MaxView

XPK Start: Sun Apr 19 21:54:35 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-19 21:55:00.025365: 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)
I0419 21:55:00.203655 133851560957760 max_utils.py:273] Attempting to initialize the jax distributed system...
I0419 21:55:09.242253 133851560957760 distributed.py:149] Starting JAX distributed service on [::]:8482
I0419 21:55:09.244718 133851560957760 distributed.py:172] Connecting to JAX distributed service on mt-02-synthetic-3435g-slice-job-0-0.mt-02-synthetic-3435g:8482
I0419 21:55:11.791782 133851560957760 max_utils.py:284] Jax distributed system initialized!
I0419 21:55:18.009413 133851560957760 max_utils.py:800] System Information: Jax Version: 0.9.2
I0419 21:55:18.009516 133851560957760 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0419 21:55:18.009557 133851560957760 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
I0419 21:55:18.009591 133851560957760 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0419 21:55:18.725381 133851560957760 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0419 21:55:18.725973 133851560957760 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0419 21:55:18.726149 133851560957760 checkpointing.py:688] Setting up checkpoint logger...
I0419 21:55:18.726199 133851560957760 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0419 21:55:18.726240 133851560957760 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0419 21:55:18.726586 133851560957760 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 0x79bc2fb1c440>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0419 21:55:22.731044 133851560957760 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0419 21:55:22.731232 133851560957760 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 0x79a79444ae40>}, handler_registry=None
I0419 21:55:22.731476 133851560957760 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79a79444ae40>` 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`.
I0419 21:55:22.731526 133851560957760 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79a794449f40>` 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`.
I0419 21:55:22.731563 133851560957760 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 0x79a79444ae40>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79a79444ae40>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79a794449f40>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79a794449f40>}).
I0419 21:55:22.731884 133851560957760 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0419 21:55:22.731953 133851560957760 async_checkpointer.py:192] [process=4][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x79a77c5a02c0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0419 21:55:25.023127 133851560957760 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525/linen_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525_02_synthetic/checkpoints
I0419 21:55:25.025363 133851560957760 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_feat_nnx_trainstate_and_training_loop_20260419_214525/linen_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x79a79444a120>
I0419 21:55:25.025471 133851560957760 checkpointing.py:302] Checkpoint manager created!
I0419 21:55:25.985000 133851560957760 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0419 21:55:25.985099 133851560957760 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 21:55:26.365862 133851560957760 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0419 21:55:26.365954 133851560957760 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 21:55:26.382078 133851560957760 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0419 21:55:26.382149 133851560957760 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 21:55:26.405936 133851560957760 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0419 21:55:26.406000 133851560957760 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 21:55:26.422096 133851560957760 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0419 21:55:26.422152 133851560957760 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 21:55:26.438301 133851560957760 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0419 21:55:26.438374 133851560957760 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 21:55:26.462887 133851560957760 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0419 21:55:26.462957 133851560957760 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 21:55:26.484244 133851560957760 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0419 21:55:26.484309 133851560957760 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0419 21:55:26.704752 133851560957760 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0419 21:55:26.704865 133851560957760 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0419 21:55:28.129768 133851560957760 maxtext_utils.py:1821]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0419 21:55:28.129893 133851560957760 maxtext_utils.py:1821]  params/params/decoder/layers/mlp/wi_0/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0419 21:55:28.129946 133851560957760 maxtext_utils.py:1821]  params/params/decoder/layers/mlp/wi_1/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0419 21:55:28.130025 133851560957760 maxtext_utils.py:1821]  params/params/decoder/layers/mlp/wo/kernel
    Shape:     float32[7168,16,2048]
    Logical:   P('mlp', 'layers', 'embed')
    Physical:  (None, None, 'fsdp')
I0419 21:55:28.130082 133851560957760 maxtext_utils.py:1821]  params/params/decoder/layers/post_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0419 21:55:28.130121 133851560957760 maxtext_utils.py:1821]  params/params/decoder/layers/pre_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0419 21:55:28.130174 133851560957760 maxtext_utils.py:1821]  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)
I0419 21:55:28.130250 133851560957760 maxtext_utils.py:1821]  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')
I0419 21:55:28.130308 133851560957760 maxtext_utils.py:1821]  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)
I0419 21:55:28.130358 133851560957760 maxtext_utils.py:1821]  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)
I0419 21:55:28.130412 133851560957760 maxtext_utils.py:1821]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0419 21:55:28.130461 133851560957760 maxtext_utils.py:1821]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0419 21:55:28.619460 133851560957760 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0419 21:55:28.619553 133851560957760 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0419 21:55:28.634883 133851560957760 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0419 21:55:28.634951 133851560957760 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0419 21:55:39.719828 133851560957760 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.
I0419 21:55:39.720678 133851560957760 metric_logger.py:301] number parameters: 1.104 billion
I0419 21:55:51.120846 133851560957760 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0419 21:55:51.274641 133851560957760 checkpointing.py:798] Waited 0.15378189086914062 seconds for step 0 to finish before starting checkpointing.
I0419 21:55:51.277129 133851560957760 checkpoint_manager.py:2009] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0419 21:55:51.278763 133851560957760 checkpoint_manager.py:1512] [process=4] Saving checkpoint at step 0
I0419 21:55:51.280163 133851560957760 event_tracking.py:70] [process=4] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525/linen_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525_02_synthetic/checkpoints/0.
I0419 21:55:52.429149 133851560957760 signaling_client.py:364] Using JaxDistributedSignalingClient
I0419 21:55:52.430104 133851560957760 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0419 21:55:52.430162 133851560957760 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
I0419 21:55:52.701761 133851560957760 base_pytree_checkpoint_handler.py:154] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.272685s
I0419 21:55:52.701945 133851560957760 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/blocking_gbytes_per_sec: 5.548 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.27805042266845703 s) (per-host)
I0419 21:55:52.701997 133851560957760 base_pytree_checkpoint_handler.py:768] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.278113s (batch_requests_ready=0.002158s, total_serialization_initiated=0.275885s, others=0.000070s)
I0419 21:55:52.702089 133851560957760 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.282245s (all_items=0.000017s, per_item={'items': '0.00001693'}, temp_paths=0.282228)
I0419 21:55:52.702902 133851560957760 event_tracking.py:125] [process=4] [async] Finished blocking save in 1.42 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525/linen_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525_02_synthetic/checkpoints/0.
I0419 21:55:52.703211 133722212820736 async_checkpointer.py:76] [process=4][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 22:15:52.703174
I0419 21:55:52.716385 133851560957760 checkpoint_manager.py:1560] [process=4][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0419 21:55:52.716717 133720594585344 async_checkpointer.py:280] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0419 21:55:52.716879 133851560957760 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525/linen_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776635751.277111, 'wait_for_prev_duration_secs': 6.318092346191406e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776635751.2788033, 'checkpointer_blocking_duration_secs': 1.42458176612854, 'get_old_steps_start_time': 1776635752.7034106, 'get_old_steps_duration_secs': 3.0279159545898438e-05, 'checkpoint_manager_blocking_start_time': 1776635751.2750933, 'checkpoint_manager_blocking_duration_secs': 1.4417459964752197}
I0419 21:55:52.716983 133851560957760 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0419 21:55:52.717031 133851560957760 max_utils.py:750] 
Memstats: After params initialized:
I0419 21:55:52.717080 133851560957760 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_16(process=4,(0,4,0,0))
I0419 21:55:52.717112 133851560957760 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_17(process=4,(1,4,0,0))
I0419 21:55:52.717139 133851560957760 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_20(process=4,(0,5,0,0))
I0419 21:55:52.717164 133851560957760 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_21(process=4,(1,5,0,0))
I0419 21:55:53.022276 133851560957760 metric_logger.py:196] completed step: 0, seconds: 11.400, TFLOP/s/device: 1.192, Tokens/s/device: 179.648, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0419 21:55:53.222239 133851560957760 metric_logger.py:196] completed step: 1, seconds: 1.900, TFLOP/s/device: 7.151, Tokens/s/device: 1077.922, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0419 21:55:53.644811 133851560957760 metric_logger.py:196] completed step: 2, seconds: 0.053, TFLOP/s/device: 258.453, Tokens/s/device: 38956.839, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0419 21:55:53.792209 133851560957760 metric_logger.py:196] completed step: 3, seconds: 0.423, TFLOP/s/device: 32.146, Tokens/s/device: 4845.365, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0419 21:55:54.087461 133851560957760 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.791, Tokens/s/device: 13383.522, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0419 21:55:54.093722 133851560957760 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.250, Tokens/s/device: 13904.920, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0419 21:55:57.783259    2838 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0419 21:56:01.879247 133721714624256 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525/linen_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525_02_synthetic/checkpoints/0/items/array_metadatas/process_4
I0419 21:56:21.860599 133851560957760 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.857, Tokens/s/device: 6912.007, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0419 21:56:22.008147 133851560957760 metric_logger.py:196] completed step: 7, seconds: 27.620, TFLOP/s/device: 0.492, Tokens/s/device: 74.149, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0419 21:56:22.155493 133851560957760 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.188, Tokens/s/device: 13443.348, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0419 21:56:22.302242 133851560957760 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0419 21:56:22.302828 133851560957760 checkpointing.py:798] Waited 0.0006084442138671875 seconds for step 9 to finish before starting checkpointing.
I0419 21:56:22.304963 133851560957760 checkpoint_manager.py:2020] [process=4][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0419 21:56:31.088802 133722212820736 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/gbytes_per_sec: 40.854 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 38.664873123168945 s) (per-host)
I0419 21:56:31.088909 133722212820736 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 38.385584s.
I0419 21:56:43.847727 133722212820736 async_checkpointer.py:160] [process=4][thread=async_save] Background save thread done. Time taken: 51.144384s.
I0419 21:56:43.847990 133720594585344 async_checkpointer.py:288] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0419 21:56:43.848116 133720594585344 async_checkpointer.py:298] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0419 21:56:43.848167 133720594585344 checkpoint_manager.py:2137] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0419 21:56:43.850917 133720594585344 checkpoint_manager.py:2146] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0419 21:56:43.851059 133851560957760 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.
W0419 21:56:43.851145 133851560957760 checkpoint_manager.py:1452] Waiting for previous save to complete took 21.546183 seconds. If this number is high, consider checkpointing less frequently.
I0419 21:56:43.853056 133851560957760 checkpoint_manager.py:1512] [process=4] Saving checkpoint at step 9
I0419 21:56:43.855093 133851560957760 event_tracking.py:70] [process=4] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525/linen_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525_02_synthetic/checkpoints/9.
I0419 21:56:44.586125 133851560957760 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0419 21:56:44.586221 133851560957760 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
I0419 21:56:44.619973 133851560957760 base_pytree_checkpoint_handler.py:154] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.034864s
I0419 21:56:44.620106 133851560957760 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/blocking_gbytes_per_sec: 40.272 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.03830432891845703 s) (per-host)
I0419 21:56:44.620152 133851560957760 base_pytree_checkpoint_handler.py:768] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.038359s (batch_requests_ready=0.001772s, total_serialization_initiated=0.036526s, others=0.000062s)
I0419 21:56:44.620231 133851560957760 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.042540s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.042525)
I0419 21:56:44.620898 133851560957760 event_tracking.py:125] [process=4] [async] Finished blocking save in 0.77 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525/linen_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525_02_synthetic/checkpoints/9.
I0419 21:56:44.621227 133720594585344 async_checkpointer.py:76] [process=4][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 22:16:44.621190
I0419 21:56:44.629439 133851560957760 checkpoint_manager.py:1560] [process=4][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0419 21:56:44.629750 133700738070272 async_checkpointer.py:280] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0419 21:56:44.629886 133851560957760 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525/linen_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776635782.3049362, 'wait_for_prev_duration_secs': 21.546182870864868, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776635803.853095, 'checkpointer_blocking_duration_secs': 0.7683017253875732, 'get_old_steps_start_time': 1776635804.621421, 'get_old_steps_duration_secs': 2.956390380859375e-05, 'checkpoint_manager_blocking_start_time': 1776635782.3030517, 'checkpoint_manager_blocking_duration_secs': 22.326798915863037}
I0419 21:56:44.629999 133851560957760 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0419 21:56:44.630044 133851560957760 checkpoint_manager.py:2020] [process=4][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0419 21:56:54.383736 133721714624256 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525/linen_xpk_feat_nnx_trainstate_and_training_loop_20260419_214525_02_synthetic/checkpoints/9/items/array_metadatas/process_4
I0419 21:57:29.319290 133720594585344 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/gbytes_per_sec: 35.308 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 44.73744559288025 s) (per-host)
I0419 21:57:29.319429 133720594585344 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 44.698089s.
I0419 21:57:39.893374 133720594585344 async_checkpointer.py:160] [process=4][thread=async_save] Background save thread done. Time taken: 55.271999s.
I0419 21:57:39.893607 133700738070272 async_checkpointer.py:288] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0419 21:57:39.893678 133700738070272 async_checkpointer.py:298] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0419 21:57:39.893721 133700738070272 checkpoint_manager.py:2137] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0419 21:57:39.895934 133700738070272 checkpoint_manager.py:2146] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0419 21:57:39.896111 133851560957760 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.
I0419 21:57:39.896266 133851560957760 checkpoint_manager.py:2009] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0419 21:57:39.897351 133851560957760 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.183, Tokens/s/device: 13894.825, 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: Sun Apr 19 21:57:51 UTC 2026
EXIT_CODE=0