokens/s/device: 241573.531, total_weights: 262144, loss: 10.563, lm_loss: 10.563, perplexity: 38662.707
I0420 06:15:57.005661 139347201169216 metric_logger.py:196] completed step: 3, seconds: 0.585, TFLOP/s/device: 92.946, Tokens/s/device: 14009.789, total_weights: 262144, loss: 10.272, lm_loss: 10.272, perplexity: 28909.668
I0420 06:15:58.217051 139347201169216 metric_logger.py:196] completed step: 4, seconds: 0.578, TFLOP/s/device: 94.034, Tokens/s/device: 14173.795, total_weights: 262144, loss: 10.022, lm_loss: 10.022, perplexity: 22524.992
I0420 06:15:58.217978 2576 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0420 06:15:58.228410 139347201169216 metric_logger.py:196] completed step: 5, seconds: 0.578, TFLOP/s/device: 93.996, Tokens/s/device: 14168.157, total_weights: 262144, loss: 9.820, lm_loss: 9.820, perplexity: 18401.865
I0420 06:16:00.010019 139218747651840 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_06_grad_accum/checkpoints/0/items/array_metadatas/process_1
I0420 06:16:33.508983 139219250951936 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/gbytes_per_sec: 40.630 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 38.87806725502014 s) (per-host)
I0420 06:16:33.509088 139219250951936 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 38.598547s.
I0420 06:16:44.549484 139347201169216 metric_logger.py:196] completed step: 6, seconds: 1.217, TFLOP/s/device: 44.661, Tokens/s/device: 6731.843, total_weights: 262144, loss: 9.667, lm_loss: 9.667, perplexity: 15787.604
I0420 06:16:45.127628 139347201169216 metric_logger.py:196] completed step: 7, seconds: 45.744, TFLOP/s/device: 1.188, Tokens/s/device: 179.084, total_weights: 262144, loss: 9.561, lm_loss: 9.561, perplexity: 14203.827
I0420 06:16:45.706096 139347201169216 metric_logger.py:196] completed step: 8, seconds: 0.583, TFLOP/s/device: 93.211, Tokens/s/device: 14049.771, total_weights: 262144, loss: 9.496, lm_loss: 9.496, perplexity: 13302.920
I0420 06:16:46.283579 139347201169216 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0420 06:16:46.284271 139347201169216 checkpointing.py:776] Waited 0.0007114410400390625 seconds for step 9 to finish before starting checkpointing.
I0420 06:16:46.286491 139347201169216 checkpoint_manager.py:2020] [process=1][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 06:16:47.236913 139219250951936 async_checkpointer.py:160] [process=1][thread=async_save] Background save thread done. Time taken: 52.326354s.
I0420 06:16:47.237189 139218739259136 async_checkpointer.py:288] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 06:16:47.237303 139218739259136 async_checkpointer.py:298] [process=1][thread=save_finalize] No errors found in background save thread=async_save.
I0420 06:16:47.237353 139218739259136 checkpoint_manager.py:2137] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0420 06:16:47.239930 139218739259136 checkpoint_manager.py:2146] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0420 06:16:47.240131 139347201169216 checkpoint_manager.py:2032] [process=1][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
I0420 06:16:47.241531 139347201169216 checkpoint_manager.py:1512] [process=1] Saving checkpoint at step 9
I0420 06:16:47.243475 139347201169216 event_tracking.py:70] [process=1] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_06_grad_accum/checkpoints/9.
I0420 06:16:47.530928 139347201169216 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0420 06:16:47.531023 139347201169216 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 06:16:47.566626 139347201169216 base_pytree_checkpoint_handler.py:154] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.036603s
I0420 06:16:47.566806 139347201169216 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/blocking_gbytes_per_sec: 38.217 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04036402702331543 s) (per-host)
I0420 06:16:47.566857 139347201169216 base_pytree_checkpoint_handler.py:768] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.040425s (batch_requests_ready=0.002005s, total_serialization_initiated=0.038351s, others=0.000069s)
I0420 06:16:47.566945 139347201169216 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.044637s (all_items=0.000022s, per_item={'items': '0.00002241'}, temp_paths=0.044615)
I0420 06:16:47.567608 139347201169216 event_tracking.py:125] [process=1] [async] Finished blocking save in 0.33 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_06_grad_accum/checkpoints/9.
I0420 06:16:47.567955 139218739259136 async_checkpointer.py:76] [process=1][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 06:36:47.567919
I0420 06:16:47.569945 139347201169216 checkpoint_manager.py:1560] [process=1][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 06:16:47.570229 139218694416128 async_checkpointer.py:280] [process=1][thread=save_finalize] Waiting for background save thread=async_save.
I0420 06:16:47.570365 139347201169216 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_06_grad_accum/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776665806.286461, 'wait_for_prev_duration_secs': 0.9537875652313232, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776665807.2415707, 'checkpointer_blocking_duration_secs': 0.32653355598449707, 'get_old_steps_start_time': 1776665807.568126, 'get_old_steps_duration_secs': 2.9087066650390625e-05, 'checkpoint_manager_blocking_start_time': 1776665806.284502, 'checkpoint_manager_blocking_duration_secs': 1.2858288288116455}
I0420 06:16:47.570528 139347201169216 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0420 06:16:47.570572 139347201169216 checkpoint_manager.py:2020] [process=1][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 06:16:53.531999 139223034226432 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_06_grad_accum/checkpoints/9/items/array_metadatas/process_1
I0420 06:17:31.066528 139218739259136 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/gbytes_per_sec: 36.279 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.54006004333496 s) (per-host)
I0420 06:17:31.066630 139218739259136 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.498563s.
I0420 06:17:38.929529 139218739259136 async_checkpointer.py:160] [process=1][thread=async_save] Background save thread done. Time taken: 51.361444s.
I0420 06:17:38.929826 139218694416128 async_checkpointer.py:288] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 06:17:38.929945 139218694416128 async_checkpointer.py:298] [process=1][thread=save_finalize] No errors found in background save thread=async_save.
I0420 06:17:38.929993 139218694416128 checkpoint_manager.py:2137] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0420 06:17:38.931767 139218694416128 checkpoint_manager.py:2146] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0420 06:17:38.931961 139347201169216 checkpoint_manager.py:2032] [process=1][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0420 06:17:38.932108 139347201169216 checkpoint_manager.py:2009] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 06:17:38.933213 139347201169216 metric_logger.py:196] completed step: 9, seconds: 0.578, TFLOP/s/device: 93.975, Tokens/s/device: 14164.899, total_weights: 262144, loss: 9.457, lm_loss: 9.457, perplexity: 12802.546
Per train step:
Total TFLOPs: 54.35
split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Mon Apr 20 06:17:49 UTC 2026
EXIT_CODE=0