XPK Start: Thu Apr 23 22:28:24 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
`rope_parameters`'s factor field must be a float >= 1, got 40
`rope_parameters`'s beta_fast field must be a float, got 32
`rope_parameters`'s beta_slow field must be a float, got 1
DeepseekV32Config got `key=rope_scaling` in kwargs but hasn't set it as attribute. For RoPE standardization you need to set `self.rope_parameters` in model's config.
2026-04-23 22:28:49.588000: 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)
I0423 22:28:49.798228 140535217510208 max_utils.py:273] Attempting to initialize the jax distributed system...
I0423 22:28:58.839818 140535217510208 distributed.py:149] Starting JAX distributed service on [::]:8482
I0423 22:28:58.842147 140535217510208 distributed.py:172] Connecting to JAX distributed service on mt-16-zarr3-true-sv-x0jkm-slice-job-0-0.mt-16-zarr3-true-sv-x0jkm:8482
I0423 22:28:59.789026 140535217510208 max_utils.py:284] Jax distributed system initialized!
I0423 22:29:05.988592 140535217510208 max_utils.py:800] System Information: Jax Version: 0.9.2
I0423 22:29:05.988696 140535217510208 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0423 22:29:05.988738 140535217510208 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
I0423 22:29:05.988775 140535217510208 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0423 22:29:06.691405 140535217510208 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0423 22:29:06.691679 140535217510208 checkpointing.py:677] Setting up checkpoint logger...
I0423 22:29:06.691732 140535217510208 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0423 22:29:06.691776 140535217510208 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0423 22:29:06.692124 140535217510208 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 0x7fd050f7a690>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0423 22:29:09.607527 140535217510208 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0423 22:29:09.607761 140535217510208 checkpoint_manager.py:708] [process=6][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fbbf0536d50>}, handler_registry=None
I0423 22:29:09.608002 140535217510208 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fbbf0536d50>` 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`.
I0423 22:29:09.608052 140535217510208 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fbc380a5430>` 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`.
I0423 22:29:09.608089 140535217510208 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 0x7fbbf0536d50>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fbbf0536d50>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fbc380a5430>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fbc380a5430>}).
I0423 22:29:09.608437 140535217510208 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0423 22:29:09.608508 140535217510208 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7fbbf03972e0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0423 22:29:10.355252 140535217510208 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints
I0423 22:29:10.668134 140535217510208 checkpoint_manager.py:929] [process=6][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=5), 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_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7fbbf0538a40>
I0423 22:29:10.668318 140535217510208 checkpointing.py:301] Checkpoint manager created!
I0423 22:29:11.601165 140535217510208 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0423 22:29:11.601287 140535217510208 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 22:29:11.980828 140535217510208 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0423 22:29:11.980922 140535217510208 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 22:29:11.997464 140535217510208 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0423 22:29:11.997530 140535217510208 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 22:29:12.021292 140535217510208 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 22:29:12.021358 140535217510208 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 22:29:12.037846 140535217510208 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 22:29:12.037907 140535217510208 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 22:29:12.054488 140535217510208 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 22:29:12.054553 140535217510208 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 22:29:12.079410 140535217510208 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0423 22:29:12.079479 140535217510208 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 22:29:12.100739 140535217510208 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0423 22:29:12.100809 140535217510208 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0423 22:29:12.312156 140535217510208 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0423 22:29:12.312266 140535217510208 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0423 22:29:13.764323 140535217510208 maxtext_utils.py:1668] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0423 22:29:13.764454 140535217510208 maxtext_utils.py:1668] params/params/decoder/layers/mlp/wi_0/kernel
Shape: float32[2048,16,7168]
Logical: P('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0423 22:29:13.764509 140535217510208 maxtext_utils.py:1668] params/params/decoder/layers/mlp/wi_1/kernel
Shape: float32[2048,16,7168]
Logical: P('embed', 'layers', 'mlp')
Physical: ('fsdp', None, None)
I0423 22:29:13.764573 140535217510208 maxtext_utils.py:1668] params/params/decoder/layers/mlp/wo/kernel
Shape: float32[7168,16,2048]
Logical: P('mlp', 'layers', 'embed')
Physical: (None, None, 'fsdp')
I0423 22:29:13.764627 140535217510208 maxtext_utils.py:1668] params/params/decoder/layers/post_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: P('norm', 'layers')
Physical: (None, None)
I0423 22:29:13.764666 140535217510208 maxtext_utils.py:1668] params/params/decoder/layers/pre_self_attention_layer_norm/scale
Shape: float32[2048,16]
Logical: P('norm', 'layers')
Physical: (None, None)
I0423 22:29:13.764718 140535217510208 maxtext_utils.py:1668] 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)
I0423 22:29:13.764769 140535217510208 maxtext_utils.py:1668] 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')
I0423 22:29:13.764809 140535217510208 maxtext_utils.py:1668] 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)
I0423 22:29:13.764845 140535217510208 maxtext_utils.py:1668] 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)
I0423 22:29:13.764892 140535217510208 maxtext_utils.py:1668] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0423 22:29:13.764940 140535217510208 maxtext_utils.py:1668] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0423 22:29:14.255996 140535217510208 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0423 22:29:14.256106 140535217510208 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0423 22:29:14.271786 140535217510208 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0423 22:29:14.271847 140535217510208 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0423 22:29:25.319010 140535217510208 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.
I0423 22:29:25.319850 140535217510208 metric_logger.py:301] number parameters: 1.104 billion
I0423 22:29:36.566369 140535217510208 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0423 22:29:36.774393 140535217510208 checkpointing.py:776] Waited 0.20800328254699707 seconds for step 0 to finish before starting checkpointing.
I0423 22:29:36.776646 140535217510208 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 22:29:36.778530 140535217510208 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0423 22:29:36.779959 140535217510208 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints/0.
I0423 22:29:37.085086 140535217510208 signaling_client.py:364] Using JaxDistributedSignalingClient
I0423 22:29:37.086212 140535217510208 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0423 22:29:37.086268 140535217510208 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
I0423 22:29:37.364167 140535217510208 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.279157s
I0423 22:29:37.364337 140535217510208 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.417 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28478050231933594 s) (per-host)
I0423 22:29:37.364393 140535217510208 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.284846s (batch_requests_ready=0.002480s, total_serialization_initiated=0.282293s, others=0.000073s)
I0423 22:29:37.364485 140535217510208 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.289029s (all_items=0.000017s, per_item={'items': '0.00001717'}, temp_paths=0.289012)
I0423 22:29:37.365302 140535217510208 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.59 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints/0.
I0423 22:29:37.365642 140406757537536 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 22:49:37.365603
I0423 22:29:37.387323 140535217510208 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 22:29:37.387595 140404053759744 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0423 22:29:37.387722 140535217510208 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776983376.7766278, 'wait_for_prev_duration_secs': 6.031990051269531e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776983376.7785673, 'checkpointer_blocking_duration_secs': 0.5872254371643066, 'get_old_steps_start_time': 1776983377.365818, 'get_old_steps_duration_secs': 3.0517578125e-05, 'checkpoint_manager_blocking_start_time': 1776983376.7748108, 'checkpoint_manager_blocking_duration_secs': 0.6128733158111572}
I0423 22:29:37.387828 140535217510208 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0423 22:29:37.387878 140535217510208 max_utils.py:750]
Memstats: After params initialized:
I0423 22:29:37.387928 140535217510208 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0423 22:29:37.387961 140535217510208 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0423 22:29:37.387991 140535217510208 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0423 22:29:37.388019 140535217510208 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0423 22:29:37.696990 140535217510208 metric_logger.py:196] completed step: 0, seconds: 11.246, TFLOP/s/device: 1.208, Tokens/s/device: 182.103, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0423 22:29:37.891669 140535217510208 metric_logger.py:196] completed step: 1, seconds: 1.129, TFLOP/s/device: 12.034, Tokens/s/device: 1813.848, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0423 22:29:38.304472 140535217510208 metric_logger.py:196] completed step: 2, seconds: 0.047, TFLOP/s/device: 288.413, Tokens/s/device: 43472.723, total_weights: 65536, loss: 10.560, lm_loss: 10.560, perplexity: 38547.719
I0423 22:29:38.452002 140535217510208 metric_logger.py:196] completed step: 3, seconds: 0.413, TFLOP/s/device: 32.899, Tokens/s/device: 4958.946, total_weights: 65536, loss: 9.980, lm_loss: 9.980, perplexity: 21590.506
I0423 22:29:38.456442 140535217510208 checkpointing.py:772] Waiting for step 5 to finish before checkpoint...
I0423 22:29:38.746079 140535217510208 checkpointing.py:776] Waited 0.28960204124450684 seconds for step 5 to finish before starting checkpointing.
I0423 22:29:38.749120 140535217510208 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 22:29:45.470755 2801 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0423 22:29:48.354212 140404062152448 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints/0/items/array_metadatas/process_6
I0423 22:30:19.541287 140406757537536 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.200 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.461708307266235 s) (per-host)
I0423 22:30:19.541397 140406757537536 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.175642s.
I0423 22:30:28.392926 140406757537536 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 51.027155s.
I0423 22:30:28.393187 140404053759744 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 22:30:28.393253 140404053759744 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0423 22:30:28.393307 140404053759744 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0423 22:30:28.395133 140404053759744 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0423 22:30:28.395312 140535217510208 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0423 22:30:28.395443 140535217510208 checkpoint_manager.py:1452] Waiting for previous save to complete took 49.646342 seconds. If this number is high, consider checkpointing less frequently.
I0423 22:30:28.397511 140535217510208 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 5
I0423 22:30:28.400050 140535217510208 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints/5.
I0423 22:30:28.706602 140535217510208 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0423 22:30:28.706696 140535217510208 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
I0423 22:30:28.748308 140535217510208 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.042966s
I0423 22:30:28.748476 140535217510208 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 33.300 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.046323537826538086 s) (per-host)
I0423 22:30:28.748535 140535217510208 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.046386s (batch_requests_ready=0.001735s, total_serialization_initiated=0.044580s, others=0.000070s)
I0423 22:30:28.748626 140535217510208 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.051402s (all_items=0.000015s, per_item={'items': '0.00001526'}, temp_paths=0.051387)
I0423 22:30:28.749300 140535217510208 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.35 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints/5.
I0423 22:30:28.749637 140383925286656 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 22:50:28.749598
I0423 22:30:28.751617 140535217510208 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=5] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 22:30:28.751887 140404053759744 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0423 22:30:28.752042 140535217510208 standard_logger.py:34] {'step': 5, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776983378.7490714, 'wait_for_prev_duration_secs': 49.646342039108276, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776983428.3975492, 'checkpointer_blocking_duration_secs': 0.35223937034606934, 'get_old_steps_start_time': 1776983428.7498116, 'get_old_steps_duration_secs': 2.956390380859375e-05, 'checkpoint_manager_blocking_start_time': 1776983378.7466786, 'checkpoint_manager_blocking_duration_secs': 50.00533103942871}
I0423 22:30:28.752168 140535217510208 checkpointing.py:408] Started an asynchronous checkpoint save for step 5
I0423 22:30:28.752939 140535217510208 metric_logger.py:196] completed step: 4, seconds: 0.154, TFLOP/s/device: 88.431, Tokens/s/device: 13329.341, total_weights: 65536, loss: 9.460, lm_loss: 9.460, perplexity: 12840.355
I0423 22:30:28.758535 140535217510208 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.350, Tokens/s/device: 13919.947, total_weights: 65536, loss: 8.959, lm_loss: 8.959, perplexity: 7776.467
I0423 22:30:33.852569 140404062152448 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints/5/items/array_metadatas/process_6
I0423 22:30:51.121581 140535217510208 metric_logger.py:196] completed step: 6, seconds: 50.301, TFLOP/s/device: 0.270, Tokens/s/device: 40.715, total_weights: 65536, loss: 8.469, lm_loss: 8.469, perplexity: 4762.740
I0423 22:30:51.269175 140535217510208 metric_logger.py:196] completed step: 7, seconds: 22.217, TFLOP/s/device: 0.612, Tokens/s/device: 92.182, total_weights: 65536, loss: 8.003, lm_loss: 8.003, perplexity: 2989.151
I0423 22:30:51.416388 140535217510208 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.414, Tokens/s/device: 13477.497, total_weights: 65536, loss: 7.572, lm_loss: 7.572, perplexity: 1943.098
I0423 22:30:51.563275 140535217510208 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0423 22:30:51.563859 140535217510208 checkpointing.py:776] Waited 0.0006043910980224609 seconds for step 9 to finish before starting checkpointing.
I0423 22:30:51.566002 140535217510208 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=5][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 22:31:10.923701 140383925286656 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.412 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.22152829170227 s) (per-host)
I0423 22:31:10.923806 140383925286656 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.174054s.
I0423 22:31:19.402355 140383925286656 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 50.652585s.
I0423 22:31:19.402714 140404053759744 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 22:31:19.402838 140404053759744 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0423 22:31:19.402888 140404053759744 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is syncing with other hosts...
I0423 22:31:19.404299 140404053759744 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is done on all hosts.
I0423 22:31:19.404462 140535217510208 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=5][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=5.
W0423 22:31:19.404603 140535217510208 checkpoint_manager.py:1452] Waiting for previous save to complete took 27.838604 seconds. If this number is high, consider checkpointing less frequently.
I0423 22:31:19.406038 140535217510208 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0423 22:31:19.408044 140535217510208 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints/9.
I0423 22:31:19.717250 140535217510208 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0423 22:31:19.717343 140535217510208 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
I0423 22:31:19.758958 140535217510208 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.042719s
I0423 22:31:19.759141 140535217510208 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 33.455 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04610943794250488 s) (per-host)
I0423 22:31:19.759193 140535217510208 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.046172s (batch_requests_ready=0.001623s, total_serialization_initiated=0.044480s, others=0.000069s)
I0423 22:31:19.759291 140535217510208 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.050270s (all_items=0.000011s, per_item={'items': '0.00001144'}, temp_paths=0.050259)
I0423 22:31:19.760003 140535217510208 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.35 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints/9.
I0423 22:31:19.760354 140404053759744 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 22:51:19.760316
I0423 22:31:19.762320 140535217510208 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 22:31:19.762570 140405693736704 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0423 22:31:19.762723 140535217510208 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776983451.5659704, 'wait_for_prev_duration_secs': 27.838603973388672, 'time_between_consecutive_saves_sec': 23.170796632766724, 'checkpointer_blocking_start_time': 1776983479.4060776, 'checkpointer_blocking_duration_secs': 0.35442185401916504, 'get_old_steps_start_time': 1776983479.7605178, 'get_old_steps_duration_secs': 2.9325485229492188e-05, 'checkpoint_manager_blocking_start_time': 1776983451.5640786, 'checkpoint_manager_blocking_duration_secs': 28.19861388206482}
I0423 22:31:19.762835 140535217510208 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0423 22:31:19.762878 140535217510208 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 22:31:24.865252 140404062152448 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260423_211617/linen_xpk_test_pipeline_scan_nnx_20260423_211617_16_zarr3_true/checkpoints/9/items/array_metadatas/process_6
I0423 22:32:02.495973 140404053759744 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 36.921 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.78290319442749 s) (per-host)
I0423 22:32:02.496113 140404053759744 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.735630s.
I0423 22:32:09.620570 140404053759744 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 49.860086s.
I0423 22:32:09.620852 140405693736704 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 22:32:09.620964 140405693736704 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0423 22:32:09.621007 140405693736704 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0423 22:32:09.623925 140405693736704 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0423 22:32:09.624134 140535217510208 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0423 22:32:09.624293 140535217510208 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 22:32:09.625249 140535217510208 metric_logger.py:196] completed step: 9, seconds: 0.148, TFLOP/s/device: 92.076, Tokens/s/device: 13878.724, total_weights: 65536, loss: 7.181, lm_loss: 7.181, perplexity: 1314.495
Per train step:
Total TFLOPs: 13.59
split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Thu Apr 23 22:32:20 UTC 2026
EXIT_CODE=0