XPK Start: Sat Apr 18 00:54:24 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 00:54:48.104797: 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 00:54:48.636398 135104514455360 max_utils.py:273] Attempting to initialize the jax distributed system...
I0418 00:54:57.678519 135104514455360 distributed.py:149] Starting JAX distributed service on [::]:8482
I0418 00:54:57.680811 135104514455360 distributed.py:172] Connecting to JAX distributed service on mt-11-optimizer-offload-lf0zn-slice-job-0-0.mt-11-optimizer-offload-lf0zn:8482
I0418 00:54:58.944988 135104514455360 max_utils.py:284] Jax distributed system initialized!
I0418 00:55:05.169156 135104514455360 max_utils.py:800] System Information: Jax Version: 0.9.2
I0418 00:55:05.169253 135104514455360 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0418 00:55:05.169293 135104514455360 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 00:55:05.169326 135104514455360 train_utils.py:348] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0418 00:55:05.853282 135104514455360 maxtext_utils.py:1551] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0418 00:55:05.853556 135104514455360 checkpointing.py:677] Setting up checkpoint logger...
I0418 00:55:05.853611 135104514455360 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0418 00:55:05.853654 135104514455360 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0418 00:55:05.854004 135104514455360 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 0x7adfc103fe00>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0418 00:55:09.119218 135104514455360 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0418 00:55:09.119451 135104514455360 checkpoint_manager.py:708] [process=0][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7acba838ee70>}, handler_registry=None
I0418 00:55:09.119688 135104514455360 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7acba838ee70>` 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 00:55:09.119735 135104514455360 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7acba8398c20>` 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 00:55:09.119770 135104514455360 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 0x7acba838ee70>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7acba838ee70>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7acba8398c20>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7acba8398c20>}).
I0418 00:55:09.120112 135104514455360 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0418 00:55:09.120181 135104514455360 async_checkpointer.py:192] [process=0][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7acb846f7880> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0418 00:55:10.751803 135104514455360 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints
I0418 00:55:12.043587 135104514455360 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_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7adeb0713bc0>
I0418 00:55:12.043753 135104514455360 checkpointing.py:301] Checkpoint manager created!
I0418 00:55:12.950358 135104514455360 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0418 00:55:12.950470 135104514455360 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 00:55:13.331708 135104514455360 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 00:55:13.331799 135104514455360 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 00:55:13.347912 135104514455360 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 00:55:13.347968 135104514455360 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 00:55:13.371368 135104514455360 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 00:55:13.371434 135104514455360 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 00:55:13.387611 135104514455360 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 00:55:13.387673 135104514455360 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 00:55:13.403988 135104514455360 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 00:55:13.404052 135104514455360 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 00:55:13.429322 135104514455360 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0418 00:55:13.429394 135104514455360 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 00:55:13.450208 135104514455360 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0418 00:55:13.450269 135104514455360 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0418 00:55:13.665229 135104514455360 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0418 00:55:13.665338 135104514455360 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0418 00:55:15.083093 135104514455360 maxtext_utils.py:1654] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0418 00:55:15.083218 135104514455360 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 00:55:15.083271 135104514455360 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 00:55:15.083330 135104514455360 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 00:55:15.083382 135104514455360 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 00:55:15.083420 135104514455360 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 00:55:15.083470 135104514455360 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 00:55:15.083521 135104514455360 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 00:55:15.083561 135104514455360 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 00:55:15.083598 135104514455360 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 00:55:15.083645 135104514455360 maxtext_utils.py:1654] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0418 00:55:15.083692 135104514455360 maxtext_utils.py:1654] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0418 00:55:15.574015 135104514455360 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0418 00:55:15.574111 135104514455360 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0418 00:55:15.589322 135104514455360 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0418 00:55:15.589384 135104514455360 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0418 00:55:26.479830 135104514455360 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 00:55:26.487725 135104514455360 metric_logger.py:301] number parameters: 1.104 billion
I0418 00:55:38.044487 135104514455360 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0418 00:55:38.198164 135104514455360 checkpointing.py:776] Waited 0.15366077423095703 seconds for step 0 to finish before starting checkpointing.
I0418 00:55:38.200588 135104514455360 checkpoint_manager.py:2009] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 00:55:38.202609 135104514455360 checkpoint_manager.py:1512] [process=0] Saving checkpoint at step 0
I0418 00:55:38.203947 135104514455360 event_tracking.py:70] [process=0] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0.
I0418 00:55:38.918936 135104514455360 signaling_client.py:364] Using JaxDistributedSignalingClient
I0418 00:55:38.932925 135104514455360 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 00:55:38.932989 135104514455360 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 00:55:39.203051 135104514455360 base_pytree_checkpoint_handler.py:154] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.270811s
I0418 00:55:39.203888 135104514455360 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/blocking_gbytes_per_sec: 5.564 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.27721738815307617 s) (per-host)
I0418 00:55:39.203985 135104514455360 base_pytree_checkpoint_handler.py:768] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.277313s (batch_requests_ready=0.002187s, total_serialization_initiated=0.274365s, others=0.000761s)
I0418 00:55:39.204069 135104514455360 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.281138s (all_items=0.000017s, per_item={'items': '0.00001669'}, temp_paths=0.281121)
I0418 00:55:39.204899 134979760523008 atomicity.py:140] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0
I0418 00:55:39.205759 135104514455360 event_tracking.py:125] [process=0] [async] Finished blocking save in 1.00 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0.
I0418 00:55:39.206072 134979768915712 async_checkpointer.py:76] [process=0][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-18 01:15:39.206034
I0418 00:55:39.221152 135104514455360 checkpoint_manager.py:1560] [process=0][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 00:55:39.221409 134978176280320 async_checkpointer.py:280] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0418 00:55:39.221555 135104514455360 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776473738.2005687, 'wait_for_prev_duration_secs': 6.318092346191406e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776473738.2026484, 'checkpointer_blocking_duration_secs': 1.0035758018493652, 'get_old_steps_start_time': 1776473739.2062502, 'get_old_steps_duration_secs': 3.4332275390625e-05, 'checkpoint_manager_blocking_start_time': 1776473738.1986806, 'checkpoint_manager_blocking_duration_secs': 1.0228357315063477}
I0418 00:55:39.221671 135104514455360 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0418 00:55:39.221741 135104514455360 max_utils.py:750]
Memstats: After params initialized:
I0418 00:55:39.221813 135104514455360 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_0(process=0,(0,0,0,0))
I0418 00:55:39.221896 135104514455360 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_1(process=0,(1,0,0,0))
I0418 00:55:39.221948 135104514455360 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_4(process=0,(0,1,0,0))
I0418 00:55:39.221995 135104514455360 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_5(process=0,(1,1,0,0))
I0418 00:55:39.558063 135104514455360 metric_logger.py:196] completed step: 0, seconds: 11.251, TFLOP/s/device: 1.208, Tokens/s/device: 182.030, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0418 00:55:39.561521 135104514455360 metric_logger.py:281] To see full metrics 'tensorboard --logdir=gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/tensorboard/'
I0418 00:55:39.964385 135104514455360 metric_logger.py:196] completed step: 1, seconds: 1.509, TFLOP/s/device: 9.002, Tokens/s/device: 1356.898, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0418 00:55:40.110307 135104514455360 metric_logger.py:196] completed step: 2, seconds: 0.409, TFLOP/s/device: 33.186, Tokens/s/device: 5002.088, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0418 00:55:40.257771 135104514455360 metric_logger.py:196] completed step: 3, seconds: 0.010, TFLOP/s/device: 1425.871, Tokens/s/device: 214922.867, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0418 00:55:40.989760 134979270993664 atomicity.py:140] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0/items
I0418 00:55:41.112768 134977128847104 checkpoint.py:188] Wrote Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776473740719508545, 'commit_timestamp_nsecs': None, 'custom_metadata': {}}, json={"item_handlers": null, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1776473740719508545, "commit_timestamp_nsecs": null, "custom_metadata": {}} to gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0/_CHECKPOINT_METADATA
I0418 00:55:42.541391 2751 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0418 00:55:45.571698 134978693625600 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0/items/array_metadatas/process_0
I0418 00:56:03.515604 135104514455360 metric_logger.py:196] completed step: 4, seconds: 0.147, TFLOP/s/device: 92.598, Tokens/s/device: 13957.419, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0418 00:56:03.524683 135104514455360 metric_logger.py:196] completed step: 5, seconds: 0.148, TFLOP/s/device: 91.653, Tokens/s/device: 13814.875, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0418 00:56:03.670746 135104514455360 metric_logger.py:196] completed step: 6, seconds: 23.256, TFLOP/s/device: 0.584, Tokens/s/device: 88.064, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0418 00:56:03.818421 135104514455360 metric_logger.py:196] completed step: 7, seconds: 0.006, TFLOP/s/device: 2094.194, Tokens/s/device: 315659.679, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0418 00:56:03.965863 135104514455360 metric_logger.py:196] completed step: 8, seconds: 0.147, TFLOP/s/device: 92.426, Tokens/s/device: 13931.404, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0418 00:56:04.112385 135104514455360 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0418 00:56:04.113121 135104514455360 checkpointing.py:776] Waited 0.0007512569427490234 seconds for step 9 to finish before starting checkpointing.
I0418 00:56:04.115066 135104514455360 checkpoint_manager.py:2020] [process=0][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 00:56:20.794070 134978685232896 base_pytree_checkpoint_handler.py:1282] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 38.252946s (commit=35.809397s, array_metadata_write=2.443549s)
I0418 00:56:20.795165 134979768915712 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/gbytes_per_sec: 37.728 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.86846160888672 s) (per-host)
I0418 00:56:20.795277 134979768915712 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.589088s.
I0418 00:56:21.512047 134979768915712 checkpoint.py:228] Read Metadata={'item_handlers': None, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776473740719508545, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} from gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0/_CHECKPOINT_METADATA
I0418 00:56:22.247629 134977128847104 checkpoint.py:247] Updated Metadata={'item_handlers': {'items': 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler'}, 'metrics': {}, 'performance_metrics': {}, 'init_timestamp_nsecs': 1776473740719508545, 'commit_timestamp_nsecs': None, 'custom_metadata': {}} to gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0/_CHECKPOINT_METADATA
I0418 00:56:22.670767 134979768915712 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000248s.
I0418 00:56:23.034312 134979768915712 ocdbt_utils.py:49] Param validation support for Zarr3 will be added later (b/362328389).
I0418 00:56:23.955467 134979768915712 base_pytree_checkpoint_handler.py:1406] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 2.321943s. use_zarr3=True, enable_post_merge_validation=True, directory=gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0/items
I0418 00:56:23.957236 134979768915712 atomicity.py:666] Finalizing gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0/items
I0418 00:56:24.477995 134979768915712 atomicity.py:666] Finalizing gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0
I0418 00:56:25.904452 134979768915712 atomicity.py:847] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0`.
I0418 00:56:25.905178 134979768915712 event_tracking.py:138] [process=0] [async] Finished save (blocking + background) in 47.70 seconds @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/0
I0418 00:56:25.906625 134979768915712 async_checkpointer.py:160] [process=0][thread=async_save] Background save thread done. Time taken: 46.700434s.
I0418 00:56:25.906854 134978176280320 async_checkpointer.py:288] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 00:56:25.906980 134978176280320 async_checkpointer.py:298] [process=0][thread=save_finalize] No errors found in background save thread=async_save.
I0418 00:56:25.907046 134978176280320 checkpoint_manager.py:2137] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0418 00:56:25.908746 134978176280320 checkpoint_manager.py:2146] [process=0][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0418 00:56:25.908887 135104514455360 checkpoint_manager.py:2032] [process=0][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0418 00:56:25.908974 135104514455360 checkpoint_manager.py:1452] Waiting for previous save to complete took 21.793910 seconds. If this number is high, consider checkpointing less frequently.
I0418 00:56:25.910719 135104514455360 checkpoint_manager.py:1512] [process=0] Saving checkpoint at step 9
I0418 00:56:25.912880 135104514455360 event_tracking.py:70] [process=0] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/9.
I0418 00:56:26.644567 135104514455360 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 00:56:26.644666 135104514455360 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 00:56:26.670660 135104514455360 base_pytree_checkpoint_handler.py:154] [process=0][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.026845s
I0418 00:56:26.671412 135104514455360 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/blocking_gbytes_per_sec: 48.847 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.03157949447631836 s) (per-host)
I0418 00:56:26.671483 135104514455360 base_pytree_checkpoint_handler.py:768] [process=0][thread=MainThread] Initiated Pytree async_save. Time taken: 0.031666s (batch_requests_ready=0.002070s, total_serialization_initiated=0.028897s, others=0.000699s)
I0418 00:56:26.671554 135104514455360 composite_checkpoint_handler.py:715] [process=0][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.035596s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.035579)
I0418 00:56:26.672255 135104514455360 event_tracking.py:125] [process=0] [async] Finished blocking save in 0.76 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/9.
I0418 00:56:26.672569 134977660380928 async_checkpointer.py:76] [process=0][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-18 01:16:26.672529
I0418 00:56:26.680106 135104514455360 checkpoint_manager.py:1560] [process=0][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 00:56:26.680407 134978176280320 async_checkpointer.py:280] [process=0][thread=save_finalize] Waiting for background save thread=async_save.
I0418 00:56:26.680587 135104514455360 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776473764.115035, 'wait_for_prev_duration_secs': 21.793909549713135, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776473785.9107597, 'checkpointer_blocking_duration_secs': 0.7619459629058838, 'get_old_steps_start_time': 1776473786.6727312, 'get_old_steps_duration_secs': 3.170967102050781e-05, 'checkpoint_manager_blocking_start_time': 1776473764.11335, 'checkpoint_manager_blocking_duration_secs': 22.56719970703125}
I0418 00:56:26.680695 135104514455360 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0418 00:56:26.680740 135104514455360 checkpoint_manager.py:2020] [process=0][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 00:56:26.875300 134979768915712 atomicity.py:140] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/9
I0418 00:56:28.704151 134979245815552 atomicity.py:140] Creating tmp directory gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/9/items
I0418 00:56:32.494665 134978152519424 array_metadata_store.py:203] [process=0][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/9/items/array_metadatas/process_0
I0418 00:57:11.791629 134977668773632 base_pytree_checkpoint_handler.py:1282] [process=0][thread=write_metadata_after_commits] Commit + Array metadata written. Time taken: 41.476881s (commit=40.301160s, array_metadata_write=1.175720s)
I0418 00:57:11.792570 134977660380928 base_pytree_checkpoint_handler.py:130] [process=0] /jax/orbax/write/gbytes_per_sec: 34.983 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 45.15273451805115 s) (per-host)
I0418 00:57:11.792634 134977660380928 async_checkpointer.py:90] [process=0][thread=async_save] 3 Handler Commit operations completed. Time taken: 45.119956s.
I0418 00:57:14.340596 134977660380928 ocdbt_utils.py:49] Param validation support for Zarr3 will be added later (b/362328389).
I0418 00:57:14.555011 134977660380928 array_metadata_store.py:411] [process=0][thread=async_save] Validated ArrayMetadata from all 8 hosts. Time taken: 0.000259s.
I0418 00:57:15.275465 134977660380928 base_pytree_checkpoint_handler.py:1406] [process=0][thread=async_save] Pytree save finalize (merge_ocdbt + ArrayMetadata validation) completed. Time taken: 2.188953s. use_zarr3=True, enable_post_merge_validation=True, directory=gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/9/items
I0418 00:57:15.277161 134977660380928 atomicity.py:666] Finalizing gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/9/items
I0418 00:57:15.824227 134977660380928 atomicity.py:666] Finalizing gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/9
I0418 00:57:18.907746 134977660380928 atomicity.py:847] [process=0][thread=async_save] Finished saving checkpoint (finalized tmp dir) to `gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/9`.
I0418 00:57:18.908615 134977660380928 event_tracking.py:138] [process=0] [async] Finished save (blocking + background) in 53.00 seconds @ gs://lance-maxtext/linen_ckpt_xpk_main_20260417_235819/linen_xpk_main_20260417_235819_11_optimizer_offload_false/checkpoints/9
I0418 00:57:18.910002 134977660380928 async_checkpointer.py:160] [process=0][thread=async_save] Background save thread done. Time taken: 52.237321s.
I0418 00:57:18.910220 134978176280320 async_checkpointer.py:288] [process=0][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 00:57:18.910334 134978176280320 async_checkpointer.py:298] [process=0][thread=save_finalize] No errors found in background save thread=async_save.
I0418 00:57:18.910395 134978176280320 checkpoint_manager.py:2137] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0418 00:57:18.912620 134978176280320 checkpoint_manager.py:2146] [process=0][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0418 00:57:18.912792 135104514455360 checkpoint_manager.py:2032] [process=0][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0418 00:57:18.912960 135104514455360 checkpoint_manager.py:2009] [process=0][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 00:57:18.913877 135104514455360 metric_logger.py:196] completed step: 9, seconds: 0.148, TFLOP/s/device: 91.900, Tokens/s/device: 13852.158, 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 00:57:29 UTC 2026
EXIT_CODE=0