MaxView

XPK Start: Mon Apr 20 15:38:07 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-20 15:38:32.412764: 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)
I0420 15:38:32.588089 137812465477440 max_utils.py:273] Attempting to initialize the jax distributed system...
I0420 15:38:41.626066 137812465477440 distributed.py:149] Starting JAX distributed service on [::]:8482
I0420 15:38:41.628288 137812465477440 distributed.py:172] Connecting to JAX distributed service on mt-01-base-jh70z-slice-job-0-0.mt-01-base-jh70z:8482
I0420 15:38:45.272289 137812465477440 max_utils.py:284] Jax distributed system initialized!
I0420 15:38:50.362884 137812465477440 max_utils.py:800] System Information: Jax Version: 0.9.2
I0420 15:38:50.362989 137812465477440 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0420 15:38:50.363028 137812465477440 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
I0420 15:38:50.363063 137812465477440 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0420 15:38:51.421772 137812465477440 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 15:38:51.422348 137812465477440 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 15:38:51.422540 137812465477440 checkpointing.py:688] Setting up checkpoint logger...
I0420 15:38:51.422591 137812465477440 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0420 15:38:51.422635 137812465477440 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0420 15:38:51.422968 137812465477440 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 0x7d563f44c5f0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0420 15:38:55.375983 137812465477440 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0420 15:38:55.376216 137812465477440 checkpoint_manager.py:708] [process=2][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d55337021b0>}, handler_registry=None
I0420 15:38:55.376447 137812465477440 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d55337021b0>` 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`.
I0420 15:38:55.376494 137812465477440 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d5533713bf0>` 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`.
I0420 15:38:55.376542 137812465477440 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 0x7d55337021b0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d55337021b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d5533713bf0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d5533713bf0>}).
I0420 15:38:55.376901 137812465477440 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0420 15:38:55.376973 137812465477440 async_checkpointer.py:192] [process=2][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d55339b77e0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0420 15:38:56.507917 137812465477440 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260420_153543/linen_xpk_feat_nnx_post_train_fixes_20260420_153543_01_base/checkpoints
I0420 15:38:56.525294 137812465477440 checkpoint_manager.py:929] [process=2][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_20260420_153543/linen_xpk_feat_nnx_post_train_fixes_20260420_153543_01_base/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d5533702870>
I0420 15:38:56.525420 137812465477440 checkpointing.py:302] Checkpoint manager created!
I0420 15:38:57.465049 137812465477440 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0420 15:38:57.465148 137812465477440 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 15:38:57.845616 137812465477440 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 15:38:57.845705 137812465477440 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 15:38:57.862097 137812465477440 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 15:38:57.862158 137812465477440 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 15:38:57.886541 137812465477440 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 15:38:57.886604 137812465477440 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 15:38:57.903001 137812465477440 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 15:38:57.903062 137812465477440 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 15:38:57.919456 137812465477440 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 15:38:57.919529 137812465477440 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 15:38:57.944017 137812465477440 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0420 15:38:57.944084 137812465477440 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 15:38:57.964929 137812465477440 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0420 15:38:57.964998 137812465477440 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0420 15:38:58.179017 137812465477440 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0420 15:38:58.179126 137812465477440 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0420 15:38:59.603568 137812465477440 maxtext_utils.py:1827]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0420 15:38:59.603691 137812465477440 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)
I0420 15:38:59.603744 137812465477440 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)
I0420 15:38:59.603815 137812465477440 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')
I0420 15:38:59.603891 137812465477440 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)
I0420 15:38:59.603967 137812465477440 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)
I0420 15:38:59.604025 137812465477440 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)
I0420 15:38:59.604076 137812465477440 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')
I0420 15:38:59.604118 137812465477440 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)
I0420 15:38:59.604153 137812465477440 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)
I0420 15:38:59.604200 137812465477440 maxtext_utils.py:1827]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0420 15:38:59.604245 137812465477440 maxtext_utils.py:1827]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0420 15:39:00.092117 137812465477440 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 15:39:00.092215 137812465477440 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 15:39:00.107549 137812465477440 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 15:39:00.107608 137812465477440 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 15:39:11.179942 137812465477440 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.
I0420 15:39:11.180707 137812465477440 metric_logger.py:301] number parameters: 1.104 billion
I0420 15:39:22.408551 137812465477440 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0420 15:39:22.563725 137812465477440 checkpointing.py:798] Waited 0.1551649570465088 seconds for step 0 to finish before starting checkpointing.
I0420 15:39:22.566384 137812465477440 checkpoint_manager.py:2009] [process=2][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 15:39:22.568122 137812465477440 checkpoint_manager.py:1512] [process=2] Saving checkpoint at step 0
I0420 15:39:22.569586 137812465477440 event_tracking.py:70] [process=2] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260420_153543/linen_xpk_feat_nnx_post_train_fixes_20260420_153543_01_base/checkpoints/0.
I0420 15:39:23.345305 137812465477440 signaling_client.py:364] Using JaxDistributedSignalingClient
I0420 15:39:23.346244 137812465477440 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0420 15:39:23.346300 137812465477440 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
I0420 15:39:23.626136 137812465477440 base_pytree_checkpoint_handler.py:154] [process=2][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.280908s
I0420 15:39:23.626312 137812465477440 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/blocking_gbytes_per_sec: 5.382 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2866053581237793 s) (per-host)
I0420 15:39:23.626367 137812465477440 base_pytree_checkpoint_handler.py:768] [process=2][thread=MainThread] Initiated Pytree async_save. Time taken: 0.286669s (batch_requests_ready=0.002455s, total_serialization_initiated=0.284143s, others=0.000072s)
I0420 15:39:23.626459 137812465477440 composite_checkpoint_handler.py:715] [process=2][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.291036s (all_items=0.000018s, per_item={'items': '0.00001788'}, temp_paths=0.291018)
I0420 15:39:23.627320 137812465477440 event_tracking.py:125] [process=2] [async] Finished blocking save in 1.06 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260420_153543/linen_xpk_feat_nnx_post_train_fixes_20260420_153543_01_base/checkpoints/0.
I0420 15:39:23.627626 137687205910272 async_checkpointer.py:76] [process=2][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 15:59:23.627592
I0420 15:39:23.633083 137812465477440 checkpoint_manager.py:1560] [process=2][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 15:39:23.633395 137684542547712 async_checkpointer.py:280] [process=2][thread=save_finalize] Waiting for background save thread=async_save.
I0420 15:39:23.633571 137812465477440 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260420_153543/linen_xpk_feat_nnx_post_train_fixes_20260420_153543_01_base/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776699562.5663586, 'wait_for_prev_duration_secs': 8.654594421386719e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776699562.5681713, 'checkpointer_blocking_duration_secs': 1.0596089363098145, 'get_old_steps_start_time': 1776699563.6278114, 'get_old_steps_duration_secs': 3.1948089599609375e-05, 'checkpoint_manager_blocking_start_time': 1776699562.564257, 'checkpoint_manager_blocking_duration_secs': 1.0692768096923828}
I0420 15:39:23.633679 137812465477440 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0420 15:39:23.633730 137812465477440 max_utils.py:750] 
Memstats: After params initialized:
I0420 15:39:23.633779 137812465477440 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_8(process=2,(0,2,0,0))
I0420 15:39:23.633818 137812465477440 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_9(process=2,(1,2,0,0))
I0420 15:39:23.633844 137812465477440 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_12(process=2,(0,3,0,0))
I0420 15:39:23.633869 137812465477440 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_13(process=2,(1,3,0,0))
I0420 15:39:23.944674 137812465477440 metric_logger.py:196] completed step: 0, seconds: 11.228, TFLOP/s/device: 1.210, Tokens/s/device: 182.406, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0420 15:39:24.128460 137812465477440 metric_logger.py:196] completed step: 1, seconds: 1.535, TFLOP/s/device: 8.854, Tokens/s/device: 1334.540, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0420 15:39:24.560319 137812465477440 metric_logger.py:196] completed step: 2, seconds: 0.036, TFLOP/s/device: 373.550, Tokens/s/device: 56305.501, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0420 15:39:24.707617 137812465477440 metric_logger.py:196] completed step: 3, seconds: 0.432, TFLOP/s/device: 31.444, Tokens/s/device: 4739.644, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0420 15:39:25.002629 137812465477440 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.825, Tokens/s/device: 13388.684, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0420 15:39:25.008539 137812465477440 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.214, Tokens/s/device: 13899.446, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0420 15:39:27.068190    2765 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0420 15:39:29.311600 137685052692224 array_metadata_store.py:203] [process=2][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260420_153543/linen_xpk_feat_nnx_post_train_fixes_20260420_153543_01_base/checkpoints/0/items/array_metadatas/process_2
I0420 15:39:47.206369 137812465477440 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.976, Tokens/s/device: 6930.016, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0420 15:39:47.353627 137812465477440 metric_logger.py:196] completed step: 7, seconds: 22.051, TFLOP/s/device: 0.616, Tokens/s/device: 92.875, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0420 15:39:47.500880 137812465477440 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.268, Tokens/s/device: 13455.449, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0420 15:39:47.647733 137812465477440 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0420 15:39:47.648427 137812465477440 checkpointing.py:798] Waited 0.0007150173187255859 seconds for step 9 to finish before starting checkpointing.
I0420 15:39:47.650693 137812465477440 checkpoint_manager.py:2020] [process=2][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 15:40:00.345221 137687205910272 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/gbytes_per_sec: 42.685 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.005473375320435 s) (per-host)
I0420 15:40:00.345345 137687205910272 async_checkpointer.py:90] [process=2][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.717604s.
I0420 15:40:10.585473 137687205910272 async_checkpointer.py:160] [process=2][thread=async_save] Background save thread done. Time taken: 46.957714s.
I0420 15:40:10.585802 137684542547712 async_checkpointer.py:288] [process=2][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 15:40:10.585922 137684542547712 async_checkpointer.py:298] [process=2][thread=save_finalize] No errors found in background save thread=async_save.
I0420 15:40:10.585973 137684542547712 checkpoint_manager.py:2137] [process=2][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0420 15:40:10.588089 137684542547712 checkpoint_manager.py:2146] [process=2][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0420 15:40:10.588282 137812465477440 checkpoint_manager.py:2032] [process=2][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0420 15:40:10.588416 137812465477440 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.937723 seconds. If this number is high, consider checkpointing less frequently.
I0420 15:40:10.590246 137812465477440 checkpoint_manager.py:1512] [process=2] Saving checkpoint at step 9
I0420 15:40:10.592314 137812465477440 event_tracking.py:70] [process=2] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260420_153543/linen_xpk_feat_nnx_post_train_fixes_20260420_153543_01_base/checkpoints/9.
I0420 15:40:11.347799 137812465477440 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0420 15:40:11.347895 137812465477440 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
I0420 15:40:11.382431 137812465477440 base_pytree_checkpoint_handler.py:154] [process=2][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.035879s
I0420 15:40:11.382617 137812465477440 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/blocking_gbytes_per_sec: 39.237 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.03931474685668945 s) (per-host)
I0420 15:40:11.382671 137812465477440 base_pytree_checkpoint_handler.py:768] [process=2][thread=MainThread] Initiated Pytree async_save. Time taken: 0.039380s (batch_requests_ready=0.001780s, total_serialization_initiated=0.037528s, others=0.000071s)
I0420 15:40:11.382764 137812465477440 composite_checkpoint_handler.py:715] [process=2][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.043574s (all_items=0.000015s, per_item={'items': '0.00001526'}, temp_paths=0.043558)
I0420 15:40:11.383462 137812465477440 event_tracking.py:125] [process=2] [async] Finished blocking save in 0.79 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260420_153543/linen_xpk_feat_nnx_post_train_fixes_20260420_153543_01_base/checkpoints/9.
I0420 15:40:11.383787 137684542547712 async_checkpointer.py:76] [process=2][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 16:00:11.383754
I0420 15:40:11.385733 137812465477440 checkpoint_manager.py:1560] [process=2][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 15:40:11.386014 137685052692224 async_checkpointer.py:280] [process=2][thread=save_finalize] Waiting for background save thread=async_save.
I0420 15:40:11.386151 137812465477440 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260420_153543/linen_xpk_feat_nnx_post_train_fixes_20260420_153543_01_base/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776699587.650663, 'wait_for_prev_duration_secs': 22.93772292137146, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776699610.5902858, 'checkpointer_blocking_duration_secs': 0.7936451435089111, 'get_old_steps_start_time': 1776699611.3839526, 'get_old_steps_duration_secs': 3.1948089599609375e-05, 'checkpoint_manager_blocking_start_time': 1776699587.648677, 'checkpoint_manager_blocking_duration_secs': 23.737438917160034}
I0420 15:40:11.386311 137812465477440 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0420 15:40:11.386357 137812465477440 checkpoint_manager.py:2020] [process=2][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 15:40:16.961746 137665529755392 array_metadata_store.py:203] [process=2][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260420_153543/linen_xpk_feat_nnx_post_train_fixes_20260420_153543_01_base/checkpoints/9/items/array_metadatas/process_2
I0420 15:40:53.753597 137684542547712 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/gbytes_per_sec: 37.246 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.41026544570923 s) (per-host)
I0420 15:40:53.753709 137684542547712 async_checkpointer.py:90] [process=2][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.369812s.
I0420 15:41:03.097014 137684542547712 async_checkpointer.py:160] [process=2][thread=async_save] Background save thread done. Time taken: 51.713101s.
I0420 15:41:03.097311 137685052692224 async_checkpointer.py:288] [process=2][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 15:41:03.097433 137685052692224 async_checkpointer.py:298] [process=2][thread=save_finalize] No errors found in background save thread=async_save.
I0420 15:41:03.097480 137685052692224 checkpoint_manager.py:2137] [process=2][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0420 15:41:03.099447 137685052692224 checkpoint_manager.py:2146] [process=2][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0420 15:41:03.099651 137812465477440 checkpoint_manager.py:2032] [process=2][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0420 15:41:03.099793 137812465477440 checkpoint_manager.py:2009] [process=2][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 15:41:03.100590 137812465477440 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.253, Tokens/s/device: 13905.392, 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: Mon Apr 20 15:41:13 UTC 2026
EXIT_CODE=0