XPK Start: Fri Apr 24 07:06:10 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 07:06:35.277905: 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 07:06:35.488466 137553131562816 max_utils.py:273] Attempting to initialize the jax distributed system...
I0424 07:06:44.530469 137553131562816 distributed.py:149] Starting JAX distributed service on [::]:8482
I0424 07:06:44.533037 137553131562816 distributed.py:172] Connecting to JAX distributed service on mt-02-synthetic-0yhrp-slice-job-0-0.mt-02-synthetic-0yhrp:8482
I0424 07:06:46.362699 137553131562816 max_utils.py:284] Jax distributed system initialized!
I0424 07:06:52.740434 137553131562816 max_utils.py:800] System Information: Jax Version: 0.9.2
I0424 07:06:52.740541 137553131562816 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0424 07:06:52.740585 137553131562816 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 07:06:52.740621 137553131562816 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0424 07:06:53.437979 137553131562816 maxtext_utils.py:1604] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0424 07:06:53.438253 137553131562816 checkpointing.py:677] Setting up checkpoint logger...
I0424 07:06:53.438306 137553131562816 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0424 07:06:53.438349 137553131562816 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0424 07:06:53.438678 137553131562816 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 0x7d1a06d066c0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0424 07:06:56.368603 137553131562816 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0424 07:06:56.368838 137553131562816 checkpoint_manager.py:708] [process=4][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d05883b1610>}, handler_registry=None
I0424 07:06:56.369086 137553131562816 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d05883b1610>` 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 07:06:56.369134 137553131562816 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d04ac718d10>` 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 07:06:56.369170 137553131562816 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 0x7d05883b1610>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d05883b1610>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d04ac718d10>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d04ac718d10>}).
I0424 07:06:56.369489 137553131562816 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0424 07:06:56.369565 137553131562816 async_checkpointer.py:192] [process=4][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d04ac389a80> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0424 07:06:57.364063 137553131562816 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_02_synthetic/checkpoints
I0424 07:06:57.403423 137553131562816 checkpoint_manager.py:929] [process=4][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_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d0588165520>
I0424 07:06:57.403560 137553131562816 checkpointing.py:301] Checkpoint manager created!
I0424 07:06:58.327107 137553131562816 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0424 07:06:58.327219 137553131562816 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 07:06:58.708178 137553131562816 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0424 07:06:58.708273 137553131562816 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 07:06:58.724666 137553131562816 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0424 07:06:58.724724 137553131562816 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 07:06:58.748585 137553131562816 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 07:06:58.748656 137553131562816 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 07:06:58.765085 137553131562816 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 07:06:58.765150 137553131562816 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 07:06:58.781593 137553131562816 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 07:06:58.781652 137553131562816 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 07:06:58.806613 137553131562816 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0424 07:06:58.806678 137553131562816 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 07:06:58.827754 137553131562816 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0424 07:06:58.827820 137553131562816 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0424 07:06:59.036724 137553131562816 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0424 07:06:59.036828 137553131562816 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0424 07:07:00.462896 137553131562816 maxtext_utils.py:1707] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0424 07:07:00.463028 137553131562816 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 07:07:00.463080 137553131562816 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 07:07:00.463135 137553131562816 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 07:07:00.463185 137553131562816 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 07:07:00.463221 137553131562816 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 07:07:00.463280 137553131562816 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 07:07:00.463332 137553131562816 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 07:07:00.463371 137553131562816 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 07:07:00.463407 137553131562816 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 07:07:00.463453 137553131562816 maxtext_utils.py:1707] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0424 07:07:00.463504 137553131562816 maxtext_utils.py:1707] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0424 07:07:00.954437 137553131562816 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0424 07:07:00.954535 137553131562816 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0424 07:07:00.970328 137553131562816 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0424 07:07:00.970390 137553131562816 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0424 07:07:12.065423 137553131562816 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 07:07:12.066236 137553131562816 metric_logger.py:301] number parameters: 1.104 billion
I0424 07:07:23.859223 137553131562816 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0424 07:07:24.011552 137553131562816 checkpointing.py:776] Waited 0.15230369567871094 seconds for step 0 to finish before starting checkpointing.
I0424 07:07:24.014088 137553131562816 checkpoint_manager.py:2009] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0424 07:07:24.015724 137553131562816 checkpoint_manager.py:1512] [process=4] Saving checkpoint at step 0
I0424 07:07:24.017144 137553131562816 event_tracking.py:70] [process=4] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_02_synthetic/checkpoints/0.
I0424 07:07:24.337249 137553131562816 signaling_client.py:364] Using JaxDistributedSignalingClient
I0424 07:07:24.338145 137553131562816 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0424 07:07:24.338202 137553131562816 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 07:07:24.614850 137553131562816 base_pytree_checkpoint_handler.py:154] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.277673s
I0424 07:07:24.615041 137553131562816 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/blocking_gbytes_per_sec: 5.446 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2832348346710205 s) (per-host)
I0424 07:07:24.615102 137553131562816 base_pytree_checkpoint_handler.py:768] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.283306s (batch_requests_ready=0.002261s, total_serialization_initiated=0.280966s, others=0.000079s)
I0424 07:07:24.615199 137553131562816 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.287350s (all_items=0.000019s, per_item={'items': '0.00001860'}, temp_paths=0.287332)
I0424 07:07:24.616103 137553131562816 event_tracking.py:125] [process=4] [async] Finished blocking save in 0.60 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_02_synthetic/checkpoints/0.
I0424 07:07:24.616398 137423310993152 async_checkpointer.py:76] [process=4][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 07:27:24.616364
I0424 07:07:24.625404 137553131562816 checkpoint_manager.py:1560] [process=4][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 07:07:24.625674 137421123786496 async_checkpointer.py:280] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0424 07:07:24.625800 137553131562816 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777014444.0140674, 'wait_for_prev_duration_secs': 7.295608520507812e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777014444.0157673, 'checkpointer_blocking_duration_secs': 0.6008036136627197, 'get_old_steps_start_time': 1777014444.6165953, 'get_old_steps_duration_secs': 2.9802322387695312e-05, 'checkpoint_manager_blocking_start_time': 1777014444.0120878, 'checkpoint_manager_blocking_duration_secs': 0.6136713027954102}
I0424 07:07:24.625919 137553131562816 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0424 07:07:24.625969 137553131562816 max_utils.py:750]
Memstats: After params initialized:
I0424 07:07:24.626020 137553131562816 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_16(process=4,(0,4,0,0))
I0424 07:07:24.626053 137553131562816 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_17(process=4,(1,4,0,0))
I0424 07:07:24.626081 137553131562816 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_20(process=4,(0,5,0,0))
I0424 07:07:24.626107 137553131562816 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_21(process=4,(1,5,0,0))
I0424 07:07:24.940943 137553131562816 metric_logger.py:196] completed step: 0, seconds: 11.793, TFLOP/s/device: 1.152, Tokens/s/device: 173.664, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0424 07:07:25.110624 137553131562816 metric_logger.py:196] completed step: 1, seconds: 1.080, TFLOP/s/device: 12.580, Tokens/s/device: 1896.186, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0424 07:07:25.571206 137553131562816 metric_logger.py:196] completed step: 2, seconds: 0.023, TFLOP/s/device: 601.840, Tokens/s/device: 90715.804, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0424 07:07:25.718741 137553131562816 metric_logger.py:196] completed step: 3, seconds: 0.461, TFLOP/s/device: 29.490, Tokens/s/device: 4445.120, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0424 07:07:26.013949 137553131562816 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.750, Tokens/s/device: 13377.315, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0424 07:07:26.019676 137553131562816 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.145, Tokens/s/device: 13889.171, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0424 07:07:28.185508 2844 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0424 07:07:30.200211 137421169620736 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_02_synthetic/checkpoints/0/items/array_metadatas/process_4
I0424 07:07:47.554302 137553131562816 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.963, Tokens/s/device: 6927.977, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0424 07:07:47.701740 137553131562816 metric_logger.py:196] completed step: 7, seconds: 21.388, TFLOP/s/device: 0.635, Tokens/s/device: 95.755, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0424 07:07:47.849061 137553131562816 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.134, Tokens/s/device: 13435.235, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0424 07:07:47.995807 137553131562816 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0424 07:07:47.996528 137553131562816 checkpointing.py:776] Waited 0.0007421970367431641 seconds for step 9 to finish before starting checkpointing.
I0424 07:07:47.998631 137553131562816 checkpoint_manager.py:2020] [process=4][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 07:08:01.804610 137423310993152 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/gbytes_per_sec: 42.153 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.472769260406494 s) (per-host)
I0424 07:08:01.804717 137423310993152 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.188206s.
I0424 07:08:11.149462 137423310993152 async_checkpointer.py:160] [process=4][thread=async_save] Background save thread done. Time taken: 46.532933s.
I0424 07:08:11.149712 137421123786496 async_checkpointer.py:288] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 07:08:11.149771 137421123786496 async_checkpointer.py:298] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0424 07:08:11.149817 137421123786496 checkpoint_manager.py:2137] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0424 07:08:11.152169 137421123786496 checkpoint_manager.py:2146] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0424 07:08:11.152364 137553131562816 checkpoint_manager.py:2032] [process=4][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0424 07:08:11.152499 137553131562816 checkpoint_manager.py:1452] Waiting for previous save to complete took 23.153870 seconds. If this number is high, consider checkpointing less frequently.
I0424 07:08:11.154422 137553131562816 checkpoint_manager.py:1512] [process=4] Saving checkpoint at step 9
I0424 07:08:11.156492 137553131562816 event_tracking.py:70] [process=4] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_02_synthetic/checkpoints/9.
I0424 07:08:11.455297 137553131562816 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0424 07:08:11.455400 137553131562816 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 07:08:11.491614 137553131562816 base_pytree_checkpoint_handler.py:154] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.037437s
I0424 07:08:11.491778 137553131562816 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/blocking_gbytes_per_sec: 37.656 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04096531867980957 s) (per-host)
I0424 07:08:11.491827 137553131562816 base_pytree_checkpoint_handler.py:768] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.041025s (batch_requests_ready=0.001827s, total_serialization_initiated=0.039131s, others=0.000067s)
I0424 07:08:11.491934 137553131562816 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.045086s (all_items=0.000014s, per_item={'items': '0.00001407'}, temp_paths=0.045072)
I0424 07:08:11.492634 137553131562816 event_tracking.py:125] [process=4] [async] Finished blocking save in 0.34 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_02_synthetic/checkpoints/9.
I0424 07:08:11.492959 137421123786496 async_checkpointer.py:76] [process=4][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 07:28:11.492925
I0424 07:08:11.494885 137553131562816 checkpoint_manager.py:1560] [process=4][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 07:08:11.495175 137399417956096 async_checkpointer.py:280] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0424 07:08:11.495317 137553131562816 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777014467.9986, 'wait_for_prev_duration_secs': 23.15386962890625, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777014491.154461, 'checkpointer_blocking_duration_secs': 0.3386404514312744, 'get_old_steps_start_time': 1777014491.4931264, 'get_old_steps_duration_secs': 3.123283386230469e-05, 'checkpoint_manager_blocking_start_time': 1777014467.996763, 'checkpoint_manager_blocking_duration_secs': 23.498518705368042}
I0424 07:08:11.495483 137553131562816 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0424 07:08:11.495531 137553131562816 checkpoint_manager.py:2020] [process=4][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 07:08:16.626771 137421169620736 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260424_070227/linen_xpk_main_20260424_070227_02_synthetic/checkpoints/9/items/array_metadatas/process_4
I0424 07:08:53.273910 137421123786496 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/gbytes_per_sec: 37.769 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.82304573059082 s) (per-host)
I0424 07:08:53.274022 137421123786496 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.780948s.
I0424 07:09:00.816888 137421123786496 async_checkpointer.py:160] [process=4][thread=async_save] Background save thread done. Time taken: 49.323776s.
I0424 07:09:00.817169 137399417956096 async_checkpointer.py:288] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 07:09:00.817284 137399417956096 async_checkpointer.py:298] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0424 07:09:00.817332 137399417956096 checkpoint_manager.py:2137] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0424 07:09:00.819599 137399417956096 checkpoint_manager.py:2146] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0424 07:09:00.819781 137553131562816 checkpoint_manager.py:2032] [process=4][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0424 07:09:00.819947 137553131562816 checkpoint_manager.py:2009] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0424 07:09:00.820857 137553131562816 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.330, Tokens/s/device: 13917.014, 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 07:09:10 UTC 2026
EXIT_CODE=0