MaxView

XPK Start: Sat Apr 18 23:52:23 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-18 23:52:48.631900: 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)
I0418 23:52:48.813883 138105240442688 max_utils.py:273] Attempting to initialize the jax distributed system...
I0418 23:52:57.855117 138105240442688 distributed.py:149] Starting JAX distributed service on [::]:8482
I0418 23:52:57.857532 138105240442688 distributed.py:172] Connecting to JAX distributed service on mt-02-synthetic-b9n91-slice-job-0-0.mt-02-synthetic-b9n91:8482
I0418 23:53:00.394344 138105240442688 max_utils.py:284] Jax distributed system initialized!
I0418 23:53:06.582359 138105240442688 max_utils.py:800] System Information: Jax Version: 0.9.2
I0418 23:53:06.582457 138105240442688 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0418 23:53:06.582497 138105240442688 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
I0418 23:53:06.582532 138105240442688 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0418 23:53:07.281762 138105240442688 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0418 23:53:07.282361 138105240442688 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0418 23:53:07.282537 138105240442688 checkpointing.py:688] Setting up checkpoint logger...
I0418 23:53:07.282587 138105240442688 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0418 23:53:07.282628 138105240442688 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0418 23:53:07.282959 138105240442688 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 0x7d9a80d1cd10>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0418 23:53:11.323406 138105240442688 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0418 23:53:11.323634 138105240442688 checkpoint_manager.py:708] [process=1][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d86203bde80>}, handler_registry=None
I0418 23:53:11.323875 138105240442688 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d86203bde80>` 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`.
I0418 23:53:11.323923 138105240442688 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d8620594500>` 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`.
I0418 23:53:11.323958 138105240442688 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 0x7d86203bde80>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d86203bde80>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d8620594500>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d8620594500>}).
I0418 23:53:11.324286 138105240442688 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0418 23:53:11.324355 138105240442688 async_checkpointer.py:192] [process=1][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d853831f740> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0418 23:53:12.444922 138105240442688 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_02_synthetic/checkpoints
I0418 23:53:13.302140 138105240442688 checkpoint_manager.py:929] [process=1][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_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d86203bdb20>
I0418 23:53:13.302356 138105240442688 checkpointing.py:302] Checkpoint manager created!
I0418 23:53:14.236592 138105240442688 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0418 23:53:14.236696 138105240442688 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 23:53:14.611898 138105240442688 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 23:53:14.611989 138105240442688 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 23:53:14.628343 138105240442688 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 23:53:14.628401 138105240442688 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 23:53:14.652117 138105240442688 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 23:53:14.652194 138105240442688 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 23:53:14.668563 138105240442688 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 23:53:14.668622 138105240442688 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 23:53:14.684874 138105240442688 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 23:53:14.684933 138105240442688 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 23:53:14.709547 138105240442688 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0418 23:53:14.709614 138105240442688 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 23:53:14.730616 138105240442688 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0418 23:53:14.730682 138105240442688 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0418 23:53:14.944458 138105240442688 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0418 23:53:14.944565 138105240442688 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0418 23:53:16.367152 138105240442688 maxtext_utils.py:1827]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0418 23:53:16.367291 138105240442688 maxtext_utils.py:1827]  params/params/decoder/layers/mlp/wi_0/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0418 23:53:16.367344 138105240442688 maxtext_utils.py:1827]  params/params/decoder/layers/mlp/wi_1/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0418 23:53:16.367619 138105240442688 maxtext_utils.py:1827]  params/params/decoder/layers/mlp/wo/kernel
    Shape:     float32[7168,16,2048]
    Logical:   P('mlp', 'layers', 'embed')
    Physical:  (None, None, 'fsdp')
I0418 23:53:16.367807 138105240442688 maxtext_utils.py:1827]  params/params/decoder/layers/post_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0418 23:53:16.367863 138105240442688 maxtext_utils.py:1827]  params/params/decoder/layers/pre_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0418 23:53:16.367941 138105240442688 maxtext_utils.py:1827]  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)
I0418 23:53:16.368006 138105240442688 maxtext_utils.py:1827]  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')
I0418 23:53:16.368055 138105240442688 maxtext_utils.py:1827]  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)
I0418 23:53:16.368094 138105240442688 maxtext_utils.py:1827]  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)
I0418 23:53:16.368149 138105240442688 maxtext_utils.py:1827]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0418 23:53:16.368229 138105240442688 maxtext_utils.py:1827]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0418 23:53:16.863657 138105240442688 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0418 23:53:16.863755 138105240442688 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0418 23:53:16.879391 138105240442688 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0418 23:53:16.879450 138105240442688 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0418 23:53:27.666803 138105240442688 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.
I0418 23:53:27.667606 138105240442688 metric_logger.py:301] number parameters: 1.104 billion
I0418 23:53:39.204205 138105240442688 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0418 23:53:39.358393 138105240442688 checkpointing.py:798] Waited 0.1541905403137207 seconds for step 0 to finish before starting checkpointing.
I0418 23:53:39.360827 138105240442688 checkpoint_manager.py:2009] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 23:53:39.362683 138105240442688 checkpoint_manager.py:1512] [process=1] Saving checkpoint at step 0
I0418 23:53:39.364042 138105240442688 event_tracking.py:70] [process=1] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_02_synthetic/checkpoints/0.
I0418 23:53:40.095131 138105240442688 signaling_client.py:364] Using JaxDistributedSignalingClient
I0418 23:53:40.096168 138105240442688 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 23:53:40.096244 138105240442688 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
I0418 23:53:40.370796 138105240442688 base_pytree_checkpoint_handler.py:154] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.275742s
I0418 23:53:40.370973 138105240442688 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/blocking_gbytes_per_sec: 5.488 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2810819149017334 s) (per-host)
I0418 23:53:40.371027 138105240442688 base_pytree_checkpoint_handler.py:768] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.281147s (batch_requests_ready=0.002165s, total_serialization_initiated=0.278908s, others=0.000074s)
I0418 23:53:40.371119 138105240442688 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.285445s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.285428)
I0418 23:53:40.371981 138105240442688 event_tracking.py:125] [process=1] [async] Finished blocking save in 1.01 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_02_synthetic/checkpoints/0.
I0418 23:53:40.372328 137975351564032 async_checkpointer.py:76] [process=1][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 00:13:40.372290
I0418 23:53:40.386376 138105240442688 checkpoint_manager.py:1560] [process=1][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 23:53:40.386680 137972661872384 async_checkpointer.py:280] [process=1][thread=save_finalize] Waiting for background save thread=async_save.
I0418 23:53:40.386839 138105240442688 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776556419.3608103, 'wait_for_prev_duration_secs': 5.9604644775390625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776556419.3627212, 'checkpointer_blocking_duration_secs': 1.0097613334655762, 'get_old_steps_start_time': 1776556420.3725073, 'get_old_steps_duration_secs': 3.123283386230469e-05, 'checkpoint_manager_blocking_start_time': 1776556419.358902, 'checkpoint_manager_blocking_duration_secs': 1.0278980731964111}
I0418 23:53:40.386942 138105240442688 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0418 23:53:40.386998 138105240442688 max_utils.py:750] 
Memstats: After params initialized:
I0418 23:53:40.387048 138105240442688 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_2(process=1,(2,0,0,0))
I0418 23:53:40.387081 138105240442688 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_3(process=1,(3,0,0,0))
I0418 23:53:40.387108 138105240442688 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_6(process=1,(2,1,0,0))
I0418 23:53:40.387132 138105240442688 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_7(process=1,(3,1,0,0))
I0418 23:53:40.693113 138105240442688 metric_logger.py:196] completed step: 0, seconds: 11.536, TFLOP/s/device: 1.178, Tokens/s/device: 177.524, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0418 23:53:40.887562 138105240442688 metric_logger.py:196] completed step: 1, seconds: 1.487, TFLOP/s/device: 9.135, Tokens/s/device: 1376.914, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0418 23:53:41.287415 138105240442688 metric_logger.py:196] completed step: 2, seconds: 0.047, TFLOP/s/device: 288.860, Tokens/s/device: 43540.192, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0418 23:53:41.434641 138105240442688 metric_logger.py:196] completed step: 3, seconds: 0.400, TFLOP/s/device: 33.976, Tokens/s/device: 5121.191, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0418 23:53:41.730195 138105240442688 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.592, Tokens/s/device: 13353.590, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0418 23:53:41.735899 138105240442688 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.213, Tokens/s/device: 13899.352, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0418 23:53:46.471561    2818 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0418 23:53:49.286918 137972670265088 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_02_synthetic/checkpoints/0/items/array_metadatas/process_1
I0418 23:54:05.590207 138105240442688 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.950, Tokens/s/device: 6926.102, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0418 23:54:05.737780 138105240442688 metric_logger.py:196] completed step: 7, seconds: 23.708, TFLOP/s/device: 0.573, Tokens/s/device: 86.386, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0418 23:54:05.885210 138105240442688 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.204, Tokens/s/device: 13445.820, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0418 23:54:06.031965 138105240442688 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0418 23:54:06.032607 138105240442688 checkpointing.py:798] Waited 0.0006625652313232422 seconds for step 9 to finish before starting checkpointing.
I0418 23:54:06.034972 138105240442688 checkpoint_manager.py:2020] [process=1][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 23:54:20.418786 137975351564032 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/gbytes_per_sec: 39.168 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 40.32885432243347 s) (per-host)
I0418 23:54:20.418907 137975351564032 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 40.046466s.
I0418 23:54:31.623595 137975351564032 async_checkpointer.py:160] [process=1][thread=async_save] Background save thread done. Time taken: 51.251136s.
I0418 23:54:31.623866 137972661872384 async_checkpointer.py:288] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 23:54:31.624007 137972661872384 async_checkpointer.py:298] [process=1][thread=save_finalize] No errors found in background save thread=async_save.
I0418 23:54:31.624063 137972661872384 checkpoint_manager.py:2137] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0418 23:54:31.625836 137972661872384 checkpoint_manager.py:2146] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0418 23:54:31.626003 138105240442688 checkpoint_manager.py:2032] [process=1][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0418 23:54:31.626131 138105240442688 checkpoint_manager.py:1452] Waiting for previous save to complete took 25.591158 seconds. If this number is high, consider checkpointing less frequently.
I0418 23:54:31.628081 138105240442688 checkpoint_manager.py:1512] [process=1] Saving checkpoint at step 9
I0418 23:54:31.630129 138105240442688 event_tracking.py:70] [process=1] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_02_synthetic/checkpoints/9.
I0418 23:54:32.768154 138105240442688 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 23:54:32.768262 138105240442688 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
I0418 23:54:32.810192 138105240442688 base_pytree_checkpoint_handler.py:154] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.043166s
I0418 23:54:32.810362 138105240442688 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/blocking_gbytes_per_sec: 33.136 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04655265808105469 s) (per-host)
I0418 23:54:32.810414 138105240442688 base_pytree_checkpoint_handler.py:768] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.046614s (batch_requests_ready=0.001768s, total_serialization_initiated=0.044776s, others=0.000070s)
I0418 23:54:32.810501 138105240442688 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.050678s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=0.050663)
I0418 23:54:32.811170 138105240442688 event_tracking.py:125] [process=1] [async] Finished blocking save in 1.18 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_02_synthetic/checkpoints/9.
I0418 23:54:32.811528 137972661872384 async_checkpointer.py:76] [process=1][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 00:14:32.811491
I0418 23:54:32.813667 138105240442688 checkpoint_manager.py:1560] [process=1][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 23:54:32.813971 137975312025344 async_checkpointer.py:280] [process=1][thread=save_finalize] Waiting for background save thread=async_save.
I0418 23:54:32.814132 138105240442688 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776556446.034944, 'wait_for_prev_duration_secs': 25.591157913208008, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776556471.6281226, 'checkpointer_blocking_duration_secs': 1.183556318283081, 'get_old_steps_start_time': 1776556472.8117023, 'get_old_steps_duration_secs': 2.9087066650390625e-05, 'checkpoint_manager_blocking_start_time': 1776556446.0328662, 'checkpoint_manager_blocking_duration_secs': 26.781230211257935}
I0418 23:54:32.814262 138105240442688 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0418 23:54:32.814309 138105240442688 checkpoint_manager.py:2020] [process=1][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 23:54:40.973543 137972670265088 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_02_synthetic/checkpoints/9/items/array_metadatas/process_1
I0418 23:55:15.495543 137972661872384 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/gbytes_per_sec: 36.965 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.731693983078 s) (per-host)
I0418 23:55:15.495668 137972661872384 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.684024s.
I0418 23:55:23.593390 137972661872384 async_checkpointer.py:160] [process=1][thread=async_save] Background save thread done. Time taken: 50.781730s.
I0418 23:55:23.593678 137975312025344 async_checkpointer.py:288] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 23:55:23.593804 137975312025344 async_checkpointer.py:298] [process=1][thread=save_finalize] No errors found in background save thread=async_save.
I0418 23:55:23.593852 137975312025344 checkpoint_manager.py:2137] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0418 23:55:23.596243 137975312025344 checkpoint_manager.py:2146] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0418 23:55:23.596431 138105240442688 checkpoint_manager.py:2032] [process=1][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0418 23:55:23.596583 138105240442688 checkpoint_manager.py:2009] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 23:55:23.597564 138105240442688 metric_logger.py:196] completed step: 9, seconds: 0.148, TFLOP/s/device: 92.092, Tokens/s/device: 13881.076, 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: Sat Apr 18 23:55:33 UTC 2026
EXIT_CODE=0