XPK Start: Mon Apr 20 06:18:32 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 06:18:56.286028: 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 06:18:56.823451 137946811504448 max_utils.py:273] Attempting to initialize the jax distributed system...
I0420 06:19:05.865581 137946811504448 distributed.py:149] Starting JAX distributed service on [::]:8482
I0420 06:19:05.867967 137946811504448 distributed.py:172] Connecting to JAX distributed service on mt-07-eval-3ko6e-slice-job-0-0.mt-07-eval-3ko6e:8482
I0420 06:19:07.562508 137946811504448 max_utils.py:284] Jax distributed system initialized!
I0420 06:19:13.675007 137946811504448 max_utils.py:800] System Information: Jax Version: 0.9.2
I0420 06:19:13.675110 137946811504448 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0420 06:19:13.675151 137946811504448 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 06:19:13.675187 137946811504448 train_utils.py:348] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0420 06:19:14.447207 137946811504448 maxtext_utils.py:1551] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 06:19:14.447512 137946811504448 checkpointing.py:677] Setting up checkpoint logger...
I0420 06:19:14.447568 137946811504448 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0420 06:19:14.447615 137946811504448 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0420 06:19:14.447959 137946811504448 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 0x7d75a5d92a20>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0420 06:19:17.579974 137946811504448 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0420 06:19:17.580208 137946811504448 checkpoint_manager.py:708] [process=2][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d606c5b5af0>}, handler_registry=None
I0420 06:19:17.580447 137946811504448 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d606c5b5af0>` 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 06:19:17.580512 137946811504448 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d614021ce00>` 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 06:19:17.580550 137946811504448 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 0x7d606c5b5af0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d606c5b5af0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d614021ce00>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d614021ce00>}).
I0420 06:19:17.580871 137946811504448 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0420 06:19:17.580943 137946811504448 async_checkpointer.py:192] [process=2][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d604c51b7e0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0420 06:19:18.296594 137946811504448 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_07_eval/checkpoints
I0420 06:19:18.730435 137946811504448 checkpoint_manager.py:929] [process=2][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_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d747aae9c40>
I0420 06:19:18.730635 137946811504448 checkpointing.py:301] Checkpoint manager created!
I0420 06:19:19.652753 137946811504448 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0420 06:19:19.652871 137946811504448 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 06:19:20.033174 137946811504448 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 06:19:20.033265 137946811504448 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 06:19:20.049696 137946811504448 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 06:19:20.049757 137946811504448 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 06:19:20.073678 137946811504448 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 06:19:20.073743 137946811504448 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 06:19:20.090249 137946811504448 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 06:19:20.090312 137946811504448 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 06:19:20.113429 137946811504448 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 06:19:20.113535 137946811504448 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 06:19:20.139798 137946811504448 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0420 06:19:20.139868 137946811504448 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 06:19:20.160865 137946811504448 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0420 06:19:20.160929 137946811504448 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0420 06:19:20.383670 137946811504448 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0420 06:19:20.383789 137946811504448 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0420 06:19:21.817003 137946811504448 maxtext_utils.py:1654] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0420 06:19:21.817125 137946811504448 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 06:19:21.817180 137946811504448 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 06:19:21.817239 137946811504448 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 06:19:21.817301 137946811504448 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 06:19:21.817342 137946811504448 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 06:19:21.817395 137946811504448 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 06:19:21.817448 137946811504448 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 06:19:21.817506 137946811504448 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 06:19:21.817553 137946811504448 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 06:19:21.817605 137946811504448 maxtext_utils.py:1654] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0420 06:19:21.817655 137946811504448 maxtext_utils.py:1654] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0420 06:19:22.319468 137946811504448 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 06:19:22.319574 137946811504448 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 06:19:22.334954 137946811504448 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 06:19:22.335013 137946811504448 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 06:19:33.321103 137946811504448 max_utils.py:791] Total memory size: 1.5 GB, Output size: 0.4 GB, Temp size: 1.1 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0420 06:19:33.323501 137946811504448 metric_logger.py:301] number parameters: 1.104 billion
I0420 06:19:44.475430 137946811504448 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0420 06:19:44.627908 137946811504448 checkpointing.py:776] Waited 0.15245842933654785 seconds for step 0 to finish before starting checkpointing.
I0420 06:19:44.630672 137946811504448 checkpoint_manager.py:2009] [process=2][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 06:19:44.632270 137946811504448 checkpoint_manager.py:1512] [process=2] Saving checkpoint at step 0
I0420 06:19:44.633678 137946811504448 event_tracking.py:70] [process=2] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_07_eval/checkpoints/0.
I0420 06:19:44.979087 137946811504448 signaling_client.py:364] Using JaxDistributedSignalingClient
I0420 06:19:44.980053 137946811504448 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0420 06:19:44.980109 137946811504448 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 06:19:45.256725 137946811504448 base_pytree_checkpoint_handler.py:154] [process=2][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.277714s
I0420 06:19:45.256896 137946811504448 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/blocking_gbytes_per_sec: 5.442 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2834746837615967 s) (per-host)
I0420 06:19:45.256950 137946811504448 base_pytree_checkpoint_handler.py:768] [process=2][thread=MainThread] Initiated Pytree async_save. Time taken: 0.283537s (batch_requests_ready=0.002440s, total_serialization_initiated=0.281027s, others=0.000070s)
I0420 06:19:45.257042 137946811504448 composite_checkpoint_handler.py:715] [process=2][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.287465s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.287448)
I0420 06:19:45.257860 137946811504448 event_tracking.py:125] [process=2] [async] Finished blocking save in 0.63 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_07_eval/checkpoints/0.
I0420 06:19:45.258120 137818383931136 async_checkpointer.py:76] [process=2][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 06:39:45.258098
I0420 06:19:45.265630 137946811504448 checkpoint_manager.py:1560] [process=2][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 06:19:45.265877 137816239503104 async_checkpointer.py:280] [process=2][thread=save_finalize] Waiting for background save thread=async_save.
I0420 06:19:45.265995 137946811504448 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776665984.630655, 'wait_for_prev_duration_secs': 6.794929504394531e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776665984.6323078, 'checkpointer_blocking_duration_secs': 0.6259200572967529, 'get_old_steps_start_time': 1776665985.2582526, 'get_old_steps_duration_secs': 3.123283386230469e-05, 'checkpoint_manager_blocking_start_time': 1776665984.628316, 'checkpoint_manager_blocking_duration_secs': 0.6376402378082275}
I0420 06:19:45.266105 137946811504448 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0420 06:19:45.266157 137946811504448 max_utils.py:750]
Memstats: After params initialized:
I0420 06:19:45.266205 137946811504448 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_8(process=2,(0,2,0,0))
I0420 06:19:45.266237 137946811504448 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_9(process=2,(1,2,0,0))
I0420 06:19:45.266264 137946811504448 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_12(process=2,(0,3,0,0))
I0420 06:19:45.266301 137946811504448 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_13(process=2,(1,3,0,0))
I0420 06:19:45.580301 137946811504448 metric_logger.py:196] completed step: 0, seconds: 11.152, TFLOP/s/device: 1.218, Tokens/s/device: 183.647, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0420 06:19:45.756082 137946811504448 metric_logger.py:196] completed step: 1, seconds: 1.103, TFLOP/s/device: 12.314, Tokens/s/device: 1856.113, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0420 06:19:46.175252 137946811504448 metric_logger.py:196] completed step: 2, seconds: 0.028, TFLOP/s/device: 479.010, Tokens/s/device: 72201.657, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0420 06:19:46.322533 137946811504448 metric_logger.py:196] completed step: 3, seconds: 0.419, TFLOP/s/device: 32.393, Tokens/s/device: 4882.689, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0420 06:19:46.617763 137946811504448 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.699, Tokens/s/device: 13369.630, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0420 06:19:46.623719 137946811504448 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.274, Tokens/s/device: 13908.603, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0420 06:19:48.652718 2833 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0420 06:19:50.881720 137817795004160 array_metadata_store.py:203] [process=2][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_07_eval/checkpoints/0/items/array_metadatas/process_2
I0420 06:20:08.799435 137946811504448 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.951, Tokens/s/device: 6926.290, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0420 06:20:08.946750 137946811504448 metric_logger.py:196] completed step: 7, seconds: 22.029, TFLOP/s/device: 0.617, Tokens/s/device: 92.970, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0420 06:20:09.094133 137946811504448 metric_logger.py:196] completed step: 8, seconds: 0.153, TFLOP/s/device: 88.968, Tokens/s/device: 13410.250, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0420 06:20:09.240830 137946811504448 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0420 06:20:09.241440 137946811504448 checkpointing.py:776] Waited 0.0006325244903564453 seconds for step 9 to finish before starting checkpointing.
I0420 06:20:09.243559 137946811504448 checkpoint_manager.py:2020] [process=2][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 06:20:21.911314 137818383931136 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/gbytes_per_sec: 42.764 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.93784976005554 s) (per-host)
I0420 06:20:21.911441 137818383931136 async_checkpointer.py:90] [process=2][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.653250s.
I0420 06:20:30.847848 137818383931136 async_checkpointer.py:160] [process=2][thread=async_save] Background save thread done. Time taken: 45.589640s.
I0420 06:20:30.848080 137816239503104 async_checkpointer.py:288] [process=2][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 06:20:30.848140 137816239503104 async_checkpointer.py:298] [process=2][thread=save_finalize] No errors found in background save thread=async_save.
I0420 06:20:30.848196 137816239503104 checkpoint_manager.py:2137] [process=2][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0420 06:20:30.849864 137816239503104 checkpoint_manager.py:2146] [process=2][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0420 06:20:30.850039 137946811504448 checkpoint_manager.py:2032] [process=2][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0420 06:20:30.850180 137946811504448 checkpoint_manager.py:1452] Waiting for previous save to complete took 21.606621 seconds. If this number is high, consider checkpointing less frequently.
I0420 06:20:30.852458 137946811504448 checkpoint_manager.py:1512] [process=2] Saving checkpoint at step 9
I0420 06:20:30.854467 137946811504448 event_tracking.py:70] [process=2] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_07_eval/checkpoints/9.
I0420 06:20:31.138386 137946811504448 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0420 06:20:31.138489 137946811504448 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 06:20:31.174917 137946811504448 base_pytree_checkpoint_handler.py:154] [process=2][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.037614s
I0420 06:20:31.175105 137946811504448 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/blocking_gbytes_per_sec: 37.343 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04130864143371582 s) (per-host)
I0420 06:20:31.175180 137946811504448 base_pytree_checkpoint_handler.py:768] [process=2][thread=MainThread] Initiated Pytree async_save. Time taken: 0.041396s (batch_requests_ready=0.001993s, total_serialization_initiated=0.039304s, others=0.000098s)
I0420 06:20:31.175290 137946811504448 composite_checkpoint_handler.py:715] [process=2][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.045624s (all_items=0.000015s, per_item={'items': '0.00001478'}, temp_paths=0.045609)
I0420 06:20:31.175979 137946811504448 event_tracking.py:125] [process=2] [async] Finished blocking save in 0.32 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_07_eval/checkpoints/9.
I0420 06:20:31.176321 137816239503104 async_checkpointer.py:76] [process=2][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 06:40:31.176283
I0420 06:20:31.178429 137946811504448 checkpoint_manager.py:1560] [process=2][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 06:20:31.178735 137817795004160 async_checkpointer.py:280] [process=2][thread=save_finalize] Waiting for background save thread=async_save.
I0420 06:20:31.178896 137946811504448 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776666009.2435288, 'wait_for_prev_duration_secs': 21.60662078857422, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776666030.8525116, 'checkpointer_blocking_duration_secs': 0.32398462295532227, 'get_old_steps_start_time': 1776666031.17652, 'get_old_steps_duration_secs': 4.1484832763671875e-05, 'checkpoint_manager_blocking_start_time': 1776666009.2416918, 'checkpoint_manager_blocking_duration_secs': 21.93716859817505}
I0420 06:20:31.179002 137946811504448 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0420 06:20:31.179046 137946811504448 checkpoint_manager.py:2020] [process=2][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 06:20:36.770922 137796329129728 array_metadata_store.py:203] [process=2][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260420_060028/linen_xpk_main_20260420_060028_07_eval/checkpoints/9/items/array_metadatas/process_2
I0420 06:21:13.559109 137816239503104 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/gbytes_per_sec: 37.232 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.425278425216675 s) (per-host)
I0420 06:21:13.559235 137816239503104 async_checkpointer.py:90] [process=2][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.382795s.
I0420 06:21:21.503290 137816239503104 async_checkpointer.py:160] [process=2][thread=async_save] Background save thread done. Time taken: 50.326835s.
I0420 06:21:21.503593 137817795004160 async_checkpointer.py:288] [process=2][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 06:21:21.503736 137817795004160 async_checkpointer.py:298] [process=2][thread=save_finalize] No errors found in background save thread=async_save.
I0420 06:21:21.503787 137817795004160 checkpoint_manager.py:2137] [process=2][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0420 06:21:21.505390 137817795004160 checkpoint_manager.py:2146] [process=2][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0420 06:21:21.505578 137946811504448 checkpoint_manager.py:2032] [process=2][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0420 06:21:21.505729 137946811504448 checkpoint_manager.py:2009] [process=2][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 06:21:21.506696 137946811504448 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.235, Tokens/s/device: 13902.654, 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: Mon Apr 20 06:21:32 UTC 2026
EXIT_CODE=0