MaxView

XPK Start: Sat Apr 18 20:48:13 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-18 20:48:37.839556: 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)
I0418 20:48:38.017621 132523731334976 max_utils.py:273] Attempting to initialize the jax distributed system...
I0418 20:48:47.058374 132523731334976 distributed.py:149] Starting JAX distributed service on [::]:8482
I0418 20:48:47.060651 132523731334976 distributed.py:172] Connecting to JAX distributed service on mt-03-dropout-jb0qt-slice-job-0-0.mt-03-dropout-jb0qt:8482
I0418 20:48:49.744584 132523731334976 max_utils.py:284] Jax distributed system initialized!
I0418 20:48:54.804112 132523731334976 max_utils.py:800] System Information: Jax Version: 0.9.2
I0418 20:48:54.804228 132523731334976 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0418 20:48:54.804270 132523731334976 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
I0418 20:48:54.804304 132523731334976 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0418 20:48:55.503057 132523731334976 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0418 20:48:55.503647 132523731334976 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0418 20:48:55.503828 132523731334976 checkpointing.py:688] Setting up checkpoint logger...
I0418 20:48:55.503880 132523731334976 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0418 20:48:55.503922 132523731334976 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0418 20:48:55.504260 132523731334976 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 0x788701f14620>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0418 20:48:59.271553 132523731334976 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0418 20:48:59.271743 132523731334976 checkpoint_manager.py:708] [process=1][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7885d31bf0b0>}, handler_registry=None
I0418 20:48:59.271976 132523731334976 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7885d31bf0b0>` 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`.
I0418 20:48:59.272023 132523731334976 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x787344175580>` 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`.
I0418 20:48:59.272057 132523731334976 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 0x7885d31bf0b0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7885d31bf0b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x787344175580>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x787344175580>}).
I0418 20:48:59.272376 132523731334976 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0418 20:48:59.272447 132523731334976 async_checkpointer.py:192] [process=1][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x78724c38f880> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0418 20:49:00.851127 132523731334976 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152_03_dropout/checkpoints
I0418 20:49:02.149612 132523731334976 checkpoint_manager.py:929] [process=1][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_linen_converter_and_sharding_tools_20260418_204152/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152_03_dropout/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x787344170c20>
I0418 20:49:02.149791 132523731334976 checkpointing.py:302] Checkpoint manager created!
I0418 20:49:02.496643 132523731334976 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0418 20:49:02.496746 132523731334976 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None).
I0418 20:49:02.878232 132523731334976 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 20:49:02.878316 132523731334976 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None).
I0418 20:49:02.894760 132523731334976 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,128,2048]....................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 20:49:02.894818 132523731334976 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,128,2048]....................................... ('fsdp', None, None).
I0418 20:49:02.918562 132523731334976 attentions.py:1154] attentions/query Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 20:49:02.918636 132523731334976 attentions.py:1154] attentions/query Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0418 20:49:02.935048 132523731334976 attentions.py:1155] attentions/key Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 20:49:02.935109 132523731334976 attentions.py:1155] attentions/key Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0418 20:49:02.951597 132523731334976 attentions.py:1156] attentions/value Logical: bfloat16[32,128,16,128]..................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 20:49:02.951656 132523731334976 attentions.py:1156] attentions/value Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0418 20:49:02.976395 132523731334976 attentions.py:1197] attentions/out Logical: bfloat16[32,128,16,128]..................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0418 20:49:02.976460 132523731334976 attentions.py:1197] attentions/out Physical: bfloat16[32,128,16,128]..................................... ('fsdp', None, None, None).
I0418 20:49:03.003548 132523731334976 linears.py:525] linears/x Logical: bfloat16[32,128,7168]....................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0418 20:49:03.003619 132523731334976 linears.py:525] linears/x Physical: bfloat16[32,128,7168]....................................... ('fsdp', None, None).
I0418 20:49:03.227474 132523731334976 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0418 20:49:03.227576 132523731334976 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0418 20:49:04.665222 132523731334976 maxtext_utils.py:1827]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0418 20:49:04.665347 132523731334976 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)
I0418 20:49:04.665400 132523731334976 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)
I0418 20:49:04.665458 132523731334976 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')
I0418 20:49:04.665510 132523731334976 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)
I0418 20:49:04.665548 132523731334976 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)
I0418 20:49:04.665600 132523731334976 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)
I0418 20:49:04.665650 132523731334976 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')
I0418 20:49:04.665691 132523731334976 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)
I0418 20:49:04.665729 132523731334976 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)
I0418 20:49:04.665778 132523731334976 maxtext_utils.py:1827]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)
I0418 20:49:04.665826 132523731334976 maxtext_utils.py:1827]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')

I0418 20:49:05.192974 132523731334976 train.py:157] train/xent Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length').
I0418 20:49:05.193070 132523731334976 train.py:157] train/xent Physical: float32[32,128]............................................. ('fsdp', None).
I0418 20:49:05.208846 132523731334976 train.py:164] train/z_loss Logical: float32[32,128]............................................. ('activation_embed_and_logits_batch', 'activation_length').
I0418 20:49:05.208906 132523731334976 train.py:164] train/z_loss Physical: float32[32,128]............................................. ('fsdp', None).
I0418 20:49:08.825521 132523731334976 max_utils.py:791] Total memory size: 0.8 GB, Output size: 0.4 GB, Temp size: 0.4 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0418 20:49:08.826307 132523731334976 metric_logger.py:301] number parameters: 1.104 billion
I0418 20:49:13.015444 132523731334976 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0418 20:49:13.090173 132523731334976 checkpointing.py:798] Waited 0.07471489906311035 seconds for step 0 to finish before starting checkpointing.
I0418 20:49:13.092856 132523731334976 checkpoint_manager.py:2009] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 20:49:13.094951 132523731334976 checkpoint_manager.py:1512] [process=1] Saving checkpoint at step 0
I0418 20:49:13.096417 132523731334976 event_tracking.py:70] [process=1] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152_03_dropout/checkpoints/0.
I0418 20:49:13.830512 132523731334976 signaling_client.py:364] Using JaxDistributedSignalingClient
I0418 20:49:13.831557 132523731334976 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 20:49:13.831614 132523731334976 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
I0418 20:49:14.111063 132523731334976 base_pytree_checkpoint_handler.py:154] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.280625s
I0418 20:49:14.111243 132523731334976 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/blocking_gbytes_per_sec: 5.391 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28615832328796387 s) (per-host)
I0418 20:49:14.111299 132523731334976 base_pytree_checkpoint_handler.py:768] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.286224s (batch_requests_ready=0.002169s, total_serialization_initiated=0.283981s, others=0.000074s)
I0418 20:49:14.111391 132523731334976 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.290385s (all_items=0.000017s, per_item={'items': '0.00001717'}, temp_paths=0.290367)
I0418 20:49:14.112283 132523731334976 event_tracking.py:125] [process=1] [async] Finished blocking save in 1.02 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152_03_dropout/checkpoints/0.
I0418 20:49:14.112625 132397769144064 async_checkpointer.py:76] [process=1][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-18 21:09:14.112586
I0418 20:49:14.122319 132523731334976 checkpoint_manager.py:1560] [process=1][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 20:49:14.122623 132396678616832 async_checkpointer.py:280] [process=1][thread=save_finalize] Waiting for background save thread=async_save.
I0418 20:49:14.122784 132523731334976 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776545353.092839, 'wait_for_prev_duration_secs': 6.008148193359375e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776545353.0949934, 'checkpointer_blocking_duration_secs': 1.0177953243255615, 'get_old_steps_start_time': 1776545354.1128154, 'get_old_steps_duration_secs': 2.8848648071289062e-05, 'checkpoint_manager_blocking_start_time': 1776545353.0906808, 'checkpoint_manager_blocking_duration_secs': 1.0320613384246826}
I0418 20:49:14.122898 132523731334976 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0418 20:49:14.122946 132523731334976 max_utils.py:750] 
Memstats: After params initialized:
I0418 20:49:14.122994 132523731334976 max_utils.py:756] 	Using (GB) 0.41 / 31.25 (1.312000%) on TPU_2(process=1,(2,0,0,0))
I0418 20:49:14.123027 132523731334976 max_utils.py:756] 	Using (GB) 0.41 / 31.25 (1.312000%) on TPU_3(process=1,(3,0,0,0))
I0418 20:49:14.123054 132523731334976 max_utils.py:756] 	Using (GB) 0.41 / 31.25 (1.312000%) on TPU_6(process=1,(2,1,0,0))
I0418 20:49:14.123080 132523731334976 max_utils.py:756] 	Using (GB) 0.41 / 31.25 (1.312000%) on TPU_7(process=1,(3,1,0,0))
I0418 20:49:14.433643 132523731334976 metric_logger.py:196] completed step: 0, seconds: 4.189, TFLOP/s/device: 0.191, Tokens/s/device: 30.556, total_weights: 4096, loss: 10.889, lm_loss: 10.889, perplexity: 53560.699
I0418 20:49:14.537653 132523731334976 metric_logger.py:196] completed step: 1, seconds: 1.417, TFLOP/s/device: 0.565, Tokens/s/device: 90.355, total_weights: 4096, loss: 10.902, lm_loss: 10.902, perplexity: 54298.523
I0418 20:49:14.946604 132523731334976 metric_logger.py:196] completed step: 2, seconds: 0.034, TFLOP/s/device: 23.535, Tokens/s/device: 3761.498, total_weights: 4096, loss: 9.956, lm_loss: 9.956, perplexity: 21082.051
I0418 20:49:15.017612 132523731334976 metric_logger.py:196] completed step: 3, seconds: 0.410, TFLOP/s/device: 1.955, Tokens/s/device: 312.450, total_weights: 4096, loss: 9.139, lm_loss: 9.139, perplexity: 9314.691
I0418 20:49:15.159491 132523731334976 metric_logger.py:196] completed step: 4, seconds: 0.076, TFLOP/s/device: 10.579, Tokens/s/device: 1690.751, total_weights: 4096, loss: 8.459, lm_loss: 8.459, perplexity: 4715.493
I0418 20:49:15.165626 132523731334976 metric_logger.py:196] completed step: 5, seconds: 0.071, TFLOP/s/device: 11.293, Tokens/s/device: 1804.825, total_weights: 4096, loss: 7.924, lm_loss: 7.924, perplexity: 2762.373
I0418 20:49:18.656760    2758 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0418 20:49:22.653315 132396697364224 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152_03_dropout/checkpoints/0/items/array_metadatas/process_1
I0418 20:49:26.543023 132523731334976 metric_logger.py:196] completed step: 6, seconds: 0.142, TFLOP/s/device: 5.632, Tokens/s/device: 900.166, total_weights: 4096, loss: 7.553, lm_loss: 7.553, perplexity: 1906.617
I0418 20:49:26.613945 132523731334976 metric_logger.py:196] completed step: 7, seconds: 11.308, TFLOP/s/device: 0.071, Tokens/s/device: 11.319, total_weights: 4096, loss: 7.267, lm_loss: 7.267, perplexity: 1431.653
I0418 20:49:26.684820 132523731334976 metric_logger.py:196] completed step: 8, seconds: 0.075, TFLOP/s/device: 10.662, Tokens/s/device: 1704.122, total_weights: 4096, loss: 7.111, lm_loss: 7.111, perplexity: 1225.983
I0418 20:49:26.754988 132523731334976 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0418 20:49:26.755651 132523731334976 checkpointing.py:798] Waited 0.0006816387176513672 seconds for step 9 to finish before starting checkpointing.
I0418 20:49:26.757601 132523731334976 checkpoint_manager.py:2020] [process=1][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 20:49:47.550076 132397769144064 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/gbytes_per_sec: 46.838 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 33.72495412826538 s) (per-host)
I0418 20:49:47.550203 132397769144064 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 33.437466s.
I0418 20:49:56.685311 132397769144064 async_checkpointer.py:160] [process=1][thread=async_save] Background save thread done. Time taken: 42.572557s.
I0418 20:49:56.685596 132396678616832 async_checkpointer.py:288] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 20:49:56.685719 132396678616832 async_checkpointer.py:298] [process=1][thread=save_finalize] No errors found in background save thread=async_save.
I0418 20:49:56.685770 132396678616832 checkpoint_manager.py:2137] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0418 20:49:56.688409 132396678616832 checkpoint_manager.py:2146] [process=1][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0418 20:49:56.688587 132523731334976 checkpoint_manager.py:2032] [process=1][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0418 20:49:56.688722 132523731334976 checkpoint_manager.py:1452] Waiting for previous save to complete took 29.931118 seconds. If this number is high, consider checkpointing less frequently.
I0418 20:49:56.690612 132523731334976 checkpoint_manager.py:1512] [process=1] Saving checkpoint at step 9
I0418 20:49:56.692561 132523731334976 event_tracking.py:70] [process=1] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152_03_dropout/checkpoints/9.
I0418 20:49:57.825125 132523731334976 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 20:49:57.825248 132523731334976 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
I0418 20:49:57.861285 132523731334976 base_pytree_checkpoint_handler.py:154] [process=1][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.037022s
I0418 20:49:57.861440 132523731334976 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/blocking_gbytes_per_sec: 38.112 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04047441482543945 s) (per-host)
I0418 20:49:57.861490 132523731334976 base_pytree_checkpoint_handler.py:768] [process=1][thread=MainThread] Initiated Pytree async_save. Time taken: 0.040534s (batch_requests_ready=0.001786s, total_serialization_initiated=0.038681s, others=0.000067s)
I0418 20:49:57.861577 132523731334976 composite_checkpoint_handler.py:715] [process=1][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.044592s (all_items=0.000015s, per_item={'items': '0.00001478'}, temp_paths=0.044577)
I0418 20:49:57.863557 132523731334976 event_tracking.py:125] [process=1] [async] Finished blocking save in 1.17 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152_03_dropout/checkpoints/9.
I0418 20:49:57.863861 132396678616832 async_checkpointer.py:76] [process=1][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-18 21:09:57.863827
I0418 20:49:57.865810 132523731334976 checkpoint_manager.py:1560] [process=1][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 20:49:57.866100 132396697364224 async_checkpointer.py:280] [process=1][thread=save_finalize] Waiting for background save thread=async_save.
I0418 20:49:57.866246 132523731334976 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152_03_dropout/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776545366.7575731, 'wait_for_prev_duration_secs': 29.93111777305603, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776545396.6906505, 'checkpointer_blocking_duration_secs': 1.1733551025390625, 'get_old_steps_start_time': 1776545397.8640335, 'get_old_steps_duration_secs': 3.218650817871094e-05, 'checkpoint_manager_blocking_start_time': 1776545366.755881, 'checkpoint_manager_blocking_duration_secs': 31.110332250595093}
I0418 20:49:57.866428 132523731334976 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0418 20:49:57.866474 132523731334976 checkpoint_manager.py:2020] [process=1][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 20:50:09.580395 132371282048768 array_metadata_store.py:203] [process=1][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260418_204152_03_dropout/checkpoints/9/items/array_metadatas/process_1
I0418 20:50:41.498627 132396678616832 base_pytree_checkpoint_handler.py:130] [process=1] /jax/orbax/write/gbytes_per_sec: 36.165 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.67762351036072 s) (per-host)
I0418 20:50:41.498750 132396678616832 async_checkpointer.py:90] [process=1][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.634774s.
I0418 20:50:52.283441 132396678616832 async_checkpointer.py:160] [process=1][thread=async_save] Background save thread done. Time taken: 54.419449s.
I0418 20:50:52.283725 132396697364224 async_checkpointer.py:288] [process=1][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 20:50:52.283856 132396697364224 async_checkpointer.py:298] [process=1][thread=save_finalize] No errors found in background save thread=async_save.
I0418 20:50:52.283916 132396697364224 checkpoint_manager.py:2137] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0418 20:50:52.286526 132396697364224 checkpoint_manager.py:2146] [process=1][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0418 20:50:52.286711 132523731334976 checkpoint_manager.py:2032] [process=1][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0418 20:50:52.286871 132523731334976 checkpoint_manager.py:2009] [process=1][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 20:50:52.287934 132523731334976 metric_logger.py:196] completed step: 9, seconds: 0.071, TFLOP/s/device: 11.316, Tokens/s/device: 1808.523, total_weights: 4096, loss: 6.981, lm_loss: 6.981, perplexity: 1075.867
Per train step:
 Total TFLOPs: 0.80 
 split as 99.60% learnable weight flops and 0.40% attention flops
XPK End: Sat Apr 18 20:51:04 UTC 2026
EXIT_CODE=0