MaxView

XPK Start: Tue Apr 21 08:37:53 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-21 08:38:17.518440: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303)
I0421 08:38:18.058357 137384520496960 max_utils.py:273] Attempting to initialize the jax distributed system...
I0421 08:38:27.100553 137384520496960 distributed.py:149] Starting JAX distributed service on [::]:8482
I0421 08:38:27.102908 137384520496960 distributed.py:172] Connecting to JAX distributed service on mt-02-synthetic-3xw1o-slice-job-0-0.mt-02-synthetic-3xw1o:8482
I0421 08:38:28.642452 137384520496960 max_utils.py:284] Jax distributed system initialized!
I0421 08:38:34.815122 137384520496960 max_utils.py:800] System Information: Jax Version: 0.9.2
I0421 08:38:34.815224 137384520496960 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0421 08:38:34.815268 137384520496960 max_utils.py:802] System Information: Jax Backend: PJRT C API
TFRT TPU v6 lite
Built on Mar 4 2026 11:32:08 (1772652728) cl/878335365
I0421 08:38:34.815302 137384520496960 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0421 08:38:35.550950 137384520496960 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 08:38:35.551535 137384520496960 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 08:38:35.551726 137384520496960 checkpointing.py:688] Setting up checkpoint logger...
I0421 08:38:35.551777 137384520496960 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0421 08:38:35.551819 137384520496960 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0421 08:38:35.552151 137384520496960 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 0x7cf29be50ec0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0421 08:38:38.237655 137384520496960 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0421 08:38:38.237833 137384520496960 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 0x7cdde4457da0>}, handler_registry=None
I0421 08:38:38.238066 137384520496960 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7cdde4457da0>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`.
I0421 08:38:38.238113 137384520496960 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7cdde4459250>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`.
I0421 08:38:38.238147 137384520496960 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 0x7cdde4457da0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7cdde4457da0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7cdde4459250>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7cdde4459250>}).
I0421 08:38:38.238470 137384520496960 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0421 08:38:38.238545 137384520496960 async_checkpointer.py:192] [process=4][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7cddb87e3ec0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0421 08:38:39.930904 137384520496960 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_02_synthetic/checkpoints
I0421 08:38:40.368377 137384520496960 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_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7cdde445b830>
I0421 08:38:40.368559 137384520496960 checkpointing.py:302] Checkpoint manager created!
I0421 08:38:41.312183 137384520496960 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0421 08:38:41.312289 137384520496960 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 08:38:41.693679 137384520496960 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 08:38:41.693770 137384520496960 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 08:38:41.710119 137384520496960 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 08:38:41.710177 137384520496960 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 08:38:41.733793 137384520496960 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 08:38:41.733860 137384520496960 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 08:38:41.750256 137384520496960 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 08:38:41.750319 137384520496960 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 08:38:41.766553 137384520496960 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 08:38:41.766613 137384520496960 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 08:38:41.791992 137384520496960 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0421 08:38:41.792059 137384520496960 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 08:38:41.812889 137384520496960 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0421 08:38:41.812952 137384520496960 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0421 08:38:42.027412 137384520496960 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0421 08:38:42.027520 137384520496960 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0421 08:38:43.439329 137384520496960 maxtext_utils.py:1821]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 08:38:43.439451 137384520496960 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)
I0421 08:38:43.439503 137384520496960 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)
I0421 08:38:43.439591 137384520496960 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')
I0421 08:38:43.439737 137384520496960 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)
I0421 08:38:43.439787 137384520496960 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)
I0421 08:38:43.439842 137384520496960 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)
I0421 08:38:43.439894 137384520496960 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')
I0421 08:38:43.439934 137384520496960 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)
I0421 08:38:43.439969 137384520496960 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)
I0421 08:38:43.440017 137384520496960 maxtext_utils.py:1821]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)
I0421 08:38:43.440061 137384520496960 maxtext_utils.py:1821]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')

I0421 08:38:43.940359 137384520496960 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 08:38:43.940455 137384520496960 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 08:38:43.955883 137384520496960 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 08:38:43.955945 137384520496960 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 08:38:54.588920 137384520496960 max_utils.py:791] Total memory size: 1.5 GB, Output size: 0.4 GB, Temp size: 1.1 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0421 08:38:54.589750 137384520496960 metric_logger.py:301] number parameters: 1.104 billion
I0421 08:39:05.998478 137384520496960 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0421 08:39:06.168505 137384520496960 checkpointing.py:798] Waited 0.1700148582458496 seconds for step 0 to finish before starting checkpointing.
I0421 08:39:06.170873 137384520496960 checkpoint_manager.py:2009] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 08:39:06.173356 137384520496960 checkpoint_manager.py:1512] [process=4] Saving checkpoint at step 0
I0421 08:39:06.174718 137384520496960 event_tracking.py:70] [process=4] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_02_synthetic/checkpoints/0.
I0421 08:39:06.516538 137384520496960 signaling_client.py:364] Using JaxDistributedSignalingClient
I0421 08:39:06.517469 137384520496960 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 08:39:06.517525 137384520496960 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False
I0421 08:39:06.794325 137384520496960 base_pytree_checkpoint_handler.py:154] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.277862s
I0421 08:39:06.794503 137384520496960 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/blocking_gbytes_per_sec: 5.443 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28342580795288086 s) (per-host)
I0421 08:39:06.794556 137384520496960 base_pytree_checkpoint_handler.py:768] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.283496s (batch_requests_ready=0.002345s, total_serialization_initiated=0.281073s, others=0.000078s)
I0421 08:39:06.794668 137384520496960 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.287622s (all_items=0.000018s, per_item={'items': '0.00001788'}, temp_paths=0.287605)
I0421 08:39:06.795518 137384520496960 event_tracking.py:125] [process=4] [async] Finished blocking save in 0.62 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_02_synthetic/checkpoints/0.
I0421 08:39:06.795879 137257149445888 async_checkpointer.py:76] [process=4][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 08:59:06.795841
I0421 08:39:06.814899 137384520496960 checkpoint_manager.py:1560] [process=4][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 08:39:06.815195 137255536551680 async_checkpointer.py:280] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0421 08:39:06.815340 137384520496960 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776760746.1708546, 'wait_for_prev_duration_secs': 6.29425048828125e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776760746.1733952, 'checkpointer_blocking_duration_secs': 0.6226463317871094, 'get_old_steps_start_time': 1776760746.796069, 'get_old_steps_duration_secs': 2.7179718017578125e-05, 'checkpoint_manager_blocking_start_time': 1776760746.169033, 'checkpoint_manager_blocking_duration_secs': 0.6462666988372803}
I0421 08:39:06.815445 137384520496960 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0421 08:39:06.815496 137384520496960 max_utils.py:750] 
Memstats: After params initialized:
I0421 08:39:06.815548 137384520496960 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_16(process=4,(0,4,0,0))
I0421 08:39:06.815586 137384520496960 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_17(process=4,(1,4,0,0))
I0421 08:39:06.815613 137384520496960 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_20(process=4,(0,5,0,0))
I0421 08:39:06.815654 137384520496960 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_21(process=4,(1,5,0,0))
I0421 08:39:07.127078 137384520496960 metric_logger.py:196] completed step: 0, seconds: 11.409, TFLOP/s/device: 1.191, Tokens/s/device: 179.513, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0421 08:39:07.307455 137384520496960 metric_logger.py:196] completed step: 1, seconds: 1.127, TFLOP/s/device: 12.056, Tokens/s/device: 1817.180, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0421 08:39:07.737265 137384520496960 metric_logger.py:196] completed step: 2, seconds: 0.033, TFLOP/s/device: 408.869, Tokens/s/device: 61629.202, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0421 08:39:07.884570 137384520496960 metric_logger.py:196] completed step: 3, seconds: 0.430, TFLOP/s/device: 31.604, Tokens/s/device: 4763.688, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0421 08:39:08.179675 137384520496960 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.726, Tokens/s/device: 13373.820, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0421 08:39:08.185708 137384520496960 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.231, Tokens/s/device: 13901.994, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0421 08:39:10.507624    2778 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0421 08:39:13.327842 137255544944384 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_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_02_synthetic/checkpoints/0/items/array_metadatas/process_4
I0421 08:39:30.520562 137384520496960 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.966, Tokens/s/device: 6928.563, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0421 08:39:30.667773 137384520496960 metric_logger.py:196] completed step: 7, seconds: 22.188, TFLOP/s/device: 0.612, Tokens/s/device: 92.302, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0421 08:39:30.815241 137384520496960 metric_logger.py:196] completed step: 8, seconds: 0.153, TFLOP/s/device: 89.078, Tokens/s/device: 13426.779, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0421 08:39:30.961836 137384520496960 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0421 08:39:30.962494 137384520496960 checkpointing.py:798] Waited 0.0006811618804931641 seconds for step 9 to finish before starting checkpointing.
I0421 08:39:30.964505 137384520496960 checkpoint_manager.py:2020] [process=4][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 08:39:44.992618 137257149445888 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/gbytes_per_sec: 41.048 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 38.48152494430542 s) (per-host)
I0421 08:39:44.992745 137257149445888 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 38.196750s.
I0421 08:39:52.939374 137257149445888 async_checkpointer.py:160] [process=4][thread=async_save] Background save thread done. Time taken: 46.143361s.
I0421 08:39:52.939682 137255536551680 async_checkpointer.py:288] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 08:39:52.939799 137255536551680 async_checkpointer.py:298] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0421 08:39:52.939848 137255536551680 checkpoint_manager.py:2137] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0421 08:39:52.941709 137255536551680 checkpoint_manager.py:2146] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0421 08:39:52.941874 137384520496960 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.
W0421 08:39:52.942006 137384520496960 checkpoint_manager.py:1452] Waiting for previous save to complete took 21.977501 seconds. If this number is high, consider checkpointing less frequently.
I0421 08:39:52.943931 137384520496960 checkpoint_manager.py:1512] [process=4] Saving checkpoint at step 9
I0421 08:39:52.945885 137384520496960 event_tracking.py:70] [process=4] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_02_synthetic/checkpoints/9.
I0421 08:39:53.659625 137384520496960 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 08:39:53.659737 137384520496960 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False
I0421 08:39:53.703681 137384520496960 base_pytree_checkpoint_handler.py:154] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.045111s
I0421 08:39:53.703848 137384520496960 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/blocking_gbytes_per_sec: 31.676 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04869818687438965 s) (per-host)
I0421 08:39:53.703899 137384520496960 base_pytree_checkpoint_handler.py:768] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.048759s (batch_requests_ready=0.001934s, total_serialization_initiated=0.046757s, others=0.000068s)
I0421 08:39:53.703986 137384520496960 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.053003s (all_items=0.000015s, per_item={'items': '0.00001526'}, temp_paths=0.052988)
I0421 08:39:53.704721 137384520496960 event_tracking.py:125] [process=4] [async] Finished blocking save in 0.76 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_02_synthetic/checkpoints/9.
I0421 08:39:53.705075 137255536551680 async_checkpointer.py:76] [process=4][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 08:59:53.705038
I0421 08:39:53.707112 137384520496960 checkpoint_manager.py:1560] [process=4][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 08:39:53.707372 137234598586112 async_checkpointer.py:280] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0421 08:39:53.707533 137384520496960 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776760770.9644752, 'wait_for_prev_duration_secs': 21.977501392364502, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776760792.94397, 'checkpointer_blocking_duration_secs': 0.7612550258636475, 'get_old_steps_start_time': 1776760793.7052512, 'get_old_steps_duration_secs': 2.9087066650390625e-05, 'checkpoint_manager_blocking_start_time': 1776760770.9627464, 'checkpoint_manager_blocking_duration_secs': 22.744752645492554}
I0421 08:39:53.707668 137384520496960 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0421 08:39:53.707717 137384520496960 checkpoint_manager.py:2020] [process=4][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 08:39:59.220779 137255544944384 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_20260421_083238/linen_xpk_feat_nnx_trainstate_and_training_loop_20260421_083238_02_synthetic/checkpoints/9/items/array_metadatas/process_4
I0421 08:40:35.374727 137255536551680 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/gbytes_per_sec: 37.862 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.7195565700531 s) (per-host)
I0421 08:40:35.374833 137255536551680 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.669646s.
I0421 08:40:45.553373 137255536551680 async_checkpointer.py:160] [process=4][thread=async_save] Background save thread done. Time taken: 51.848169s.
I0421 08:40:45.553697 137234598586112 async_checkpointer.py:288] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 08:40:45.553819 137234598586112 async_checkpointer.py:298] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0421 08:40:45.553866 137234598586112 checkpoint_manager.py:2137] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0421 08:40:45.555595 137234598586112 checkpoint_manager.py:2146] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0421 08:40:45.555791 137384520496960 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.
I0421 08:40:45.555940 137384520496960 checkpoint_manager.py:2009] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 08:40:45.556941 137384520496960 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.382, Tokens/s/device: 13924.868, total_weights: 65536, loss: 8.179, lm_loss: 8.179, perplexity: 3564.635
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Tue Apr 21 08:40:58 UTC 2026
EXIT_CODE=0