MaxView

Timing Test Reference Updates
← Back to run

Log Summary

XPK Start: Sat Apr 25 18:14:30 UTC 2026
2026-04-25 18:14:34.536536: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:467] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
E0000 00:00:1777140874.549359      10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1777140874.553160      10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
W0000 00:00:1777140874.564288      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1777140874.564308      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1777140874.564310      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1777140874.564312      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
2026-04-25 18:14:53.698080: 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)
I0425 18:14:54.222148 133335477552960 max_utils.py:273] Attempting to initialize the jax distributed system...
INFO:2026-04-25 18:15:03,262:jax._src.distributed:140: Starting JAX distributed service on [::]:8482
I0425 18:15:03.262721 133335477552960 distributed.py:140] Starting JAX distributed service on [::]:8482
INFO:2026-04-25 18:15:03,265:jax._src.distributed:157: Connecting to JAX distributed service on mt-05-fp8-x4lue-slice-job-0-0.mt-05-fp8-x4lue:8482
I0425 18:15:03.265003 133335477552960 distributed.py:157] Connecting to JAX distributed service on mt-05-fp8-x4lue-slice-job-0-0.mt-05-fp8-x4lue:8482
I0425 18:15:04.946344 133335477552960 max_utils.py:284] Jax distributed system initialized!
I0425 18:15:11.876888 133335477552960 max_utils.py:800] System Information: Jax Version: 0.8.1
I0425 18:15:11.876995 133335477552960 max_utils.py:801] System Information: Jaxlib Version: 0.8.1
I0425 18:15:11.877039 133335477552960 max_utils.py:802] System Information: Jax Backend: PJRT C API
TFRT TPU v6 lite
Built on Nov 12 2025 14:16:36 (1762985796) cl/831091709
I0425 18:15:11.877074 133335477552960 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0425 18:15:12.493946 133335477552960 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0425 18:15:12.494248 133335477552960 checkpointing.py:677] Setting up checkpoint logger...
I0425 18:15:12.494306 133335477552960 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0425 18:15:12.494351 133335477552960 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None
I0425 18:15:12.494717 133335477552960 base_pytree_checkpoint_handler.py:415] 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 0x79441bf536e0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0425 18:15:15.430572 133335477552960 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0425 18:15:15.430973 133335477552960 checkpoint_manager.py:709] [process=5][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x793176e2dbe0>}, handler_registry=None
I0425 18:15:15.431213 133335477552960 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x793176e2dbe0>` 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`.
I0425 18:15:15.431263 133335477552960 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x792fdc0d3080>` 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`.
I0425 18:15:15.431300 133335477552960 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 0x793176e2dbe0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x793176e2dbe0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x792fdc0d3080>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x792fdc0d3080>}).
I0425 18:15:15.431845 133335477552960 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33
I0425 18:15:15.431931 133335477552960 async_checkpointer.py:177] [process=5][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x792f5c712840> timeout: 600 secs and primary_host=0 for async checkpoint writes
I0425 18:15:17.075299 133335477552960 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints
I0425 18:15:17.086405 133335477552960 checkpoint_manager.py:929] [process=5][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_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x79317c860260>
I0425 18:15:17.086522 133335477552960 checkpointing.py:301] Checkpoint manager created!
I0425 18:15:18.171867 133335477552960 nnx_wrappers.py:442] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0425 18:15:18.171974 133335477552960 nnx_wrappers.py:442] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 18:15:18.549893 133335477552960 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0425 18:15:18.549983 133335477552960 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 18:15:18.565618 133335477552960 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0425 18:15:18.565692 133335477552960 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 18:15:18.624579 133335477552960 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0425 18:15:18.624672 133335477552960 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 18:15:18.640486 133335477552960 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0425 18:15:18.640547 133335477552960 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 18:15:18.656135 133335477552960 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0425 18:15:18.656196 133335477552960 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 18:15:18.679886 133335477552960 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0425 18:15:18.679959 133335477552960 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 18:15:18.739481 133335477552960 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0425 18:15:18.739566 133335477552960 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0425 18:15:19.204250 133335477552960 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0425 18:15:19.204377 133335477552960 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0425 18:15:21.248554 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.248699 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.248755 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.248796 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.248831 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.248865 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.248898 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.248929 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.248962 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.248992 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249022 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249052 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249082 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249113 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249144 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249172 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249201 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249237 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249266 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249294 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249321 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249349 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249376 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249404 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249431 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249457 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249484 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249511 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249536 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249563 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249589 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249615 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249641 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249691 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249726 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249770 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249801 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249829 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249862 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249889 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249916 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249942 133335477552960 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0425 18:15:21.249993 133335477552960 maxtext_utils.py:1620]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   PartitionSpec('norm',)
    Physical:  (None,)
I0425 18:15:21.250047 133335477552960 maxtext_utils.py:1620]  params/params/decoder/layers/mlp/wi_0/kernel
    Shape:     float32[2048,16,7168]
    Logical:   PartitionSpec('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0425 18:15:21.250084 133335477552960 maxtext_utils.py:1620]  params/params/decoder/layers/mlp/wi_1/kernel
    Shape:     float32[2048,16,7168]
    Logical:   PartitionSpec('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0425 18:15:21.250132 133335477552960 maxtext_utils.py:1620]  params/params/decoder/layers/mlp/wo/kernel
    Shape:     float32[7168,16,2048]
    Logical:   PartitionSpec('mlp', 'layers', 'embed')
    Physical:  (None, None, 'fsdp')
I0425 18:15:21.250176 133335477552960 maxtext_utils.py:1620]  params/params/decoder/layers/post_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   PartitionSpec('norm', 'layers')
    Physical:  (None, None)
I0425 18:15:21.250214 133335477552960 maxtext_utils.py:1620]  params/params/decoder/layers/pre_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   PartitionSpec('norm', 'layers')
    Physical:  (None, None)
I0425 18:15:21.250261 133335477552960 maxtext_utils.py:1620]  params/params/decoder/layers/self_attention/key/kernel
    Shape:     float32[2048,16,16,128]
    Logical:   PartitionSpec('embed', 'layers', 'kv_heads', 'kv_head_dim')
    Physical:  ('fsdp', None, None, None)
I0425 18:15:21.250308 133335477552960 maxtext_utils.py:1620]  params/params/decoder/layers/self_attention/out/kernel
    Shape:     float32[16,16,128,2048]
    Logical:   PartitionSpec('heads', 'layers', 'kv', 'embed')
    Physical:  (None, None, None, 'fsdp')
I0425 18:15:21.250343 133335477552960 maxtext_utils.py:1620]  params/params/decoder/layers/self_attention/query/kernel
    Shape:     float32[2048,16,16,128]
    Logical:   PartitionSpec('embed', 'layers', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None, None)
I0425 18:15:21.250375 133335477552960 maxtext_utils.py:1620]  params/params/decoder/layers/self_attention/value/kernel
    Shape:     float32[2048,16,16,128]
    Logical:   PartitionSpec('embed', 'layers', 'kv_heads', 'kv_head_dim')
    Physical:  ('fsdp', None, None, None)
I0425 18:15:21.250421 133335477552960 maxtext_utils.py:1620]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   PartitionSpec('embed', 'vocab')
    Physical:  ('fsdp', None)

I0425 18:15:21.250464 133335477552960 maxtext_utils.py:1620]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   PartitionSpec('vocab', 'embed')
    Physical:  (None, 'fsdp')
I0425 18:15:23.055155 133335477552960 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0425 18:15:23.055249 133335477552960 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0425 18:15:23.070339 133335477552960 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0425 18:15:23.070396 133335477552960 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0425 18:15:36.344259 133335477552960 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.
I0425 18:15:36.345300 133335477552960 metric_logger.py:289] number parameters: 1.105 billion
I0425 18:15:51.354938 133335477552960 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0425 18:15:51.525825 133335477552960 checkpointing.py:776] Waited 0.17088103294372559 seconds for step 0 to finish before starting checkpointing.
I0425 18:15:51.528440 133335477552960 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 18:15:51.530459 133335477552960 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 0
I0425 18:15:51.533220 133335477552960 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints/0.
I0425 18:15:52.976837 133335477552960 signaling_client.py:364] Using JaxDistributedSignalingClient
I0425 18:15:52.977873 133335477552960 jax_array_handlers.py:358] Scheduling D2H of 81 prioritized jax.Array.
I0425 18:15:52.977951 133335477552960 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
I0425 18:15:53.338879 133335477552960 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.362169s
I0425 18:15:53.339060 133335477552960 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.456 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 1.059333324432373 s) (per-host)
I0425 18:15:53.339123 133335477552960 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 1.059411s (batch_requests_ready=0.691750s, total_serialization_initiated=0.367575s, others=0.000086s)
I0425 18:15:53.339256 133335477552960 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 1.063515s (all_items=0.000020s, per_item={'items': '0.00002003'}, temp_paths=1.063495)
I0425 18:15:53.340309 133205222995712 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started.
I0425 18:15:53.340478 133335477552960 async_checkpointer.py:561] Finished blocking save. Time taken: 1.809947s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints/0.
I0425 18:15:53.370937 133335477552960 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 18:15:53.371214 133212227495680 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0425 18:15:53.371374 133335477552960 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777140951.5284219, 'wait_for_prev_duration_secs': 6.580352783203125e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777140951.5305, 'checkpointer_blocking_duration_secs': 1.810081958770752, 'get_old_steps_start_time': 1777140953.3406038, 'get_old_steps_duration_secs': 2.4318695068359375e-05, 'checkpoint_manager_blocking_start_time': 1777140951.5264978, 'checkpoint_manager_blocking_duration_secs': 1.8448371887207031}
I0425 18:15:53.371485 133335477552960 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0425 18:15:53.371537 133335477552960 max_utils.py:750] 
Memstats: After params initialized:
I0425 18:15:53.371588 133335477552960 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_18(process=5,(2,4,0,0))
I0425 18:15:53.371623 133335477552960 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_19(process=5,(3,4,0,0))
I0425 18:15:53.371673 133335477552960 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_22(process=5,(2,5,0,0))
I0425 18:15:53.371702 133335477552960 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_23(process=5,(3,5,0,0))
I0425 18:15:53.638545 133335477552960 metric_logger.py:185] completed step: 0, seconds: 15.010, TFLOP/s/device: 0.905, Tokens/s/device: 136.447, total_weights: 65536, loss: 10.874
I0425 18:15:53.912387 133335477552960 metric_logger.py:185] completed step: 1, seconds: 2.282, TFLOP/s/device: 5.955, Tokens/s/device: 897.583, total_weights: 65536, loss: 10.874
I0425 18:15:54.321626 133335477552960 metric_logger.py:185] completed step: 2, seconds: 0.117, TFLOP/s/device: 115.974, Tokens/s/device: 17480.816, total_weights: 65536, loss: 10.816
I0425 18:15:54.479762 133335477552960 metric_logger.py:185] completed step: 3, seconds: 0.409, TFLOP/s/device: 33.226, Tokens/s/device: 5008.143, total_weights: 65536, loss: 10.431
I0425 18:15:54.797029 133335477552960 metric_logger.py:185] completed step: 4, seconds: 0.164, TFLOP/s/device: 83.010, Tokens/s/device: 12512.143, total_weights: 65536, loss: 9.992
I0425 18:15:54.803069 133335477552960 metric_logger.py:185] completed step: 5, seconds: 0.158, TFLOP/s/device: 85.963, Tokens/s/device: 12957.269, total_weights: 65536, loss: 9.549
I0425 18:16:15.094258 133335477552960 metric_logger.py:185] completed step: 6, seconds: 0.318, TFLOP/s/device: 42.781, Tokens/s/device: 6448.383, total_weights: 65536, loss: 9.111
I0425 18:16:15.252426 133335477552960 metric_logger.py:185] completed step: 7, seconds: 20.134, TFLOP/s/device: 0.675, Tokens/s/device: 101.718, total_weights: 65536, loss: 8.685
I0425 18:16:15.410678 133335477552960 metric_logger.py:185] completed step: 8, seconds: 0.163, TFLOP/s/device: 83.471, Tokens/s/device: 12581.630, total_weights: 65536, loss: 8.281
I0425 18:16:15.415566 133335477552960 checkpointing.py:772] Waiting for step 10 to finish before checkpoint...
I0425 18:16:15.727442 133335477552960 checkpointing.py:776] Waited 0.3118398189544678 seconds for step 10 to finish before starting checkpointing.
I0425 18:16:15.730148 133335477552960 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 18:16:17.359951    2734 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0425 18:16:19.289974 133212235888384 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints/0/items/array_metadatas/process_5
I0425 18:16:34.055557 133205222995712 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.815 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.77578163146973 s) (per-host)
I0425 18:16:34.055692 133205222995712 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 40.715273s.
I0425 18:16:43.040618 133205222995712 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 49.700184s.
I0425 18:16:43.040949 133212227495680 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 18:16:43.041074 133212227495680 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0425 18:16:43.041136 133212227495680 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0425 18:16:43.042620 133212227495680 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0425 18:16:43.042806 133335477552960 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0425 18:16:43.042937 133335477552960 checkpoint_manager.py:1458] Waiting for previous save to complete took 27.312796 seconds. If this number is high, consider checkpointing less frequently.
I0425 18:16:43.044679 133335477552960 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 10
I0425 18:16:43.047983 133335477552960 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints/10.
I0425 18:16:44.054065 133335477552960 jax_array_handlers.py:358] Scheduling D2H of 81 prioritized jax.Array.
I0425 18:16:44.054164 133335477552960 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
I0425 18:16:44.107416 133335477552960 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.054312s
I0425 18:16:44.107587 133335477552960 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 4.564 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.33803319931030273 s) (per-host)
I0425 18:16:44.107640 133335477552960 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.338096s (batch_requests_ready=0.280225s, total_serialization_initiated=0.057799s, others=0.000073s)
I0425 18:16:44.107787 133335477552960 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.342318s (all_items=0.000016s, per_item={'items': '0.00001597'}, temp_paths=0.342302)
I0425 18:16:44.108676 133205214603008 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started.
I0425 18:16:44.108838 133335477552960 async_checkpointer.py:561] Finished blocking save. Time taken: 1.064090s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints/10.
I0425 18:16:44.119184 133335477552960 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=10] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 18:16:44.119430 133212227495680 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0425 18:16:44.119559 133335477552960 standard_logger.py:34] {'step': 10, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777140975.730117, 'wait_for_prev_duration_secs': 27.312796115875244, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777141003.0447183, 'checkpointer_blocking_duration_secs': 1.0642290115356445, 'get_old_steps_start_time': 1777141004.1089683, 'get_old_steps_duration_secs': 2.5510787963867188e-05, 'checkpoint_manager_blocking_start_time': 1777140975.7282243, 'checkpoint_manager_blocking_duration_secs': 28.391299962997437}
I0425 18:16:44.119682 133335477552960 checkpointing.py:408] Started an asynchronous checkpoint save for step 10
I0425 18:16:44.120098 133335477552960 metric_logger.py:185] completed step: 9, seconds: 0.158, TFLOP/s/device: 85.864, Tokens/s/device: 12942.366, total_weights: 65536, loss: 7.908
I0425 18:16:44.126979 133335477552960 metric_logger.py:185] completed step: 10, seconds: 0.158, TFLOP/s/device: 85.945, Tokens/s/device: 12954.564, total_weights: 65536, loss: 7.568
I0425 18:16:44.284553 133335477552960 metric_logger.py:185] completed step: 11, seconds: 28.709, TFLOP/s/device: 0.473, Tokens/s/device: 71.336, total_weights: 65536, loss: 7.288
I0425 18:16:44.442801 133335477552960 metric_logger.py:185] completed step: 12, seconds: 0.006, TFLOP/s/device: 2224.117, Tokens/s/device: 335243.084, total_weights: 65536, loss: 7.032
I0425 18:16:44.601003 133335477552960 metric_logger.py:185] completed step: 13, seconds: 0.159, TFLOP/s/device: 85.719, Tokens/s/device: 12920.565, total_weights: 65536, loss: 6.815
I0425 18:16:44.759090 133335477552960 metric_logger.py:185] completed step: 14, seconds: 0.158, TFLOP/s/device: 85.866, Tokens/s/device: 12942.693, total_weights: 65536, loss: 6.635
I0425 18:16:44.917335 133335477552960 metric_logger.py:185] completed step: 15, seconds: 0.158, TFLOP/s/device: 85.880, Tokens/s/device: 12944.820, total_weights: 65536, loss: 6.492
I0425 18:16:45.075357 133335477552960 metric_logger.py:185] completed step: 16, seconds: 0.158, TFLOP/s/device: 85.852, Tokens/s/device: 12940.567, total_weights: 65536, loss: 6.380
I0425 18:16:45.233876 133335477552960 metric_logger.py:185] completed step: 17, seconds: 0.158, TFLOP/s/device: 86.023, Tokens/s/device: 12966.293, total_weights: 65536, loss: 6.293
I0425 18:16:45.391872 133335477552960 metric_logger.py:185] completed step: 18, seconds: 0.158, TFLOP/s/device: 85.976, Tokens/s/device: 12959.237, total_weights: 65536, loss: 6.223
I0425 18:16:45.549783 133335477552960 checkpointing.py:772] Waiting for step 19 to finish before checkpoint...
I0425 18:16:45.550755 133335477552960 checkpointing.py:776] Waited 0.000982522964477539 seconds for step 19 to finish before starting checkpointing.
I0425 18:16:45.552768 133335477552960 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=10][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 18:16:49.320581 133204125583104 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints/10/items/array_metadatas/process_5
I0425 18:17:25.733098 133205214603008 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.646 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.96349620819092 s) (per-host)
I0425 18:17:25.733226 133205214603008 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.624427s.
I0425 18:17:33.719901 133205214603008 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 49.611095s.
I0425 18:17:33.720121 133212227495680 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 18:17:33.720179 133212227495680 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0425 18:17:33.720233 133212227495680 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=10] CheckpointManager Save Finalize is syncing with other hosts...
I0425 18:17:33.722121 133212227495680 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=10] CheckpointManager Save Finalize is done on all hosts.
I0425 18:17:33.722311 133335477552960 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=10][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=10.
W0425 18:17:33.722441 133335477552960 checkpoint_manager.py:1458] Waiting for previous save to complete took 48.169685 seconds. If this number is high, consider checkpointing less frequently.
I0425 18:17:33.724018 133335477552960 checkpoint_manager.py:1518] [process=5] Saving checkpoint at step 19
I0425 18:17:33.727343 133335477552960 async_checkpointer.py:452] [process=5] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints/19.
I0425 18:17:34.515459 133335477552960 jax_array_handlers.py:358] Scheduling D2H of 81 prioritized jax.Array.
I0425 18:17:34.515558 133335477552960 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
I0425 18:17:34.568791 133335477552960 base_pytree_checkpoint_handler.py:153] [process=5][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.054582s
I0425 18:17:34.568956 133335477552960 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/blocking_gbytes_per_sec: 4.967 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3105804920196533 s) (per-host)
I0425 18:17:34.569004 133335477552960 base_pytree_checkpoint_handler.py:737] [process=5][thread=MainThread] Initiated Pytree async_save. Time taken: 0.310638s (batch_requests_ready=0.252527s, total_serialization_initiated=0.058048s, others=0.000063s)
I0425 18:17:34.569101 133335477552960 composite_checkpoint_handler.py:715] [process=5][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.314517s (all_items=0.000012s, per_item={'items': '0.00001192'}, temp_paths=0.314505)
I0425 18:17:34.569957 133204125583104 async_checkpointer.py:79] [process=5][thread=async_save] Background save thread started.
I0425 18:17:34.570111 133335477552960 async_checkpointer.py:561] Finished blocking save. Time taken: 0.846026s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints/19.
I0425 18:17:34.997736 133335477552960 checkpoint_manager.py:1566] [process=5][thread=MainThread][step=19] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 18:17:34.998083 133212227495680 async_checkpointer.py:265] [process=5][thread=save_finalize] Waiting for background save thread=async_save.
I0425 18:17:34.998254 133335477552960 standard_logger.py:34] {'step': 19, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777141005.5527394, 'wait_for_prev_duration_secs': 48.16968536376953, 'time_between_consecutive_saves_sec': 2.5100677013397217, 'checkpointer_blocking_start_time': 1777141053.7240567, 'checkpointer_blocking_duration_secs': 0.846165657043457, 'get_old_steps_start_time': 1777141054.5702372, 'get_old_steps_duration_secs': 2.09808349609375e-05, 'checkpoint_manager_blocking_start_time': 1777141005.5510175, 'checkpoint_manager_blocking_duration_secs': 49.44720196723938}
I0425 18:17:34.998366 133335477552960 checkpointing.py:408] Started an asynchronous checkpoint save for step 19
I0425 18:17:34.998409 133335477552960 checkpoint_manager.py:2024] [process=5][thread=MainThread][step=19][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 18:17:40.368316 133203087693568 array_metadata_store.py:203] [process=5][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260425_180433_05_fp8/checkpoints/19/items/array_metadatas/process_5
I0425 18:18:16.659434 133204125583104 base_pytree_checkpoint_handler.py:129] [process=5] /jax/checkpoint/write/gbytes_per_sec: 37.258 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.40101146697998 s) (per-host)
I0425 18:18:16.659548 133204125583104 async_checkpointer.py:90] [process=5][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.089482s.
I0425 18:18:26.251430 133204125583104 async_checkpointer.py:144] [process=5][thread=async_save] Background save thread done. Time taken: 51.681348s.
I0425 18:18:26.251738 133212227495680 async_checkpointer.py:273] [process=5][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 18:18:26.251856 133212227495680 async_checkpointer.py:283] [process=5][thread=save_finalize] No errors found in background save thread=async_save.
I0425 18:18:26.251897 133212227495680 checkpoint_manager.py:2133] [process=5][thread=save_finalize][step=19] CheckpointManager Save Finalize is syncing with other hosts...
I0425 18:18:26.253604 133212227495680 checkpoint_manager.py:2142] [process=5][thread=save_finalize][step=19] CheckpointManager Save Finalize is done on all hosts.
I0425 18:18:26.253774 133335477552960 checkpoint_manager.py:2036] [process=5][thread=MainThread][step=19][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=19.
I0425 18:18:26.253917 133335477552960 checkpoint_manager.py:2013] [process=5][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 18:18:26.254571 133335477552960 metric_logger.py:185] completed step: 19, seconds: 0.159, TFLOP/s/device: 85.703, Tokens/s/device: 12918.120, total_weights: 65536, loss: 6.167
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Sat Apr 25 18:18:36 UTC 2026
EXIT_CODE=0