MaxView

XPK Start: Sun Apr 19 00:19:32 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 00:19:56.769573: 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 00:19:56.946577 135558986467136 max_utils.py:273] Attempting to initialize the jax distributed system...
I0419 00:20:05.988568 135558986467136 distributed.py:149] Starting JAX distributed service on [::]:8482
I0419 00:20:05.990951 135558986467136 distributed.py:172] Connecting to JAX distributed service on mt-10-shardy-true-ax637-slice-job-0-0.mt-10-shardy-true-ax637:8482
I0419 00:20:07.138900 135558986467136 max_utils.py:284] Jax distributed system initialized!
I0419 00:20:13.196182 135558986467136 max_utils.py:800] System Information: Jax Version: 0.9.2
I0419 00:20:13.196304 135558986467136 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0419 00:20:13.196345 135558986467136 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 00:20:13.196378 135558986467136 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0419 00:20:13.888772 135558986467136 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0419 00:20:13.889365 135558986467136 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0419 00:20:13.889541 135558986467136 checkpointing.py:688] Setting up checkpoint logger...
I0419 00:20:13.889593 135558986467136 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0419 00:20:13.889636 135558986467136 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0419 00:20:13.889962 135558986467136 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 0x7b4991938170>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0419 00:20:18.627497 135558986467136 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0419 00:20:18.627734 135558986467136 checkpoint_manager.py:708] [process=3][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b363c26a090>}, handler_registry=None
I0419 00:20:18.627972 135558986467136 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b363c26a090>` 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 00:20:18.628019 135558986467136 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b363c26dd60>` 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 00:20:18.628056 135558986467136 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 0x7b363c26a090>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b363c26a090>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b363c26dd60>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b363c26dd60>}).
I0419 00:20:18.628397 135558986467136 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0419 00:20:18.628469 135558986467136 async_checkpointer.py:192] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7b361c397740> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0419 00:20:20.159599 135558986467136 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_10_shardy_true/checkpoints
I0419 00:20:20.608865 135558986467136 checkpoint_manager.py:929] [process=3][thread=MainThread] CheckpointManager created,  primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_10_shardy_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7b363c268e90>
I0419 00:20:20.609041 135558986467136 checkpointing.py:302] Checkpoint manager created!
I0419 00:20:21.533473 135558986467136 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0419 00:20:21.533576 135558986467136 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 00:20:21.910540 135558986467136 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0419 00:20:21.910630 135558986467136 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 00:20:21.926824 135558986467136 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0419 00:20:21.926884 135558986467136 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 00:20:21.950483 135558986467136 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 00:20:21.950550 135558986467136 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 00:20:21.966835 135558986467136 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 00:20:21.966895 135558986467136 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 00:20:21.983046 135558986467136 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 00:20:21.983111 135558986467136 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 00:20:22.007558 135558986467136 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0419 00:20:22.007627 135558986467136 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 00:20:22.028389 135558986467136 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0419 00:20:22.028452 135558986467136 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0419 00:20:22.241983 135558986467136 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0419 00:20:22.242092 135558986467136 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0419 00:20:23.655314 135558986467136 maxtext_utils.py:1827]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0419 00:20:23.655438 135558986467136 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)
I0419 00:20:23.655492 135558986467136 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)
I0419 00:20:23.655569 135558986467136 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')
I0419 00:20:23.655625 135558986467136 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)
I0419 00:20:23.655663 135558986467136 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)
I0419 00:20:23.655714 135558986467136 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)
I0419 00:20:23.655765 135558986467136 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')
I0419 00:20:23.655804 135558986467136 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)
I0419 00:20:23.655839 135558986467136 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)
I0419 00:20:23.655886 135558986467136 maxtext_utils.py:1827]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0419 00:20:23.655933 135558986467136 maxtext_utils.py:1827]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0419 00:20:24.143797 135558986467136 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0419 00:20:24.143890 135558986467136 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0419 00:20:24.159281 135558986467136 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0419 00:20:24.159343 135558986467136 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0419 00:20:35.090813 135558986467136 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 00:20:35.091682 135558986467136 metric_logger.py:301] number parameters: 1.104 billion
I0419 00:20:46.498598 135558986467136 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0419 00:20:46.652006 135558986467136 checkpointing.py:798] Waited 0.15338850021362305 seconds for step 0 to finish before starting checkpointing.
I0419 00:20:46.654425 135558986467136 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0419 00:20:46.656153 135558986467136 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 0
I0419 00:20:46.657537 135558986467136 event_tracking.py:70] [process=3] [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_10_shardy_true/checkpoints/0.
I0419 00:20:47.378750 135558986467136 signaling_client.py:364] Using JaxDistributedSignalingClient
I0419 00:20:47.379728 135558986467136 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0419 00:20:47.379783 135558986467136 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 00:20:47.652762 135558986467136 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.274090s
I0419 00:20:47.652942 135558986467136 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 5.518 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.27953410148620605 s) (per-host)
I0419 00:20:47.652999 135558986467136 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.279601s (batch_requests_ready=0.002168s, total_serialization_initiated=0.277357s, others=0.000075s)
I0419 00:20:47.653094 135558986467136 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.284142s (all_items=0.000018s, per_item={'items': '0.00001836'}, temp_paths=0.284124)
I0419 00:20:47.653915 135558986467136 event_tracking.py:125] [process=3] [async] Finished blocking save in 1.00 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_10_shardy_true/checkpoints/0.
I0419 00:20:47.654204 135434335426304 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 00:40:47.654172
I0419 00:20:47.664110 135558986467136 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0419 00:20:47.664376 135432155928320 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0419 00:20:47.664497 135558986467136 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_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776558046.654407, 'wait_for_prev_duration_secs': 6.246566772460938e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776558046.656193, 'checkpointer_blocking_duration_secs': 0.9981780052185059, 'get_old_steps_start_time': 1776558047.6543977, 'get_old_steps_duration_secs': 3.0040740966796875e-05, 'checkpoint_manager_blocking_start_time': 1776558046.6525316, 'checkpoint_manager_blocking_duration_secs': 1.0119285583496094}
I0419 00:20:47.664601 135558986467136 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0419 00:20:47.664650 135558986467136 max_utils.py:750] 
Memstats: After params initialized:
I0419 00:20:47.664698 135558986467136 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_10(process=3,(2,2,0,0))
I0419 00:20:47.664731 135558986467136 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_11(process=3,(3,2,0,0))
I0419 00:20:47.664758 135558986467136 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_14(process=3,(2,3,0,0))
I0419 00:20:47.664782 135558986467136 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_15(process=3,(3,3,0,0))
I0419 00:20:47.974401 135558986467136 metric_logger.py:196] completed step: 0, seconds: 11.407, TFLOP/s/device: 1.191, Tokens/s/device: 179.542, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0419 00:20:48.261482 135558986467136 metric_logger.py:196] completed step: 1, seconds: 1.474, TFLOP/s/device: 9.216, Tokens/s/device: 1389.090, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0419 00:20:49.071264 135558986467136 metric_logger.py:196] completed step: 2, seconds: 0.140, TFLOP/s/device: 97.309, Tokens/s/device: 14667.440, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0419 00:20:49.218612 135558986467136 metric_logger.py:196] completed step: 3, seconds: 0.810, TFLOP/s/device: 16.777, Tokens/s/device: 2528.763, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0419 00:20:49.513872 135558986467136 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.848, Tokens/s/device: 13392.186, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0419 00:20:49.519451 135558986467136 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.301, Tokens/s/device: 13912.571, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0419 00:20:51.586947    2781 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0419 00:20:55.045039 135432164321024 array_metadata_store.py:203] [process=3][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_10_shardy_true/checkpoints/0/items/array_metadatas/process_3
I0419 00:21:13.118173 135558986467136 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.925, Tokens/s/device: 6922.380, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0419 00:21:13.265677 135558986467136 metric_logger.py:196] completed step: 7, seconds: 23.452, TFLOP/s/device: 0.579, Tokens/s/device: 87.328, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0419 00:21:13.413078 135558986467136 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.363, Tokens/s/device: 13469.785, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0419 00:21:13.559981 135558986467136 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0419 00:21:13.560644 135558986467136 checkpointing.py:798] Waited 0.0006835460662841797 seconds for step 9 to finish before starting checkpointing.
I0419 00:21:13.562762 135558986467136 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0419 00:21:25.202659 135434335426304 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 41.756 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.82921266555786 s) (per-host)
I0419 00:21:25.202786 135434335426304 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.548450s.
I0419 00:21:41.832761 135434335426304 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 54.178409s.
I0419 00:21:41.833005 135432155928320 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0419 00:21:41.833071 135432155928320 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0419 00:21:41.833123 135432155928320 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0419 00:21:41.835930 135432155928320 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0419 00:21:41.836107 135558986467136 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0419 00:21:41.836285 135558986467136 checkpoint_manager.py:1452] Waiting for previous save to complete took 28.273483 seconds. If this number is high, consider checkpointing less frequently.
I0419 00:21:41.838067 135558986467136 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 9
I0419 00:21:41.840120 135558986467136 event_tracking.py:70] [process=3] [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_10_shardy_true/checkpoints/9.
I0419 00:21:42.596640 135558986467136 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0419 00:21:42.596731 135558986467136 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 00:21:42.636191 135558986467136 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.040623s
I0419 00:21:42.636376 135558986467136 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 35.057 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04400157928466797 s) (per-host)
I0419 00:21:42.636429 135558986467136 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.044064s (batch_requests_ready=0.001727s, total_serialization_initiated=0.042267s, others=0.000070s)
I0419 00:21:42.636516 135558986467136 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.048154s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.048139)
I0419 00:21:42.637153 135558986467136 event_tracking.py:125] [process=3] [async] Finished blocking save in 0.80 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_10_shardy_true/checkpoints/9.
I0419 00:21:42.637513 135432164321024 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 00:41:42.637476
I0419 00:21:42.639415 135558986467136 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0419 00:21:42.639698 135432155928320 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0419 00:21:42.639857 135558986467136 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_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776558073.562734, 'wait_for_prev_duration_secs': 28.27348279953003, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776558101.8381066, 'checkpointer_blocking_duration_secs': 0.7995531558990479, 'get_old_steps_start_time': 1776558102.6376822, 'get_old_steps_duration_secs': 3.0040740966796875e-05, 'checkpoint_manager_blocking_start_time': 1776558073.5608788, 'checkpoint_manager_blocking_duration_secs': 29.078946590423584}
I0419 00:21:42.639971 135558986467136 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0419 00:21:42.640016 135558986467136 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0419 00:21:52.153356 135405798147840 array_metadata_store.py:203] [process=3][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_10_shardy_true/checkpoints/9/items/array_metadatas/process_3
I0419 00:22:26.793779 135432164321024 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 35.736 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 44.20136618614197 s) (per-host)
I0419 00:22:26.793903 135432164321024 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 44.156275s.
I0419 00:22:35.076051 135432164321024 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 52.438408s.
I0419 00:22:35.076351 135432155928320 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0419 00:22:35.076469 135432155928320 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0419 00:22:35.076518 135432155928320 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0419 00:22:35.078743 135432155928320 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0419 00:22:35.078924 135558986467136 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0419 00:22:35.079083 135558986467136 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0419 00:22:35.080024 135558986467136 metric_logger.py:196] completed step: 9, seconds: 0.148, TFLOP/s/device: 92.093, Tokens/s/device: 13881.264, 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 00:22:43 UTC 2026
EXIT_CODE=0