MaxView

XPK Start: Thu Apr 23 07:37:11 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. 
2026-04-23 07:37:35.671171: 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)
I0423 07:37:35.883135 137642130446144 max_utils.py:273] Attempting to initialize the jax distributed system...
I0423 07:37:44.924027 137642130446144 distributed.py:149] Starting JAX distributed service on [::]:8482
I0423 07:37:44.926364 137642130446144 distributed.py:172] Connecting to JAX distributed service on mt-07-eval-1blal-slice-job-0-0.mt-07-eval-1blal:8482
I0423 07:37:46.466779 137642130446144 max_utils.py:284] Jax distributed system initialized!
I0423 07:37:52.585108 137642130446144 max_utils.py:800] System Information: Jax Version: 0.9.2
I0423 07:37:52.585217 137642130446144 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0423 07:37:52.585260 137642130446144 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
I0423 07:37:52.585297 137642130446144 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0423 07:37:53.281212 137642130446144 maxtext_utils.py:1604] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0423 07:37:53.281491 137642130446144 checkpointing.py:677] Setting up checkpoint logger...
I0423 07:37:53.281546 137642130446144 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0423 07:37:53.281593 137642130446144 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0423 07:37:53.281958 137642130446144 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 0x7d2ebcd1ce00>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0423 07:37:56.635179 137642130446144 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0423 07:37:56.635416 137642130446144 checkpoint_manager.py:708] [process=3][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d1a286cee70>}, handler_registry=None
I0423 07:37:56.635654 137642130446144 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d1a286cee70>` 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`.
I0423 07:37:56.635710 137642130446144 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d1a286d1460>` 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`.
I0423 07:37:56.635746 137642130446144 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 0x7d1a286cee70>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d1a286cee70>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d1a286d1460>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d1a286d1460>}).
I0423 07:37:56.636081 137642130446144 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0423 07:37:56.636153 137642130446144 async_checkpointer.py:192] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d1a282398a0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0423 07:37:57.776154 137642130446144 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_07_eval/checkpoints
I0423 07:37:58.206344 137642130446144 checkpoint_manager.py:929] [process=3][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_20260423_071538/linen_xpk_main_20260423_071538_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d1a480f35f0>
I0423 07:37:58.206526 137642130446144 checkpointing.py:301] Checkpoint manager created!
I0423 07:37:59.158531 137642130446144 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0423 07:37:59.158655 137642130446144 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 07:37:59.545094 137642130446144 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch_attn', 'activation_length_attn', 'activation_embed_attn').
I0423 07:37:59.545192 137642130446144 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 07:37:59.561763 137642130446144 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch_attn', 'activation_length_attn', 'activation_embed_attn').
I0423 07:37:59.561827 137642130446144 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 07:37:59.585779 137642130446144 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 07:37:59.585851 137642130446144 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 07:37:59.602411 137642130446144 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 07:37:59.602473 137642130446144 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 07:37:59.618987 137642130446144 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 07:37:59.619050 137642130446144 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 07:37:59.643892 137642130446144 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch_attn', 'activation_length_attn', 'activation_heads', 'activation_kv').
I0423 07:37:59.643978 137642130446144 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 07:37:59.665273 137642130446144 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0423 07:37:59.665347 137642130446144 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0423 07:37:59.875787 137642130446144 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0423 07:37:59.875901 137642130446144 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0423 07:38:01.322816 137642130446144 maxtext_utils.py:1707]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0423 07:38:01.322984 137642130446144 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)
I0423 07:38:01.323047 137642130446144 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)
I0423 07:38:01.323108 137642130446144 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')
I0423 07:38:01.323163 137642130446144 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)
I0423 07:38:01.323204 137642130446144 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)
I0423 07:38:01.323258 137642130446144 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)
I0423 07:38:01.323312 137642130446144 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')
I0423 07:38:01.323353 137642130446144 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)
I0423 07:38:01.323388 137642130446144 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)
I0423 07:38:01.323436 137642130446144 maxtext_utils.py:1707]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0423 07:38:01.323490 137642130446144 maxtext_utils.py:1707]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0423 07:38:01.815015 137642130446144 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0423 07:38:01.815108 137642130446144 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0423 07:38:01.830712 137642130446144 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0423 07:38:01.830773 137642130446144 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0423 07:38:12.764209 137642130446144 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.
I0423 07:38:12.765057 137642130446144 metric_logger.py:301] number parameters: 1.104 billion
I0423 07:38:24.290251 137642130446144 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0423 07:38:24.444216 137642130446144 checkpointing.py:776] Waited 0.15394282341003418 seconds for step 0 to finish before starting checkpointing.
I0423 07:38:24.446739 137642130446144 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 07:38:24.448432 137642130446144 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 0
I0423 07:38:24.449887 137642130446144 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_07_eval/checkpoints/0.
I0423 07:38:25.222315 137642130446144 signaling_client.py:364] Using JaxDistributedSignalingClient
I0423 07:38:25.223214 137642130446144 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0423 07:38:25.223270 137642130446144 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
I0423 07:38:25.503909 137642130446144 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.281676s
I0423 07:38:25.504096 137642130446144 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 5.365 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2875196933746338 s) (per-host)
I0423 07:38:25.504150 137642130446144 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.287584s (batch_requests_ready=0.002630s, total_serialization_initiated=0.284883s, others=0.000072s)
I0423 07:38:25.504241 137642130446144 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.291787s (all_items=0.000018s, per_item={'items': '0.00001812'}, temp_paths=0.291769)
I0423 07:38:25.505016 137642130446144 event_tracking.py:125] [process=3] [async] Finished blocking save in 1.06 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_07_eval/checkpoints/0.
I0423 07:38:25.505344 137513523566336 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 07:58:25.505306
I0423 07:38:25.507267 137642130446144 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 07:38:25.507554 137511376041728 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0423 07:38:25.507681 137642130446144 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776929904.4467218, 'wait_for_prev_duration_secs': 5.984306335449219e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776929904.4484723, 'checkpointer_blocking_duration_secs': 1.05702805519104, 'get_old_steps_start_time': 1776929905.5055256, 'get_old_steps_duration_secs': 2.9087066650390625e-05, 'checkpoint_manager_blocking_start_time': 1776929904.4447458, 'checkpoint_manager_blocking_duration_secs': 1.0629007816314697}
I0423 07:38:25.507844 137642130446144 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0423 07:38:25.507895 137642130446144 max_utils.py:750] 
Memstats: After params initialized:
I0423 07:38:25.507964 137642130446144 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_10(process=3,(2,2,0,0))
I0423 07:38:25.508000 137642130446144 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_11(process=3,(3,2,0,0))
I0423 07:38:25.508027 137642130446144 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_14(process=3,(2,3,0,0))
I0423 07:38:25.508053 137642130446144 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_15(process=3,(3,3,0,0))
I0423 07:38:25.821218 137642130446144 metric_logger.py:196] completed step: 0, seconds: 11.525, TFLOP/s/device: 1.179, Tokens/s/device: 177.699, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0423 07:38:26.001144 137642130446144 metric_logger.py:196] completed step: 1, seconds: 1.530, TFLOP/s/device: 8.883, Tokens/s/device: 1338.985, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0423 07:38:26.420474 137642130446144 metric_logger.py:196] completed step: 2, seconds: 0.033, TFLOP/s/device: 416.975, Tokens/s/device: 62851.005, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0423 07:38:26.567756 137642130446144 metric_logger.py:196] completed step: 3, seconds: 0.420, TFLOP/s/device: 32.385, Tokens/s/device: 4881.397, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0423 07:38:26.862882 137642130446144 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.811, Tokens/s/device: 13386.583, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0423 07:38:26.868877 137642130446144 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.292, Tokens/s/device: 13911.248, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0423 07:38:29.296228    2803 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0423 07:38:31.963971 137511423186688 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_07_eval/checkpoints/0/items/array_metadatas/process_3
I0423 07:38:49.618314 137642130446144 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.940, Tokens/s/device: 6924.650, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0423 07:38:49.765613 137642130446144 metric_logger.py:196] completed step: 7, seconds: 22.602, TFLOP/s/device: 0.601, Tokens/s/device: 90.609, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0423 07:38:49.912856 137642130446144 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.149, Tokens/s/device: 13437.438, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0423 07:38:50.059486 137642130446144 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0423 07:38:50.060217 137642130446144 checkpointing.py:776] Waited 0.0007503032684326172 seconds for step 9 to finish before starting checkpointing.
I0423 07:38:50.063333 137642130446144 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 07:39:03.238014 137513523566336 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 41.545 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 38.02139902114868 s) (per-host)
I0423 07:39:03.238137 137513523566336 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.732682s.
I0423 07:39:10.589526 137513523566336 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 45.084052s.
I0423 07:39:10.589824 137511376041728 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 07:39:10.589965 137511376041728 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0423 07:39:10.590018 137511376041728 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0423 07:39:10.591655 137511376041728 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0423 07:39:10.591852 137642130446144 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0423 07:39:10.592000 137642130446144 checkpoint_manager.py:1452] Waiting for previous save to complete took 20.528666 seconds. If this number is high, consider checkpointing less frequently.
I0423 07:39:10.593703 137642130446144 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 9
I0423 07:39:10.595656 137642130446144 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_07_eval/checkpoints/9.
I0423 07:39:11.300869 137642130446144 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0423 07:39:11.300983 137642130446144 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
I0423 07:39:11.335996 137642130446144 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.036036s
I0423 07:39:11.336163 137642130446144 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 38.848 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.03970766067504883 s) (per-host)
I0423 07:39:11.336214 137642130446144 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.039768s (batch_requests_ready=0.001934s, total_serialization_initiated=0.037768s, others=0.000066s)
I0423 07:39:11.336305 137642130446144 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.044478s (all_items=0.000016s, per_item={'items': '0.00001574'}, temp_paths=0.044462)
I0423 07:39:11.336946 137642130446144 event_tracking.py:125] [process=3] [async] Finished blocking save in 0.74 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_07_eval/checkpoints/9.
I0423 07:39:11.337216 137490543400704 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 07:59:11.337182
I0423 07:39:11.339064 137642130446144 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 07:39:11.339346 137511376041728 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0423 07:39:11.339481 137642130446144 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776929930.0633035, 'wait_for_prev_duration_secs': 20.528665781021118, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776929950.5937479, 'checkpointer_blocking_duration_secs': 0.7436139583587646, 'get_old_steps_start_time': 1776929951.3373861, 'get_old_steps_duration_secs': 2.86102294921875e-05, 'checkpoint_manager_blocking_start_time': 1776929930.0604446, 'checkpoint_manager_blocking_duration_secs': 21.27900266647339}
I0423 07:39:11.339639 137642130446144 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0423 07:39:11.339684 137642130446144 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 07:39:17.859199 137511423186688 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_07_eval/checkpoints/9/items/array_metadatas/process_3
I0423 07:39:53.909340 137490543400704 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 37.069 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.612847089767456 s) (per-host)
I0423 07:39:53.909466 137490543400704 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.572137s.
I0423 07:40:03.227423 137490543400704 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 51.890078s.
I0423 07:40:03.227715 137511376041728 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 07:40:03.227859 137511376041728 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0423 07:40:03.227915 137511376041728 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0423 07:40:03.229553 137511376041728 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0423 07:40:03.229727 137642130446144 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0423 07:40:03.229906 137642130446144 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 07:40:03.230798 137642130446144 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.227, Tokens/s/device: 13901.522, 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: Thu Apr 23 07:40:16 UTC 2026
EXIT_CODE=0