MaxView

Timing Test Reference Updates
← Back to run

Log Summary

XPK Start: Sun Apr 19 06:45:58 UTC 2026
2026-04-19 06:46:02.383299: 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 00:00:1776581162.396092      10 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1776581162.400004      10 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
W0000 00:00:1776581162.411316      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776581162.411334      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776581162.411336      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1776581162.411338      10 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
2026-04-19 06:46:22.368196: 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)
I0419 06:46:22.896133 139392962934592 max_utils.py:273] Attempting to initialize the jax distributed system...
INFO:2026-04-19 06:46:31,936:jax._src.distributed:140: Starting JAX distributed service on [::]:8482
I0419 06:46:31.936468 139392962934592 distributed.py:140] Starting JAX distributed service on [::]:8482
INFO:2026-04-19 06:46:31,938:jax._src.distributed:157: Connecting to JAX distributed service on mt-05-fp8-k2410-slice-job-0-0.mt-05-fp8-k2410:8482
I0419 06:46:31.938854 139392962934592 distributed.py:157] Connecting to JAX distributed service on mt-05-fp8-k2410-slice-job-0-0.mt-05-fp8-k2410:8482
I0419 06:46:33.791563 139392962934592 max_utils.py:284] Jax distributed system initialized!
I0419 06:46:40.692539 139392962934592 max_utils.py:800] System Information: Jax Version: 0.8.1
I0419 06:46:40.692648 139392962934592 max_utils.py:801] System Information: Jaxlib Version: 0.8.1
I0419 06:46:40.692690 139392962934592 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
I0419 06:46:40.692727 139392962934592 train_utils.py:347] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0419 06:46:41.326196 139392962934592 maxtext_utils.py:1517] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0419 06:46:41.326511 139392962934592 checkpointing.py:677] Setting up checkpoint logger...
I0419 06:46:41.326577 139392962934592 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0419 06:46:41.326623 139392962934592 pytree_checkpoint_handler.py:589] save_device_host_concurrent_bytes=None
I0419 06:46:41.326980 139392962934592 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 0x7ec66ab751c0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0419 06:46:44.165741 139392962934592 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0419 06:46:44.166090 139392962934592 checkpoint_manager.py:709] [process=6][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7eb369268260>}, handler_registry=None
I0419 06:46:44.166337 139392962934592 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7eb369268260>` 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`.
I0419 06:46:44.166388 139392962934592 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7eb36926b7a0>` 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`.
I0419 06:46:44.166424 139392962934592 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 0x7eb369268260>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7eb369268260>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7eb36926b7a0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7eb36926b7a0>}).
I0419 06:46:44.166939 139392962934592 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.33
I0419 06:46:44.167019 139392962934592 async_checkpointer.py:177] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7eb1e8752840> timeout: 600 secs and primary_host=0 for async checkpoint writes
I0419 06:46:46.092608 139392962934592 checkpoint_manager.py:1818] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints
I0419 06:46:47.430414 139392962934592 checkpoint_manager.py:929] [process=6][thread=MainThread] CheckpointManager created,  primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=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_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7eb369268e00>
I0419 06:46:47.430611 139392962934592 checkpointing.py:301] Checkpoint manager created!
I0419 06:46:48.511743 139392962934592 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0419 06:46:48.511849 139392962934592 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 06:46:48.891273 139392962934592 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0419 06:46:48.891370 139392962934592 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 06:46:48.907113 139392962934592 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0419 06:46:48.907175 139392962934592 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0419 06:46:48.966450 139392962934592 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0419 06:46:48.966552 139392962934592 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 06:46:48.982327 139392962934592 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0419 06:46:48.982389 139392962934592 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 06:46:48.998136 139392962934592 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0419 06:46:48.998200 139392962934592 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 06:46:49.022013 139392962934592 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0419 06:46:49.022083 139392962934592 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0419 06:46:49.081840 139392962934592 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0419 06:46:49.081928 139392962934592 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0419 06:46:49.548747 139392962934592 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0419 06:46:49.548864 139392962934592 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0419 06:46:51.614166 139392962934592 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:  ()
I0419 06:46:51.614312 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614368 139392962934592 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:  ()
I0419 06:46:51.614409 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614444 139392962934592 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:  ()
I0419 06:46:51.614478 139392962934592 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:  ()
I0419 06:46:51.614511 139392962934592 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:  ()
I0419 06:46:51.614547 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614580 139392962934592 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:  ()
I0419 06:46:51.614611 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614642 139392962934592 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:  ()
I0419 06:46:51.614673 139392962934592 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:  ()
I0419 06:46:51.614704 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614735 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614765 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614794 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614825 139392962934592 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:  ()
I0419 06:46:51.614854 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614883 139392962934592 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:  ()
I0419 06:46:51.614911 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614939 139392962934592 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:  ()
I0419 06:46:51.614965 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.614992 139392962934592 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:  ()
I0419 06:46:51.615020 139392962934592 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:  ()
I0419 06:46:51.615046 139392962934592 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:  ()
I0419 06:46:51.615073 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.615100 139392962934592 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:  ()
I0419 06:46:51.615126 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.615152 139392962934592 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:  ()
I0419 06:46:51.615179 139392962934592 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:  ()
I0419 06:46:51.615206 139392962934592 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:  ()
I0419 06:46:51.615246 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.615288 139392962934592 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:  ()
I0419 06:46:51.615321 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.615350 139392962934592 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:  ()
I0419 06:46:51.615378 139392962934592 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:  ()
I0419 06:46:51.615405 139392962934592 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:  ()
I0419 06:46:51.615432 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.615459 139392962934592 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:  ()
I0419 06:46:51.615486 139392962934592 maxtext_utils.py:1620]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   PartitionSpec()
    Physical:  ()
I0419 06:46:51.615512 139392962934592 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:  ()
I0419 06:46:51.615544 139392962934592 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:  ()
I0419 06:46:51.615596 139392962934592 maxtext_utils.py:1620]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   PartitionSpec('norm',)
    Physical:  (None,)
I0419 06:46:51.615653 139392962934592 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)
I0419 06:46:51.615692 139392962934592 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)
I0419 06:46:51.615742 139392962934592 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')
I0419 06:46:51.615789 139392962934592 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)
I0419 06:46:51.615823 139392962934592 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)
I0419 06:46:51.615871 139392962934592 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)
I0419 06:46:51.615920 139392962934592 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')
I0419 06:46:51.615958 139392962934592 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)
I0419 06:46:51.615993 139392962934592 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)
I0419 06:46:51.616040 139392962934592 maxtext_utils.py:1620]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   PartitionSpec('embed', 'vocab')
    Physical:  ('fsdp', None)
I0419 06:46:51.616086 139392962934592 maxtext_utils.py:1620]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   PartitionSpec('vocab', 'embed')
    Physical:  (None, 'fsdp')

I0419 06:46:53.413369 139392962934592 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0419 06:46:53.413464 139392962934592 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0419 06:46:53.428563 139392962934592 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0419 06:46:53.428625 139392962934592 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0419 06:47:07.022708 139392962934592 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.
I0419 06:47:07.023719 139392962934592 metric_logger.py:289] number parameters: 1.105 billion
I0419 06:47:22.162426 139392962934592 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0419 06:47:22.330069 139392962934592 checkpointing.py:776] Waited 0.16762185096740723 seconds for step 0 to finish before starting checkpointing.
I0419 06:47:22.333048 139392962934592 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0419 06:47:22.334952 139392962934592 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 0
I0419 06:47:22.337841 139392962934592 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints/0.
I0419 06:47:24.247772 139392962934592 signaling_client.py:364] Using JaxDistributedSignalingClient
I0419 06:47:24.248843 139392962934592 jax_array_handlers.py:358] Scheduling D2H of 81 prioritized jax.Array.
I0419 06:47:24.248900 139392962934592 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
I0419 06:47:24.612752 139392962934592 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.365100s
I0419 06:47:24.612927 139392962934592 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 1.385 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 1.1140635013580322 s) (per-host)
I0419 06:47:24.612982 139392962934592 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 1.114128s (batch_requests_ready=0.743641s, total_serialization_initiated=0.370415s, others=0.000072s)
I0419 06:47:24.613095 139392962934592 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 1.118149s (all_items=0.000019s, per_item={'items': '0.00001860'}, temp_paths=1.118131)
I0419 06:47:24.614022 139269859428096 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started.
I0419 06:47:24.614200 139392962934592 async_checkpointer.py:561] Finished blocking save. Time taken: 2.279178s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints/0.
I0419 06:47:24.616131 139392962934592 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0419 06:47:24.616450 139262334842624 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0419 06:47:24.616612 139392962934592 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776581242.3330288, 'wait_for_prev_duration_secs': 6.556510925292969e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776581242.3349922, 'checkpointer_blocking_duration_secs': 2.2793281078338623, 'get_old_steps_start_time': 1776581244.6143417, 'get_old_steps_duration_secs': 2.6464462280273438e-05, 'checkpoint_manager_blocking_start_time': 1776581242.3308365, 'checkpoint_manager_blocking_duration_secs': 2.2857351303100586}
I0419 06:47:24.616724 139392962934592 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0419 06:47:24.616776 139392962934592 max_utils.py:750] 
Memstats: After params initialized:
I0419 06:47:24.616825 139392962934592 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_24(process=6,(0,6,0,0))
I0419 06:47:24.616859 139392962934592 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_25(process=6,(1,6,0,0))
I0419 06:47:24.616885 139392962934592 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_28(process=6,(0,7,0,0))
I0419 06:47:24.616909 139392962934592 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_29(process=6,(1,7,0,0))
I0419 06:47:24.882615 139392962934592 metric_logger.py:185] completed step: 0, seconds: 15.139, TFLOP/s/device: 0.898, Tokens/s/device: 135.283, total_weights: 65536, loss: 10.874
I0419 06:47:25.065984 139392962934592 metric_logger.py:185] completed step: 1, seconds: 2.718, TFLOP/s/device: 4.999, Tokens/s/device: 753.433, total_weights: 65536, loss: 10.874
I0419 06:47:25.468971 139392962934592 metric_logger.py:185] completed step: 2, seconds: 0.026, TFLOP/s/device: 518.514, Tokens/s/device: 78156.007, total_weights: 65536, loss: 10.816
I0419 06:47:25.627373 139392962934592 metric_logger.py:185] completed step: 3, seconds: 0.404, TFLOP/s/device: 33.662, Tokens/s/device: 5073.966, total_weights: 65536, loss: 10.431
I0419 06:47:25.944772 139392962934592 metric_logger.py:185] completed step: 4, seconds: 0.163, TFLOP/s/device: 83.152, Tokens/s/device: 12533.583, total_weights: 65536, loss: 9.992
I0419 06:47:25.950986 139392962934592 metric_logger.py:185] completed step: 5, seconds: 0.158, TFLOP/s/device: 85.763, Tokens/s/device: 12927.171, total_weights: 65536, loss: 9.549
I0419 06:47:47.483362 139392962934592 metric_logger.py:185] completed step: 6, seconds: 0.318, TFLOP/s/device: 42.772, Tokens/s/device: 6447.084, total_weights: 65536, loss: 9.111
I0419 06:47:47.641793 139392962934592 metric_logger.py:185] completed step: 7, seconds: 21.375, TFLOP/s/device: 0.636, Tokens/s/device: 95.812, total_weights: 65536, loss: 8.685
I0419 06:47:47.800150 139392962934592 metric_logger.py:185] completed step: 8, seconds: 0.163, TFLOP/s/device: 83.331, Tokens/s/device: 12560.564, total_weights: 65536, loss: 8.281
I0419 06:47:47.805013 139392962934592 checkpointing.py:772] Waiting for step 10 to finish before checkpoint...
I0419 06:47:48.116552 139392962934592 checkpointing.py:776] Waited 0.31150293350219727 seconds for step 10 to finish before starting checkpointing.
I0419 06:47:48.119268 139392962934592 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0419 06:47:48.623683    2707 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0419 06:47:51.307319 139269867820800 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints/0/items/array_metadatas/process_6
I0419 06:48:06.134445 139269859428096 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 37.053 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.63551712036133 s) (per-host)
I0419 06:48:06.134574 139269859428096 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.520437s.
I0419 06:48:15.053819 139269859428096 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 50.439665s.
I0419 06:48:15.054115 139262334842624 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0419 06:48:15.054255 139262334842624 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0419 06:48:15.054308 139262334842624 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0419 06:48:15.055755 139262334842624 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0419 06:48:15.055950 139392962934592 checkpoint_manager.py:2036] [process=6][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0419 06:48:15.056087 139392962934592 checkpoint_manager.py:1458] Waiting for previous save to complete took 26.936834 seconds. If this number is high, consider checkpointing less frequently.
I0419 06:48:15.057943 139392962934592 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 10
I0419 06:48:15.061351 139392962934592 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints/10.
I0419 06:48:16.024254 139392962934592 jax_array_handlers.py:358] Scheduling D2H of 81 prioritized jax.Array.
I0419 06:48:16.024358 139392962934592 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
I0419 06:48:16.078967 139392962934592 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.055896s
I0419 06:48:16.079139 139392962934592 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 4.946 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3119316101074219 s) (per-host)
I0419 06:48:16.079193 139392962934592 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.311995s (batch_requests_ready=0.252490s, total_serialization_initiated=0.059433s, others=0.000071s)
I0419 06:48:16.079314 139392962934592 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.316129s (all_items=0.000017s, per_item={'items': '0.00001693'}, temp_paths=0.316112)
I0419 06:48:16.080314 139263416977152 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started.
I0419 06:48:16.080475 139392962934592 async_checkpointer.py:561] Finished blocking save. Time taken: 1.022460s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints/10.
I0419 06:48:16.530017 139392962934592 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=10] Starting CheckpointManager Save Finalize thread=save_finalize
I0419 06:48:16.530484 139262334842624 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0419 06:48:16.530672 139392962934592 standard_logger.py:34] {'step': 10, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776581268.1192267, 'wait_for_prev_duration_secs': 26.93683433532715, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776581295.0579817, 'checkpointer_blocking_duration_secs': 1.0226025581359863, 'get_old_steps_start_time': 1776581296.0806074, 'get_old_steps_duration_secs': 2.6464462280273438e-05, 'checkpoint_manager_blocking_start_time': 1776581268.1173382, 'checkpoint_manager_blocking_duration_secs': 28.41329526901245}
I0419 06:48:16.530789 139392962934592 checkpointing.py:408] Started an asynchronous checkpoint save for step 10
I0419 06:48:16.531285 139392962934592 metric_logger.py:185] completed step: 9, seconds: 0.158, TFLOP/s/device: 85.834, Tokens/s/device: 12937.787, total_weights: 65536, loss: 7.908
I0419 06:48:16.538161 139392962934592 metric_logger.py:185] completed step: 10, seconds: 0.158, TFLOP/s/device: 85.884, Tokens/s/device: 12945.311, total_weights: 65536, loss: 7.568
I0419 06:48:16.696008 139392962934592 metric_logger.py:185] completed step: 11, seconds: 28.731, TFLOP/s/device: 0.473, Tokens/s/device: 71.281, total_weights: 65536, loss: 7.288
I0419 06:48:16.854107 139392962934592 metric_logger.py:185] completed step: 12, seconds: 0.006, TFLOP/s/device: 2342.205, Tokens/s/device: 353042.579, total_weights: 65536, loss: 7.032
I0419 06:48:17.012583 139392962934592 metric_logger.py:185] completed step: 13, seconds: 0.159, TFLOP/s/device: 85.446, Tokens/s/device: 12879.288, total_weights: 65536, loss: 6.815
I0419 06:48:17.170769 139392962934592 metric_logger.py:185] completed step: 14, seconds: 0.158, TFLOP/s/device: 85.961, Tokens/s/device: 12957.023, total_weights: 65536, loss: 6.635
I0419 06:48:17.329008 139392962934592 metric_logger.py:185] completed step: 15, seconds: 0.158, TFLOP/s/device: 85.754, Tokens/s/device: 12925.702, total_weights: 65536, loss: 6.492
I0419 06:48:17.487283 139392962934592 metric_logger.py:185] completed step: 16, seconds: 0.158, TFLOP/s/device: 85.854, Tokens/s/device: 12940.894, total_weights: 65536, loss: 6.380
I0419 06:48:17.645659 139392962934592 metric_logger.py:185] completed step: 17, seconds: 0.158, TFLOP/s/device: 85.940, Tokens/s/device: 12953.827, total_weights: 65536, loss: 6.293
I0419 06:48:17.804044 139392962934592 metric_logger.py:185] completed step: 18, seconds: 0.158, TFLOP/s/device: 85.860, Tokens/s/device: 12941.712, total_weights: 65536, loss: 6.223
I0419 06:48:17.962022 139392962934592 checkpointing.py:772] Waiting for step 19 to finish before checkpoint...
I0419 06:48:17.962933 139392962934592 checkpointing.py:776] Waited 0.0009312629699707031 seconds for step 19 to finish before starting checkpointing.
I0419 06:48:17.964935 139392962934592 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=10][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0419 06:48:22.842297 139263433762560 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints/10/items/array_metadatas/process_6
I0419 06:48:59.313278 139263416977152 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 36.278 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.54598927497864 s) (per-host)
I0419 06:48:59.313401 139263416977152 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.232976s.
I0419 06:49:06.502607 139263416977152 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 50.422166s.
I0419 06:49:06.502910 139262334842624 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0419 06:49:06.503042 139262334842624 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0419 06:49:06.503090 139262334842624 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=10] CheckpointManager Save Finalize is syncing with other hosts...
I0419 06:49:06.504719 139262334842624 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=10] CheckpointManager Save Finalize is done on all hosts.
I0419 06:49:06.504904 139392962934592 checkpoint_manager.py:2036] [process=6][thread=MainThread][step=10][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=10.
W0419 06:49:06.505023 139392962934592 checkpoint_manager.py:1458] Waiting for previous save to complete took 48.540099 seconds. If this number is high, consider checkpointing less frequently.
I0419 06:49:06.506557 139392962934592 checkpoint_manager.py:1518] [process=6] Saving checkpoint at step 19
I0419 06:49:06.510061 139392962934592 async_checkpointer.py:452] [process=6] Started async saving checkpoint to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints/19.
I0419 06:49:07.478784 139392962934592 jax_array_handlers.py:358] Scheduling D2H of 81 prioritized jax.Array.
I0419 06:49:07.478885 139392962934592 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
I0419 06:49:07.534732 139392962934592 base_pytree_checkpoint_handler.py:153] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.056900s
I0419 06:49:07.534921 139392962934592 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/blocking_gbytes_per_sec: 4.927 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.313122034072876 s) (per-host)
I0419 06:49:07.534999 139392962934592 base_pytree_checkpoint_handler.py:737] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.313214s (batch_requests_ready=0.252730s, total_serialization_initiated=0.060385s, others=0.000099s)
I0419 06:49:07.535162 139392962934592 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.317571s (all_items=0.000011s, per_item={'items': '0.00001073'}, temp_paths=0.317561)
I0419 06:49:07.536074 139263433762560 async_checkpointer.py:79] [process=6][thread=async_save] Background save thread started.
I0419 06:49:07.536250 139392962934592 async_checkpointer.py:561] Finished blocking save. Time taken: 1.029603s. Continuing background save to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints/19.
I0419 06:49:07.971282 139392962934592 checkpoint_manager.py:1566] [process=6][thread=MainThread][step=19] Starting CheckpointManager Save Finalize thread=save_finalize
I0419 06:49:07.971641 139262334842624 async_checkpointer.py:265] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0419 06:49:07.971811 139392962934592 standard_logger.py:34] {'step': 19, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776581297.9649074, 'wait_for_prev_duration_secs': 48.540099143981934, 'time_between_consecutive_saves_sec': 2.909113645553589, 'checkpointer_blocking_start_time': 1776581346.5065956, 'checkpointer_blocking_duration_secs': 1.029771327972412, 'get_old_steps_start_time': 1776581347.5363796, 'get_old_steps_duration_secs': 2.3126602172851562e-05, 'checkpoint_manager_blocking_start_time': 1776581297.9632533, 'checkpoint_manager_blocking_duration_secs': 50.00852131843567}
I0419 06:49:07.971924 139392962934592 checkpointing.py:408] Started an asynchronous checkpoint save for step 19
I0419 06:49:07.971968 139392962934592 checkpoint_manager.py:2024] [process=6][thread=MainThread][step=19][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0419 06:49:13.850672 139261831542528 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615/linen_xpk_feat_Migrate_Decoder_And_Tests_to_NNX_20260419_063615_05_fp8/checkpoints/19/items/array_metadatas/process_6
I0419 06:49:50.550682 139263433762560 base_pytree_checkpoint_handler.py:129] [process=6] /jax/checkpoint/write/gbytes_per_sec: 36.460 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.32884907722473 s) (per-host)
I0419 06:49:50.550798 139263433762560 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.014610s.
I0419 06:49:58.141967 139263433762560 async_checkpointer.py:144] [process=6][thread=async_save] Background save thread done. Time taken: 50.605764s.
I0419 06:49:58.142321 139262334842624 async_checkpointer.py:273] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0419 06:49:58.142462 139262334842624 async_checkpointer.py:283] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0419 06:49:58.142524 139262334842624 checkpoint_manager.py:2133] [process=6][thread=save_finalize][step=19] CheckpointManager Save Finalize is syncing with other hosts...
I0419 06:49:58.143756 139262334842624 checkpoint_manager.py:2142] [process=6][thread=save_finalize][step=19] CheckpointManager Save Finalize is done on all hosts.
I0419 06:49:58.143920 139392962934592 checkpoint_manager.py:2036] [process=6][thread=MainThread][step=19][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=19.
I0419 06:49:58.144061 139392962934592 checkpoint_manager.py:2013] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0419 06:49:58.144673 139392962934592 metric_logger.py:185] completed step: 19, seconds: 0.159, TFLOP/s/device: 85.721, Tokens/s/device: 12920.810, 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: Sun Apr 19 06:50:08 UTC 2026
EXIT_CODE=0