XPK Start: Fri Apr 24 07:21:43 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
`rope_parameters`'s factor field must be a float >= 1, got 40
`rope_parameters`'s beta_fast field must be a float, got 32
`rope_parameters`'s beta_slow field must be a float, got 1
DeepseekV32Config got `key=rope_scaling` in kwargs but hasn't set it as attribute. For RoPE standardization you need to set `self.rope_parameters` in model's config.
2026-04-24 07:22:08.376481: 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)
I0424 07:22:08.580928 140625103714112 max_utils.py:273] Attempting to initialize the jax distributed system...
I0424 07:22:17.622976 140625103714112 distributed.py:149] Starting JAX distributed service on [::]:8482
I0424 07:22:17.625506 140625103714112 distributed.py:172] Connecting to JAX distributed service on mt-07-eval-t6xja-slice-job-0-0.mt-07-eval-t6xja:8482
I0424 07:22:18.807710 140625103714112 max_utils.py:284] Jax distributed system initialized!
I0424 07:22:24.800419 140625103714112 max_utils.py:800] System Information: Jax Version: 0.9.2
I0424 07:22:24.800529 140625103714112 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0424 07:22:24.800573 140625103714112 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
I0424 07:22:24.800608 140625103714112 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0424 07:22:25.823918 140625103714112 maxtext_utils.py:1604] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1)
I0424 07:22:25.824213 140625103714112 checkpointing.py:677] Setting up checkpoint logger...
I0424 07:22:25.824268 140625103714112 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0424 07:22:25.824312 140625103714112 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0424 07:22:25.824654 140625103714112 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 0x7fe51df3c110>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0424 07:22:28.393269 140625103714112 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0424 07:22:28.393506 140625103714112 checkpoint_manager.py:708] [process=6][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fd0e85ab6b0>}, handler_registry=None
I0424 07:22:28.393743 140625103714112 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fd0e85ab6b0>` 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`.
I0424 07:22:28.393795 140625103714112 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fd0e837e090>` 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`.
I0424 07:22:28.393833 140625103714112 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 0x7fd0e85ab6b0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fd0e85ab6b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fd0e837e090>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fd0e837e090>}).
I0424 07:22:28.394180 140625103714112 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0424 07:22:28.394266 140625103714112 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7fcfd0671760> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0424 07:22:29.232524 140625103714112 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_07_eval/checkpoints
I0424 07:22:29.251498 140625103714112 checkpoint_manager.py:929] [process=6][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7fd0e837e870>
I0424 07:22:29.251625 140625103714112 checkpointing.py:301] Checkpoint manager created!
I0424 07:22:30.189830 140625103714112 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0424 07:22:30.189949 140625103714112 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 07:22:30.572884 140625103714112 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch_attn', 'activation_length_attn', 'activation_embed_attn').
I0424 07:22:30.572976 140625103714112 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 07:22:30.589530 140625103714112 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch_attn', 'activation_length_attn', 'activation_embed_attn').
I0424 07:22:30.589591 140625103714112 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 07:22:30.613392 140625103714112 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 07:22:30.613461 140625103714112 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 07:22:30.629988 140625103714112 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 07:22:30.630050 140625103714112 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 07:22:30.646587 140625103714112 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 07:22:30.646647 140625103714112 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 07:22:30.671579 140625103714112 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch_attn', 'activation_length_attn', 'activation_heads', 'activation_kv').
I0424 07:22:30.671649 140625103714112 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 07:22:30.693171 140625103714112 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0424 07:22:30.693240 140625103714112 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0424 07:22:30.907037 140625103714112 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0424 07:22:30.907155 140625103714112 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0424 07:22:32.358543 140625103714112 maxtext_utils.py:1707] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0424 07:22:32.358675 140625103714112 maxtext_utils.py:1707] params/params/decoder/layers/mlp/wi_0/kernel
Shape: float32[2048,16,7168]
Logical: P('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0424 07:22:32.358731 140625103714112 maxtext_utils.py:1707] params/params/decoder/layers/mlp/wi_1/kernel
Shape: float32[2048,16,7168]
Logical: P('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0424 07:22:32.358791 140625103714112 maxtext_utils.py:1707] params/params/decoder/layers/mlp/wo/kernel
Shape: float32[7168,16,2048]
Logical: P('mlp', 'layers', 'embed')
Physical: (None, None, 'fsdp')
I0424 07:22:32.358845 140625103714112 maxtext_utils.py:1707] params/params/decoder/layers/post_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: P('norm', 'layers')
Physical: (None, None)
I0424 07:22:32.358883 140625103714112 maxtext_utils.py:1707] params/params/decoder/layers/pre_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: P('norm', 'layers')
Physical: (None, None)
I0424 07:22:32.358939 140625103714112 maxtext_utils.py:1707] 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)
I0424 07:22:32.358995 140625103714112 maxtext_utils.py:1707] 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')
I0424 07:22:32.359035 140625103714112 maxtext_utils.py:1707] 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)
I0424 07:22:32.359071 140625103714112 maxtext_utils.py:1707] 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)
I0424 07:22:32.359145 140625103714112 maxtext_utils.py:1707] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0424 07:22:32.359205 140625103714112 maxtext_utils.py:1707] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0424 07:22:32.854302 140625103714112 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0424 07:22:32.854401 140625103714112 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0424 07:22:32.870019 140625103714112 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0424 07:22:32.870081 140625103714112 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0424 07:22:43.770426 140625103714112 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.
I0424 07:22:43.771230 140625103714112 metric_logger.py:301] number parameters: 1.104 billion
I0424 07:22:55.138598 140625103714112 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0424 07:22:55.350878 140625103714112 checkpointing.py:776] Waited 0.21225810050964355 seconds for step 0 to finish before starting checkpointing.
I0424 07:22:55.353432 140625103714112 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0424 07:22:55.355187 140625103714112 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0424 07:22:55.356918 140625103714112 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_07_eval/checkpoints/0.
I0424 07:22:55.651172 140625103714112 signaling_client.py:364] Using JaxDistributedSignalingClient
I0424 07:22:55.652159 140625103714112 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0424 07:22:55.652216 140625103714112 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
I0424 07:22:55.928416 140625103714112 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.277332s
I0424 07:22:55.928587 140625103714112 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.456 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28270626068115234 s) (per-host)
I0424 07:22:55.928638 140625103714112 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.282767s (batch_requests_ready=0.002187s, total_serialization_initiated=0.280511s, others=0.000069s)
I0424 07:22:55.928727 140625103714112 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.286914s (all_items=0.000018s, per_item={'items': '0.00001812'}, temp_paths=0.286896)
I0424 07:22:55.929483 140625103714112 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.57 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_07_eval/checkpoints/0.
I0424 07:22:55.929822 140495225353984 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 07:42:55.929785
I0424 07:22:55.935091 140625103714112 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 07:22:55.935360 140494143219456 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0424 07:22:55.935492 140625103714112 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777015375.353414, 'wait_for_prev_duration_secs': 5.9604644775390625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777015375.355267, 'checkpointer_blocking_duration_secs': 0.5747091770172119, 'get_old_steps_start_time': 1777015375.9300013, 'get_old_steps_duration_secs': 3.0040740966796875e-05, 'checkpoint_manager_blocking_start_time': 1777015375.3514335, 'checkpoint_manager_blocking_duration_secs': 0.5840201377868652}
I0424 07:22:55.935596 140625103714112 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0424 07:22:55.935647 140625103714112 max_utils.py:750]
Memstats: After params initialized:
I0424 07:22:55.935693 140625103714112 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0424 07:22:55.935728 140625103714112 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0424 07:22:55.935752 140625103714112 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0424 07:22:55.935775 140625103714112 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0424 07:22:56.251585 140625103714112 metric_logger.py:196] completed step: 0, seconds: 11.367, TFLOP/s/device: 1.195, Tokens/s/device: 180.167, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0424 07:22:56.419086 140625103714112 metric_logger.py:196] completed step: 1, seconds: 1.112, TFLOP/s/device: 12.223, Tokens/s/device: 1842.434, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0424 07:22:56.825861 140625103714112 metric_logger.py:196] completed step: 2, seconds: 0.020, TFLOP/s/device: 673.297, Tokens/s/device: 101486.620, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0424 07:22:56.973339 140625103714112 metric_logger.py:196] completed step: 3, seconds: 0.407, TFLOP/s/device: 33.392, Tokens/s/device: 5033.178, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0424 07:22:57.268280 140625103714112 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.621, Tokens/s/device: 13357.945, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0424 07:22:57.274333 140625103714112 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.181, Tokens/s/device: 13894.543, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0424 07:22:59.248360 2878 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0424 07:23:01.707972 140494151612160 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_07_eval/checkpoints/0/items/array_metadatas/process_6
I0424 07:23:18.806150 140625103714112 metric_logger.py:196] completed step: 6, seconds: 0.295, TFLOP/s/device: 45.994, Tokens/s/device: 6932.691, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0424 07:23:18.953476 140625103714112 metric_logger.py:196] completed step: 7, seconds: 21.385, TFLOP/s/device: 0.635, Tokens/s/device: 95.768, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0424 07:23:19.100585 140625103714112 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.131, Tokens/s/device: 13434.794, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0424 07:23:19.247223 140625103714112 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0424 07:23:19.247885 140625103714112 checkpointing.py:776] Waited 0.0006825923919677734 seconds for step 9 to finish before starting checkpointing.
I0424 07:23:19.249952 140625103714112 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 07:23:33.353331 140495225353984 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 41.891 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.70741868019104 s) (per-host)
I0424 07:23:33.353439 140495225353984 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.423507s.
I0424 07:23:41.666528 140495225353984 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 45.736579s.
I0424 07:23:41.666748 140494143219456 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 07:23:41.666820 140494143219456 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0424 07:23:41.666884 140494143219456 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0424 07:23:41.669747 140494143219456 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0424 07:23:41.669933 140625103714112 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0424 07:23:41.670070 140625103714112 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.420119 seconds. If this number is high, consider checkpointing less frequently.
I0424 07:23:41.672010 140625103714112 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0424 07:23:41.673894 140625103714112 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_07_eval/checkpoints/9.
I0424 07:23:41.966453 140625103714112 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0424 07:23:41.966554 140625103714112 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
I0424 07:23:42.012824 140625103714112 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.047394s
I0424 07:23:42.013002 140625103714112 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 29.896 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.05159759521484375 s) (per-host)
I0424 07:23:42.013052 140625103714112 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.051656s (batch_requests_ready=0.002477s, total_serialization_initiated=0.049112s, others=0.000067s)
I0424 07:23:42.013155 140625103714112 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.056384s (all_items=0.000015s, per_item={'items': '0.00001526'}, temp_paths=0.056369)
I0424 07:23:42.013903 140625103714112 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.34 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_07_eval/checkpoints/9.
I0424 07:23:42.014218 140494143219456 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 07:43:42.014185
I0424 07:23:42.016310 140625103714112 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 07:23:42.016595 140495207380736 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0424 07:23:42.016755 140625103714112 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777015399.249921, 'wait_for_prev_duration_secs': 22.420119285583496, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777015421.6720488, 'checkpointer_blocking_duration_secs': 0.34230971336364746, 'get_old_steps_start_time': 1777015422.0143847, 'get_old_steps_duration_secs': 3.170967102050781e-05, 'checkpoint_manager_blocking_start_time': 1777015399.2481287, 'checkpoint_manager_blocking_duration_secs': 22.768593072891235}
I0424 07:23:42.016858 140625103714112 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0424 07:23:42.016899 140625103714112 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 07:23:47.135390 140495215773440 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_07_eval/checkpoints/9/items/array_metadatas/process_6
I0424 07:24:24.763855 140494143219456 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 36.904 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.80241322517395 s) (per-host)
I0424 07:24:24.763977 140494143219456 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.749651s.
I0424 07:24:31.399799 140494143219456 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 49.385459s.
I0424 07:24:31.400086 140495207380736 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 07:24:31.400238 140495207380736 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0424 07:24:31.400291 140495207380736 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0424 07:24:31.402794 140495207380736 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0424 07:24:31.402967 140625103714112 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0424 07:24:31.403147 140625103714112 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0424 07:24:31.404057 140625103714112 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.319, Tokens/s/device: 13915.312, 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: Fri Apr 24 07:24:43 UTC 2026
EXIT_CODE=0