MaxView

XPK Start: Tue Apr 21 09:06:59 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 09:07:23.896265: 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 09:07:24.435656 134368545154880 max_utils.py:273] Attempting to initialize the jax distributed system...
I0421 09:07:33.477975 134368545154880 distributed.py:149] Starting JAX distributed service on [::]:8482
I0421 09:07:33.480302 134368545154880 distributed.py:172] Connecting to JAX distributed service on mt-10-shardy-true-9vp6f-slice-job-0-0.mt-10-shardy-true-9vp6f:8482
I0421 09:07:34.655902 134368545154880 max_utils.py:284] Jax distributed system initialized!
I0421 09:07:40.655343 134368545154880 max_utils.py:800] System Information: Jax Version: 0.9.2
I0421 09:07:40.655446 134368545154880 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0421 09:07:40.655487 134368545154880 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 09:07:40.655522 134368545154880 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0421 09:07:41.349897 134368545154880 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 09:07:41.350470 134368545154880 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 09:07:41.350646 134368545154880 checkpointing.py:688] Setting up checkpoint logger...
I0421 09:07:41.350694 134368545154880 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0421 09:07:41.350751 134368545154880 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0421 09:07:41.351084 134368545154880 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 0x7a3465d48da0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0421 09:07:44.868184 134368545154880 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0421 09:07:44.868459 134368545154880 checkpoint_manager.py:708] [process=7][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a20185df650>}, handler_registry=None
I0421 09:07:44.868696 134368545154880 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a20185df650>` 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 09:07:44.868772 134368545154880 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a20185e77d0>` 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 09:07:44.868816 134368545154880 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 0x7a20185df650>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7a20185df650>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a20185e77d0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7a20185e77d0>}).
I0421 09:07:44.869128 134368545154880 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0421 09:07:44.869198 134368545154880 async_checkpointer.py:192] [process=7][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7a1f2c16bec0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0421 09:07:46.024295 134368545154880 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_10_shardy_true/checkpoints
I0421 09:07:46.466148 134368545154880 checkpoint_manager.py:929] [process=7][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_10_shardy_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7a20185e62d0>
I0421 09:07:46.466317 134368545154880 checkpointing.py:302] Checkpoint manager created!
I0421 09:07:47.391690 134368545154880 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0421 09:07:47.391809 134368545154880 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 09:07:47.776089 134368545154880 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 09:07:47.776178 134368545154880 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 09:07:47.792604 134368545154880 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 09:07:47.792660 134368545154880 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 09:07:47.816120 134368545154880 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 09:07:47.816185 134368545154880 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 09:07:47.832552 134368545154880 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 09:07:47.832620 134368545154880 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 09:07:47.848745 134368545154880 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 09:07:47.848802 134368545154880 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 09:07:47.874131 134368545154880 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0421 09:07:47.874200 134368545154880 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 09:07:47.895102 134368545154880 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0421 09:07:47.895165 134368545154880 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0421 09:07:48.108014 134368545154880 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0421 09:07:48.108118 134368545154880 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0421 09:07:49.537965 134368545154880 maxtext_utils.py:1821]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 09:07:49.538087 134368545154880 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 09:07:49.538139 134368545154880 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 09:07:49.538212 134368545154880 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 09:07:49.538266 134368545154880 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 09:07:49.538303 134368545154880 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 09:07:49.538353 134368545154880 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 09:07:49.538403 134368545154880 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 09:07:49.538441 134368545154880 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 09:07:49.538475 134368545154880 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 09:07:49.538523 134368545154880 maxtext_utils.py:1821]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0421 09:07:49.538571 134368545154880 maxtext_utils.py:1821]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0421 09:07:50.031281 134368545154880 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 09:07:50.031376 134368545154880 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 09:07:50.046763 134368545154880 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 09:07:50.046823 134368545154880 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 09:08:00.985707 134368545154880 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 09:08:00.986557 134368545154880 metric_logger.py:301] number parameters: 1.104 billion
I0421 09:08:12.263924 134368545154880 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0421 09:08:12.417622 134368545154880 checkpointing.py:798] Waited 0.15367746353149414 seconds for step 0 to finish before starting checkpointing.
I0421 09:08:12.419952 134368545154880 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 09:08:12.421663 134368545154880 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 0
I0421 09:08:12.423084 134368545154880 event_tracking.py:70] [process=7] [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_10_shardy_true/checkpoints/0.
I0421 09:08:13.181997 134368545154880 signaling_client.py:364] Using JaxDistributedSignalingClient
I0421 09:08:13.183030 134368545154880 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 09:08:13.183087 134368545154880 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 09:08:13.465008 134368545154880 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.283082s
I0421 09:08:13.465185 134368545154880 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 5.347 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2884805202484131 s) (per-host)
I0421 09:08:13.465239 134368545154880 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.288546s (batch_requests_ready=0.002207s, total_serialization_initiated=0.286264s, others=0.000075s)
I0421 09:08:13.465330 134368545154880 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.292698s (all_items=0.000018s, per_item={'items': '0.00001764'}, temp_paths=0.292681)
I0421 09:08:13.466054 134368545154880 event_tracking.py:125] [process=7] [async] Finished blocking save in 1.04 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_10_shardy_true/checkpoints/0.
I0421 09:08:13.466292 134237941991168 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 09:28:13.466255
I0421 09:08:13.469043 134368545154880 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 09:08:13.469315 134240085890816 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save.
I0421 09:08:13.469465 134368545154880 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_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776762492.419933, 'wait_for_prev_duration_secs': 6.151199340820312e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776762492.4217007, 'checkpointer_blocking_duration_secs': 1.0447402000427246, 'get_old_steps_start_time': 1776762493.4664671, 'get_old_steps_duration_secs': 2.6464462280273438e-05, 'checkpoint_manager_blocking_start_time': 1776762492.418123, 'checkpoint_manager_blocking_duration_secs': 1.0513079166412354}
I0421 09:08:13.469562 134368545154880 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0421 09:08:13.469610 134368545154880 max_utils.py:750] 
Memstats: After params initialized:
I0421 09:08:13.469658 134368545154880 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_26(process=7,(2,6,0,0))
I0421 09:08:13.469687 134368545154880 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_27(process=7,(3,6,0,0))
I0421 09:08:13.469714 134368545154880 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_30(process=7,(2,7,0,0))
I0421 09:08:13.469750 134368545154880 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_31(process=7,(3,7,0,0))
I0421 09:08:13.783108 134368545154880 metric_logger.py:196] completed step: 0, seconds: 11.277, TFLOP/s/device: 1.205, Tokens/s/device: 181.604, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0421 09:08:13.959531 134368545154880 metric_logger.py:196] completed step: 1, seconds: 1.518, TFLOP/s/device: 8.952, Tokens/s/device: 1349.402, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0421 09:08:14.355322 134368545154880 metric_logger.py:196] completed step: 2, seconds: 0.029, TFLOP/s/device: 466.607, Tokens/s/device: 70332.086, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0421 09:08:14.504812 134368545154880 metric_logger.py:196] completed step: 3, seconds: 0.396, TFLOP/s/device: 34.324, Tokens/s/device: 5173.729, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0421 09:08:14.800002 134368545154880 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.723, Tokens/s/device: 13373.297, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0421 09:08:14.805841 134368545154880 metric_logger.py:196] completed step: 5, seconds: 0.149, TFLOP/s/device: 90.943, Tokens/s/device: 13707.982, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0421 09:08:16.772116    2812 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0421 09:08:18.809771 134237950383872 array_metadata_store.py:203] [process=7][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_10_shardy_true/checkpoints/0/items/array_metadatas/process_7
I0421 09:08:37.185788 134368545154880 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.944, Tokens/s/device: 6925.236, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0421 09:08:37.333287 134368545154880 metric_logger.py:196] completed step: 7, seconds: 22.233, TFLOP/s/device: 0.611, Tokens/s/device: 92.116, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0421 09:08:37.480766 134368545154880 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.121, Tokens/s/device: 13433.296, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0421 09:08:37.627318 134368545154880 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0421 09:08:37.627982 134368545154880 checkpointing.py:798] Waited 0.0006830692291259766 seconds for step 9 to finish before starting checkpointing.
I0421 09:08:37.630065 134368545154880 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 09:08:50.085588 134237941991168 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 42.797 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.90882754325867 s) (per-host)
I0421 09:08:50.085711 134237941991168 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.619307s.
I0421 09:08:58.327273 134237941991168 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 44.860847s.
I0421 09:08:58.327583 134240085890816 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 09:08:58.327703 134240085890816 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save.
I0421 09:08:58.327781 134240085890816 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0421 09:08:58.329354 134240085890816 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0421 09:08:58.329540 134368545154880 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0421 09:08:58.329672 134368545154880 checkpoint_manager.py:1452] Waiting for previous save to complete took 20.699611 seconds. If this number is high, consider checkpointing less frequently.
I0421 09:08:58.331536 134368545154880 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 9
I0421 09:08:58.333506 134368545154880 event_tracking.py:70] [process=7] [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_10_shardy_true/checkpoints/9.
I0421 09:08:58.616256 134368545154880 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 09:08:58.616346 134368545154880 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 09:08:58.650336 134368545154880 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.035209s
I0421 09:08:58.650495 134368545154880 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 39.991 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.038573503494262695 s) (per-host)
I0421 09:08:58.650544 134368545154880 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.038632s (batch_requests_ready=0.001741s, total_serialization_initiated=0.036825s, others=0.000066s)
I0421 09:08:58.650633 134368545154880 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.042691s (all_items=0.000021s, per_item={'items': '0.00002098'}, temp_paths=0.042670)
I0421 09:08:58.651300 134368545154880 event_tracking.py:125] [process=7] [async] Finished blocking save in 0.32 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_10_shardy_true/checkpoints/9.
I0421 09:08:58.651650 134240085890816 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 09:28:58.651611
I0421 09:08:58.658619 134368545154880 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 09:08:58.658914 134239127332608 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save.
I0421 09:08:58.659078 134368545154880 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_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776762517.6300335, 'wait_for_prev_duration_secs': 20.699610710144043, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776762538.3315752, 'checkpointer_blocking_duration_secs': 0.32019519805908203, 'get_old_steps_start_time': 1776762538.6518028, 'get_old_steps_duration_secs': 3.0040740966796875e-05, 'checkpoint_manager_blocking_start_time': 1776762517.6282113, 'checkpoint_manager_blocking_duration_secs': 21.030831575393677}
I0421 09:08:58.659187 134368545154880 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0421 09:08:58.659233 134368545154880 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 09:09:04.664029 134239135725312 array_metadata_store.py:203] [process=7][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_10_shardy_true/checkpoints/9/items/array_metadatas/process_7
I0421 09:09:40.654842 134240085890816 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 37.571 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.04286980628967 s) (per-host)
I0421 09:09:40.654973 134240085890816 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.002671s.
I0421 09:09:48.400391 134240085890816 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 49.748074s.
I0421 09:09:48.400689 134239127332608 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 09:09:48.400831 134239127332608 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save.
I0421 09:09:48.400879 134239127332608 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0421 09:09:48.403182 134239127332608 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0421 09:09:48.403376 134368545154880 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0421 09:09:48.403545 134368545154880 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 09:09:48.404531 134368545154880 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.219, Tokens/s/device: 13900.201, 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 09:10:00 UTC 2026
EXIT_CODE=0