MaxView

Timing Test Reference Updates
← Back to run

Log Summary

XPK Start: Mon Apr 20 22:43:34 UTC 2026
2026-04-20 22:43:38.606830: 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 00:00:1776725018.619801      10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1776725018.623472      10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
W0000 00:00:1776725018.634623      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776725018.634641      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776725018.634643      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776725018.634645      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
2026-04-20 22:43:57.942372: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303)
I0420 22:43:58.465340 136755829630784 max_utils.py:273] Attempting to initialize the jax distributed system...
INFO:2026-04-20 22:44:07,504:jax._src.distributed:140: Starting JAX distributed service on [::]:8482
I0420 22:44:07.504755 136755829630784 distributed.py:140] Starting JAX distributed service on [::]:8482
INFO:2026-04-20 22:44:07,507:jax._src.distributed:157: Connecting to JAX distributed service on mt-05-fp8-atazp-slice-job-0-0.mt-05-fp8-atazp:8482
I0420 22:44:07.507192 136755829630784 distributed.py:157] Connecting to JAX distributed service on mt-05-fp8-atazp-slice-job-0-0.mt-05-fp8-atazp:8482
I0420 22:44:08.949751 136755829630784 max_utils.py:284] Jax distributed system initialized!
I0420 22:44:14.860392 136755829630784 max_utils.py:800] System Information: Jax Version: 0.8.1
I0420 22:44:14.860502 136755829630784 max_utils.py:801] System Information: Jaxlib Version: 0.8.1
I0420 22:44:14.860543 136755829630784 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
I0420 22:44:14.860578 136755829630784 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0420 22:44:15.473532 136755829630784 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0420 22:44:15.473845 136755829630784 checkpointing.py:677] Setting up checkpoint logger...
I0420 22:44:15.473905 136755829630784 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0420 22:44:15.473949 136755829630784 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None
I0420 22:44:15.474304 136755829630784 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 0x7c606934f5f0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0420 22:44:18.538209 136755829630784 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0420 22:44:18.538460 136755829630784 checkpoint_manager.py:709] [process=4][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c4e48e92ea0>}, handler_registry=None
I0420 22:44:18.538691 136755829630784 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c4e48e92ea0>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`.
I0420 22:44:18.538758 136755829630784 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c4e4892f7a0>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`.
I0420 22:44:18.538794 136755829630784 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 0x7c4e48e92ea0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c4e48e92ea0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c4e4892f7a0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c4e4892f7a0>}).
I0420 22:44:18.539298 136755829630784 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33
I0420 22:44:18.539379 136755829630784 async_checkpointer.py:177] [process=4][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7c4dae606840> timeout: 600 secs and primary_host=0 for async checkpoint writes
I0420 22:44:20.371127 136755829630784 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints
I0420 22:44:20.815554 136755829630784 checkpoint_manager.py:929] [process=4][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_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7c5f2d7fbd70>
I0420 22:44:20.815761 136755829630784 checkpointing.py:301] Checkpoint manager created!
I0420 22:44:21.906424 136755829630784 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0420 22:44:21.906531 136755829630784 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 22:44:22.280215 136755829630784 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 22:44:22.280308 136755829630784 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 22:44:22.296253 136755829630784 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0420 22:44:22.296314 136755829630784 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0420 22:44:22.355460 136755829630784 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 22:44:22.355554 136755829630784 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 22:44:22.371261 136755829630784 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 22:44:22.371324 136755829630784 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 22:44:22.386791 136755829630784 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0420 22:44:22.386854 136755829630784 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 22:44:22.410654 136755829630784 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0420 22:44:22.410736 136755829630784 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0420 22:44:22.470970 136755829630784 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0420 22:44:22.471057 136755829630784 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0420 22:44:22.936663 136755829630784 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0420 22:44:22.936785 136755829630784 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0420 22:44:24.963041 136755829630784 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:  ()
I0420 22:44:24.963170 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963221 136755829630784 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:  ()
I0420 22:44:24.963262 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963298 136755829630784 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:  ()
I0420 22:44:24.963332 136755829630784 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:  ()
I0420 22:44:24.963365 136755829630784 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:  ()
I0420 22:44:24.963398 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963431 136755829630784 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:  ()
I0420 22:44:24.963462 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963492 136755829630784 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:  ()
I0420 22:44:24.963523 136755829630784 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:  ()
I0420 22:44:24.963555 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963585 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963615 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963643 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963673 136755829630784 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:  ()
I0420 22:44:24.963714 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963746 136755829630784 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:  ()
I0420 22:44:24.963774 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963802 136755829630784 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:  ()
I0420 22:44:24.963829 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963856 136755829630784 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:  ()
I0420 22:44:24.963883 136755829630784 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:  ()
I0420 22:44:24.963911 136755829630784 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:  ()
I0420 22:44:24.963938 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.963965 136755829630784 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:  ()
I0420 22:44:24.963992 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.964020 136755829630784 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:  ()
I0420 22:44:24.964048 136755829630784 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:  ()
I0420 22:44:24.964076 136755829630784 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:  ()
I0420 22:44:24.964103 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.964130 136755829630784 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:  ()
I0420 22:44:24.964156 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.964188 136755829630784 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:  ()
I0420 22:44:24.964216 136755829630784 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:  ()
I0420 22:44:24.964243 136755829630784 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:  ()
I0420 22:44:24.964270 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.964298 136755829630784 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:  ()
I0420 22:44:24.964324 136755829630784 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0420 22:44:24.964351 136755829630784 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:  ()
I0420 22:44:24.964378 136755829630784 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:  ()
I0420 22:44:24.964440 136755829630784 maxtext_utils.py:1620]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   PartitionSpec('norm',)
    Physical:  (None,)
I0420 22:44:24.964511 136755829630784 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)
I0420 22:44:24.964552 136755829630784 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)
I0420 22:44:24.964601 136755829630784 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')
I0420 22:44:24.964647 136755829630784 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)
I0420 22:44:24.964680 136755829630784 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)
I0420 22:44:24.964757 136755829630784 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)
I0420 22:44:24.964811 136755829630784 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')
I0420 22:44:24.964849 136755829630784 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)
I0420 22:44:24.964882 136755829630784 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)
I0420 22:44:24.964929 136755829630784 maxtext_utils.py:1620]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   PartitionSpec('embed', 'vocab')
    Physical:  ('fsdp', None)
I0420 22:44:24.964972 136755829630784 maxtext_utils.py:1620]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   PartitionSpec('vocab', 'embed')
    Physical:  (None, 'fsdp')

I0420 22:44:26.757936 136755829630784 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 22:44:26.758029 136755829630784 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 22:44:26.772792 136755829630784 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0420 22:44:26.772851 136755829630784 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0420 22:44:39.845386 136755829630784 max_utils.py:791] Total memory size: 1.5 GB, Output size: 0.4 GB, Temp size: 1.1 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0420 22:44:39.846306 136755829630784 metric_logger.py:289] number parameters: 1.105 billion
I0420 22:44:54.595384 136755829630784 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0420 22:44:54.763020 136755829630784 checkpointing.py:776] Waited 0.1676163673400879 seconds for step 0 to finish before starting checkpointing.
I0420 22:44:54.765807 136755829630784 checkpoint_manager.py:2013] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 22:44:54.767765 136755829630784 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 0
I0420 22:44:54.770542 136755829630784 async_checkpointer.py:452] [process=4] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints/0.
I0420 22:44:55.381557 136755829630784 signaling_client.py:364] Using JaxDistributedSignalingClient
I0420 22:44:55.382630 136755829630784 jax_array_handlers.py:358] Scheduling D2H of 81 prioritized jax.Array.
I0420 22:44:55.382706 136755829630784 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False
I0420 22:44:55.745926 136755829630784 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.364494s
I0420 22:44:55.746102 136755829630784 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 2.384 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.6471874713897705 s) (per-host)
I0420 22:44:55.746154 136755829630784 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.647249s (batch_requests_ready=0.277439s, total_serialization_initiated=0.369741s, others=0.000070s)
I0420 22:44:55.746265 136755829630784 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.651328s (all_items=0.000019s, per_item={'items': '0.00001907'}, temp_paths=0.651309)
I0420 22:44:55.747242 136632372033280 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started.
I0420 22:44:55.747417 136755829630784 async_checkpointer.py:561] Finished blocking save. Time taken: 0.979578s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints/0.
I0420 22:44:56.198465 136755829630784 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 22:44:56.198861 136630182586112 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0420 22:44:56.199028 136755829630784 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776725094.76579, 'wait_for_prev_duration_secs': 6.175041198730469e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776725094.7678063, 'checkpointer_blocking_duration_secs': 0.9797160625457764, 'get_old_steps_start_time': 1776725095.747544, 'get_old_steps_duration_secs': 2.5510787963867188e-05, 'checkpoint_manager_blocking_start_time': 1776725094.7638907, 'checkpoint_manager_blocking_duration_secs': 1.435096025466919}
I0420 22:44:56.199141 136755829630784 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0420 22:44:56.199202 136755829630784 max_utils.py:750] 
Memstats: After params initialized:
I0420 22:44:56.199260 136755829630784 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_16(process=4,(0,4,0,0))
I0420 22:44:56.199295 136755829630784 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_17(process=4,(1,4,0,0))
I0420 22:44:56.199322 136755829630784 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_20(process=4,(0,5,0,0))
I0420 22:44:56.199346 136755829630784 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_21(process=4,(1,5,0,0))
I0420 22:44:56.465940 136755829630784 metric_logger.py:185] completed step: 0, seconds: 14.749, TFLOP/s/device: 0.921, Tokens/s/device: 138.857, total_weights: 65536, loss: 10.874
I0420 22:44:56.653145 136755829630784 metric_logger.py:185] completed step: 1, seconds: 1.869, TFLOP/s/device: 7.271, Tokens/s/device: 1095.973, total_weights: 65536, loss: 10.874
I0420 22:44:57.059424 136755829630784 metric_logger.py:185] completed step: 2, seconds: 0.030, TFLOP/s/device: 448.420, Tokens/s/device: 67590.759, total_weights: 65536, loss: 10.816
I0420 22:44:57.217641 136755829630784 metric_logger.py:185] completed step: 3, seconds: 0.407, TFLOP/s/device: 33.405, Tokens/s/device: 5035.133, total_weights: 65536, loss: 10.431
I0420 22:44:57.534467 136755829630784 metric_logger.py:185] completed step: 4, seconds: 0.163, TFLOP/s/device: 83.229, Tokens/s/device: 12545.253, total_weights: 65536, loss: 9.992
I0420 22:44:57.540999 136755829630784 metric_logger.py:185] completed step: 5, seconds: 0.158, TFLOP/s/device: 85.897, Tokens/s/device: 12947.275, total_weights: 65536, loss: 9.549
I0420 22:45:17.666483 136755829630784 metric_logger.py:185] completed step: 6, seconds: 0.317, TFLOP/s/device: 42.798, Tokens/s/device: 6451.003, total_weights: 65536, loss: 9.111
I0420 22:45:17.824907 136755829630784 metric_logger.py:185] completed step: 7, seconds: 19.969, TFLOP/s/device: 0.680, Tokens/s/device: 102.560, total_weights: 65536, loss: 8.685
I0420 22:45:17.982969 136755829630784 metric_logger.py:185] completed step: 8, seconds: 0.162, TFLOP/s/device: 83.624, Tokens/s/device: 12604.706, total_weights: 65536, loss: 8.281
I0420 22:45:17.988024 136755829630784 checkpointing.py:772] Waiting for step 10 to finish before checkpoint...
I0420 22:45:18.299157 136755829630784 checkpointing.py:776] Waited 0.3110954761505127 seconds for step 10 to finish before starting checkpointing.
I0420 22:45:18.302230 136755829630784 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 22:45:18.966445    2637 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0420 22:45:21.221341 136633261201152 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints/0/items/array_metadatas/process_4
I0420 22:45:34.610800 136632372033280 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 39.982 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 39.51183247566223 s) (per-host)
I0420 22:45:34.610922 136632372033280 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 38.863567s.
I0420 22:45:44.045846 136632372033280 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 48.298476s.
I0420 22:45:44.046075 136630182586112 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 22:45:44.046133 136630182586112 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0420 22:45:44.046188 136630182586112 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0420 22:45:44.048559 136630182586112 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0420 22:45:44.048730 136755829630784 checkpoint_manager.py:2036] [process=4][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0420 22:45:44.048863 136755829630784 checkpoint_manager.py:1458] Waiting for previous save to complete took 25.746636 seconds. If this number is high, consider checkpointing less frequently.
I0420 22:45:44.050734 136755829630784 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 10
I0420 22:45:44.054089 136755829630784 async_checkpointer.py:452] [process=4] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints/10.
I0420 22:45:45.036415 136755829630784 jax_array_handlers.py:358] Scheduling D2H of 81 prioritized jax.Array.
I0420 22:45:45.036516 136755829630784 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False
I0420 22:45:45.092224 136755829630784 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.057029s
I0420 22:45:45.092391 136755829630784 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 4.961 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.31096434593200684 s) (per-host)
I0420 22:45:45.092440 136755829630784 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.311022s (batch_requests_ready=0.250458s, total_serialization_initiated=0.060498s, others=0.000065s)
I0420 22:45:45.092535 136755829630784 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.315049s (all_items=0.000016s, per_item={'items': '0.00001597'}, temp_paths=0.315033)
I0420 22:45:45.093373 136629685573376 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started.
I0420 22:45:45.093530 136755829630784 async_checkpointer.py:561] Finished blocking save. Time taken: 1.042730s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints/10.
I0420 22:45:45.514084 136755829630784 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=10] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 22:45:45.514472 136630182586112 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0420 22:45:45.514659 136755829630784 standard_logger.py:34] {'step': 10, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776725118.3022027, 'wait_for_prev_duration_secs': 25.746636390686035, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776725144.0507715, 'checkpointer_blocking_duration_secs': 1.0428645610809326, 'get_old_steps_start_time': 1776725145.0936558, 'get_old_steps_duration_secs': 2.6702880859375e-05, 'checkpoint_manager_blocking_start_time': 1776725118.3001204, 'checkpoint_manager_blocking_duration_secs': 27.214500427246094}
I0420 22:45:45.514793 136755829630784 checkpointing.py:408] Started an asynchronous checkpoint save for step 10
I0420 22:45:45.515267 136755829630784 metric_logger.py:185] completed step: 9, seconds: 0.158, TFLOP/s/device: 85.814, Tokens/s/device: 12934.846, total_weights: 65536, loss: 7.908
I0420 22:45:45.521869 136755829630784 metric_logger.py:185] completed step: 10, seconds: 0.158, TFLOP/s/device: 85.931, Tokens/s/device: 12952.434, total_weights: 65536, loss: 7.568
I0420 22:45:45.679003 136755829630784 metric_logger.py:185] completed step: 11, seconds: 27.532, TFLOP/s/device: 0.493, Tokens/s/device: 74.386, total_weights: 65536, loss: 7.288
I0420 22:45:45.837274 136755829630784 metric_logger.py:185] completed step: 12, seconds: 0.006, TFLOP/s/device: 2413.773, Tokens/s/device: 363830.165, total_weights: 65536, loss: 7.032
I0420 22:45:45.995427 136755829630784 metric_logger.py:185] completed step: 13, seconds: 0.158, TFLOP/s/device: 85.931, Tokens/s/device: 12952.434, total_weights: 65536, loss: 6.815
I0420 22:45:46.153616 136755829630784 metric_logger.py:185] completed step: 14, seconds: 0.158, TFLOP/s/device: 85.929, Tokens/s/device: 12952.188, total_weights: 65536, loss: 6.635
I0420 22:45:46.311786 136755829630784 metric_logger.py:185] completed step: 15, seconds: 0.158, TFLOP/s/device: 85.863, Tokens/s/device: 12942.202, total_weights: 65536, loss: 6.492
I0420 22:45:46.470027 136755829630784 metric_logger.py:185] completed step: 16, seconds: 0.158, TFLOP/s/device: 85.947, Tokens/s/device: 12954.892, total_weights: 65536, loss: 6.380
I0420 22:45:46.628277 136755829630784 metric_logger.py:185] completed step: 17, seconds: 0.158, TFLOP/s/device: 85.933, Tokens/s/device: 12952.762, total_weights: 65536, loss: 6.293
I0420 22:45:46.786407 136755829630784 metric_logger.py:185] completed step: 18, seconds: 0.158, TFLOP/s/device: 85.816, Tokens/s/device: 12935.173, total_weights: 65536, loss: 6.223
I0420 22:45:46.944245 136755829630784 checkpointing.py:772] Waiting for step 19 to finish before checkpoint...
I0420 22:45:46.945239 136755829630784 checkpointing.py:776] Waited 0.0010144710540771484 seconds for step 19 to finish before starting checkpointing.
I0420 22:45:46.947261 136755829630784 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=10][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 22:45:51.272834 136631298291456 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints/10/items/array_metadatas/process_4
I0420 22:46:27.570050 136629685573376 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 36.920 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.7885799407959 s) (per-host)
I0420 22:46:27.570155 136629685573376 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.476670s.
I0420 22:46:38.050850 136629685573376 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 52.957349s.
I0420 22:46:38.051135 136630182586112 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 22:46:38.051253 136630182586112 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0420 22:46:38.051301 136630182586112 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=10] CheckpointManager Save Finalize is syncing with other hosts...
I0420 22:46:38.053800 136630182586112 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=10] CheckpointManager Save Finalize is done on all hosts.
I0420 22:46:38.053997 136755829630784 checkpoint_manager.py:2036] [process=4][thread=MainThread][step=10][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=10.
W0420 22:46:38.054127 136755829630784 checkpoint_manager.py:1458] Waiting for previous save to complete took 51.106876 seconds. If this number is high, consider checkpointing less frequently.
I0420 22:46:38.055609 136755829630784 checkpoint_manager.py:1518] [process=4] Saving checkpoint at step 19
I0420 22:46:38.058984 136755829630784 async_checkpointer.py:452] [process=4] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints/19.
I0420 22:46:39.018866 136755829630784 jax_array_handlers.py:358] Scheduling D2H of 81 prioritized jax.Array.
I0420 22:46:39.018968 136755829630784 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False
I0420 22:46:39.075030 136755829630784 base_pytree_checkpoint_handler.py:153] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.057430s
I0420 22:46:39.075192 136755829630784 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/blocking_gbytes_per_sec: 5.106 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.30211448669433594 s) (per-host)
I0420 22:46:39.075245 136755829630784 base_pytree_checkpoint_handler.py:737] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.302172s (batch_requests_ready=0.241281s, total_serialization_initiated=0.060829s, others=0.000062s)
I0420 22:46:39.075343 136755829630784 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.306258s (all_items=0.000011s, per_item={'items': '0.00001073'}, temp_paths=0.306247)
I0420 22:46:39.076187 136631298291456 async_checkpointer.py:79] [process=4][thread=async_save] Background save thread started.
I0420 22:46:39.076344 136755829630784 async_checkpointer.py:561] Finished blocking save. Time taken: 1.020670s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints/19.
I0420 22:46:39.534814 136755829630784 checkpoint_manager.py:1566] [process=4][thread=MainThread][step=19] Starting CheckpointManager Save Finalize thread=save_finalize
I0420 22:46:39.535164 136630182586112 async_checkpointer.py:265] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0420 22:46:39.535332 136755829630784 standard_logger.py:34] {'step': 19, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776725146.9472342, 'wait_for_prev_duration_secs': 51.106876373291016, 'time_between_consecutive_saves_sec': 2.898634672164917, 'checkpointer_blocking_start_time': 1776725198.055648, 'checkpointer_blocking_duration_secs': 1.0208027362823486, 'get_old_steps_start_time': 1776725199.0764651, 'get_old_steps_duration_secs': 1.9550323486328125e-05, 'checkpoint_manager_blocking_start_time': 1776725146.94552, 'checkpoint_manager_blocking_duration_secs': 52.589778661727905}
I0420 22:46:39.535468 136755829630784 checkpointing.py:408] Started an asynchronous checkpoint save for step 19
I0420 22:46:39.535513 136755829630784 checkpoint_manager.py:2024] [process=4][thread=MainThread][step=19][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0420 22:46:45.155054 136630199371520 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260420_223221_05_fp8/checkpoints/19/items/array_metadatas/process_4
I0420 22:47:22.389067 136631298291456 base_pytree_checkpoint_handler.py:129] [process=4] /jax/checkpoint/write/gbytes_per_sec: 36.220 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.61594915390015 s) (per-host)
I0420 22:47:22.389194 136631298291456 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.312891s.
I0420 22:47:37.670375 136631298291456 async_checkpointer.py:144] [process=4][thread=async_save] Background save thread done. Time taken: 58.594059s.
I0420 22:47:37.670585 136630182586112 async_checkpointer.py:273] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0420 22:47:37.670655 136630182586112 async_checkpointer.py:283] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0420 22:47:37.670706 136630182586112 checkpoint_manager.py:2133] [process=4][thread=save_finalize][step=19] CheckpointManager Save Finalize is syncing with other hosts...
I0420 22:47:37.672419 136630182586112 checkpoint_manager.py:2142] [process=4][thread=save_finalize][step=19] CheckpointManager Save Finalize is done on all hosts.
I0420 22:47:37.672592 136755829630784 checkpoint_manager.py:2036] [process=4][thread=MainThread][step=19][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=19.
I0420 22:47:37.672756 136755829630784 checkpoint_manager.py:2013] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0420 22:47:37.673383 136755829630784 metric_logger.py:185] completed step: 19, seconds: 0.158, TFLOP/s/device: 85.807, Tokens/s/device: 12933.784, 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: Mon Apr 20 22:47:49 UTC 2026
EXIT_CODE=0