XPK Start: Tue Apr 21 06:15:38 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.
2026-04-21 06:16:18.635243: 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)
I0421 06:16:18.844755 132384911742784 max_utils.py:273] Attempting to initialize the jax distributed system...
I0421 06:16:27.887047 132384911742784 distributed.py:149] Starting JAX distributed service on [::]:8482
I0421 06:16:27.889302 132384911742784 distributed.py:172] Connecting to JAX distributed service on mt-01-base-9m7bb-slice-job-0-0.mt-01-base-9m7bb:8482
I0421 06:16:32.655886 132384911742784 max_utils.py:284] Jax distributed system initialized!
I0421 06:16:39.765306 132384911742784 max_utils.py:800] System Information: Jax Version: 0.9.2
I0421 06:16:39.765409 132384911742784 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0421 06:16:39.765456 132384911742784 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
I0421 06:16:39.765502 132384911742784 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0421 06:16:40.465653 132384911742784 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 06:16:40.465956 132384911742784 checkpointing.py:677] Setting up checkpoint logger...
I0421 06:16:40.466017 132384911742784 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0421 06:16:40.466068 132384911742784 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0421 06:16:40.466443 132384911742784 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 0x7866a331c3b0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0421 06:16:44.684265 132384911742784 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0421 06:16:44.684495 132384911742784 checkpoint_manager.py:708] [process=7][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7852f0729640>}, handler_registry=None
I0421 06:16:44.684759 132384911742784 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7852f0729640>` 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`.
I0421 06:16:44.684811 132384911742784 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7852f0728740>` 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`.
I0421 06:16:44.684847 132384911742784 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 0x7852f0729640>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7852f0729640>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7852f0728740>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7852f0728740>}).
I0421 06:16:44.685160 132384911742784 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0421 06:16:44.685230 132384911742784 async_checkpointer.py:192] [process=7][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7852f0579760> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0421 06:16:46.353711 132384911742784 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_01_base/checkpoints
I0421 06:16:46.363177 132384911742784 checkpoint_manager.py:929] [process=7][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_20260421_061359/linen_xpk_main_20260421_061359_01_base/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7852f072b470>
I0421 06:16:46.363296 132384911742784 checkpointing.py:301] Checkpoint manager created!
I0421 06:16:47.293864 132384911742784 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0421 06:16:47.293974 132384911742784 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 06:16:47.675949 132384911742784 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 06:16:47.676041 132384911742784 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 06:16:47.692131 132384911742784 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 06:16:47.692192 132384911742784 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 06:16:47.715620 132384911742784 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 06:16:47.715690 132384911742784 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 06:16:47.731821 132384911742784 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 06:16:47.731892 132384911742784 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 06:16:47.747999 132384911742784 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 06:16:47.748056 132384911742784 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 06:16:47.772489 132384911742784 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0421 06:16:47.772559 132384911742784 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 06:16:47.793289 132384911742784 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0421 06:16:47.793356 132384911742784 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0421 06:16:48.002451 132384911742784 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0421 06:16:48.002564 132384911742784 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0421 06:16:49.416894 132384911742784 maxtext_utils.py:1668] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0421 06:16:49.417021 132384911742784 maxtext_utils.py:1668] params/params/decoder/layers/mlp/wi_0/kernel
Shape: float32[2048,16,7168]
Logical: P('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0421 06:16:49.417074 132384911742784 maxtext_utils.py:1668] params/params/decoder/layers/mlp/wi_1/kernel
Shape: float32[2048,16,7168]
Logical: P('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0421 06:16:49.417130 132384911742784 maxtext_utils.py:1668] params/params/decoder/layers/mlp/wo/kernel
Shape: float32[7168,16,2048]
Logical: P('mlp', 'layers', 'embed')
Physical: (None, None, 'fsdp')
I0421 06:16:49.417180 132384911742784 maxtext_utils.py:1668] params/params/decoder/layers/post_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: P('norm', 'layers')
Physical: (None, None)
I0421 06:16:49.417218 132384911742784 maxtext_utils.py:1668] params/params/decoder/layers/pre_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: P('norm', 'layers')
Physical: (None, None)
I0421 06:16:49.417269 132384911742784 maxtext_utils.py:1668] 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)
I0421 06:16:49.417319 132384911742784 maxtext_utils.py:1668] 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')
I0421 06:16:49.417358 132384911742784 maxtext_utils.py:1668] 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)
I0421 06:16:49.417394 132384911742784 maxtext_utils.py:1668] 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)
I0421 06:16:49.417439 132384911742784 maxtext_utils.py:1668] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0421 06:16:49.417486 132384911742784 maxtext_utils.py:1668] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0421 06:16:49.910195 132384911742784 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 06:16:49.910290 132384911742784 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 06:16:49.925569 132384911742784 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 06:16:49.925627 132384911742784 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 06:17:00.770133 132384911742784 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.
I0421 06:17:00.770972 132384911742784 metric_logger.py:301] number parameters: 1.104 billion
I0421 06:17:12.335058 132384911742784 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0421 06:17:12.490858 132384911742784 checkpointing.py:776] Waited 0.15578174591064453 seconds for step 0 to finish before starting checkpointing.
I0421 06:17:12.493099 132384911742784 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 06:17:12.494973 132384911742784 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 0
I0421 06:17:12.496426 132384911742784 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_01_base/checkpoints/0.
I0421 06:17:13.253249 132384911742784 signaling_client.py:364] Using JaxDistributedSignalingClient
I0421 06:17:13.254201 132384911742784 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 06:17:13.254257 132384911742784 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
I0421 06:17:13.527092 132384911742784 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.273911s
I0421 06:17:13.527264 132384911742784 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 5.520 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.27945828437805176 s) (per-host)
I0421 06:17:13.527318 132384911742784 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.279523s (batch_requests_ready=0.002296s, total_serialization_initiated=0.277155s, others=0.000072s)
I0421 06:17:13.527408 132384911742784 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.284144s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.284127)
I0421 06:17:13.528271 132384911742784 event_tracking.py:125] [process=7] [async] Finished blocking save in 1.03 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_01_base/checkpoints/0.
I0421 06:17:13.528547 132258411493120 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 06:37:13.528522
I0421 06:17:13.537492 132384911742784 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 06:17:13.537796 132256765028096 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save.
I0421 06:17:13.537952 132384911742784 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_01_base/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776752232.493082, 'wait_for_prev_duration_secs': 6.103515625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776752232.495019, 'checkpointer_blocking_duration_secs': 1.0336425304412842, 'get_old_steps_start_time': 1776752233.528688, 'get_old_steps_duration_secs': 2.7418136596679688e-05, 'checkpoint_manager_blocking_start_time': 1776752232.4912667, 'checkpoint_manager_blocking_duration_secs': 1.0466454029083252}
I0421 06:17:13.538060 132384911742784 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0421 06:17:13.538110 132384911742784 max_utils.py:750]
Memstats: After params initialized:
I0421 06:17:13.538161 132384911742784 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_26(process=7,(2,6,0,0))
I0421 06:17:13.538196 132384911742784 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_27(process=7,(3,6,0,0))
I0421 06:17:13.538221 132384911742784 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_30(process=7,(2,7,0,0))
I0421 06:17:13.538246 132384911742784 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_31(process=7,(3,7,0,0))
I0421 06:17:13.847432 132384911742784 metric_logger.py:196] completed step: 0, seconds: 11.564, TFLOP/s/device: 1.175, Tokens/s/device: 177.102, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0421 06:17:14.026748 132384911742784 metric_logger.py:196] completed step: 1, seconds: 1.511, TFLOP/s/device: 8.993, Tokens/s/device: 1355.541, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0421 06:17:14.435951 132384911742784 metric_logger.py:196] completed step: 2, seconds: 0.033, TFLOP/s/device: 417.988, Tokens/s/device: 63003.753, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0421 06:17:14.583523 132384911742784 metric_logger.py:196] completed step: 3, seconds: 0.409, TFLOP/s/device: 33.204, Tokens/s/device: 5004.924, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0421 06:17:14.878410 132384911742784 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.864, Tokens/s/device: 13394.551, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0421 06:17:14.884846 132384911742784 metric_logger.py:196] completed step: 5, seconds: 0.148, TFLOP/s/device: 91.781, Tokens/s/device: 13834.286, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0421 06:17:18.339014 2811 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0421 06:17:20.298002 132257694570240 array_metadata_store.py:203] [process=7][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_01_base/checkpoints/0/items/array_metadatas/process_7
I0421 06:17:37.268166 132384911742784 metric_logger.py:196] completed step: 6, seconds: 0.295, TFLOP/s/device: 46.026, Tokens/s/device: 6937.575, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0421 06:17:37.415788 132384911742784 metric_logger.py:196] completed step: 7, seconds: 22.237, TFLOP/s/device: 0.611, Tokens/s/device: 92.099, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0421 06:17:37.563117 132384911742784 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.286, Tokens/s/device: 13458.190, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0421 06:17:37.709802 132384911742784 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0421 06:17:37.710428 132384911742784 checkpointing.py:776] Waited 0.0006597042083740234 seconds for step 9 to finish before starting checkpointing.
I0421 06:17:37.712422 132384911742784 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 06:17:51.853609 132258411493120 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 40.916 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 38.6057653427124 s) (per-host)
I0421 06:17:51.853747 132258411493120 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 38.325103s.
I0421 06:18:01.254265 132258411493120 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 47.725621s.
I0421 06:18:01.254529 132256765028096 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 06:18:01.254646 132256765028096 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save.
I0421 06:18:01.254695 132256765028096 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0421 06:18:01.256198 132256765028096 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0421 06:18:01.256346 132384911742784 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0421 06:18:01.256430 132384911742784 checkpoint_manager.py:1452] Waiting for previous save to complete took 23.544013 seconds. If this number is high, consider checkpointing less frequently.
I0421 06:18:01.258474 132384911742784 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 9
I0421 06:18:01.261169 132384911742784 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_01_base/checkpoints/9.
I0421 06:18:01.979410 132384911742784 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 06:18:01.979502 132384911742784 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
I0421 06:18:02.022875 132384911742784 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.044544s
I0421 06:18:02.023045 132384911742784 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 32.030 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04816102981567383 s) (per-host)
I0421 06:18:02.023103 132384911742784 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.048226s (batch_requests_ready=0.001937s, total_serialization_initiated=0.046216s, others=0.000072s)
I0421 06:18:02.023225 132384911742784 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.052559s (all_items=0.000015s, per_item={'items': '0.00001478'}, temp_paths=0.052544)
I0421 06:18:02.023991 132384911742784 event_tracking.py:125] [process=7] [async] Finished blocking save in 0.77 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_01_base/checkpoints/9.
I0421 06:18:02.024300 132256765028096 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 06:38:02.024266
I0421 06:18:02.026216 132384911742784 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 06:18:02.026451 132236271679232 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save.
I0421 06:18:02.026607 132384911742784 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_01_base/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776752257.712391, 'wait_for_prev_duration_secs': 23.544013023376465, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776752281.2585125, 'checkpointer_blocking_duration_secs': 0.7658848762512207, 'get_old_steps_start_time': 1776752282.024418, 'get_old_steps_duration_secs': 2.9325485229492188e-05, 'checkpoint_manager_blocking_start_time': 1776752257.7106495, 'checkpoint_manager_blocking_duration_secs': 24.315923929214478}
I0421 06:18:02.026717 132384911742784 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0421 06:18:02.026785 132384911742784 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 06:18:07.946462 132257694570240 array_metadata_store.py:203] [process=7][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_01_base/checkpoints/9/items/array_metadatas/process_7
I0421 06:18:44.363379 132256765028096 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 37.265 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.38845682144165 s) (per-host)
I0421 06:18:44.363509 132256765028096 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.338586s.
I0421 06:18:52.778325 132256765028096 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 50.753386s.
I0421 06:18:52.778615 132236271679232 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 06:18:52.778762 132236271679232 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save.
I0421 06:18:52.778821 132236271679232 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0421 06:18:52.781949 132236271679232 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0421 06:18:52.782109 132384911742784 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0421 06:18:52.782253 132384911742784 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 06:18:52.783234 132384911742784 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.164, Tokens/s/device: 13891.904, 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: Tue Apr 21 06:19:03 UTC 2026
EXIT_CODE=0