MaxView

XPK Start: Fri Apr 24 20:20:43 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
`rope_parameters`'s factor field must be a float >= 1, got 40
`rope_parameters`'s beta_fast field must be a float, got 32
`rope_parameters`'s beta_slow field must be a float, got 1
DeepseekV32Config got `key=rope_scaling` in kwargs but hasn't set it as attribute. For RoPE standardization you need to set `self.rope_parameters` in model's config. 
2026-04-24 20:21:08.867468: 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)
I0424 20:21:09.078346 133399450851136 max_utils.py:273] Attempting to initialize the jax distributed system...
I0424 20:21:18.119958 133399450851136 distributed.py:149] Starting JAX distributed service on [::]:8482
I0424 20:21:18.122326 133399450851136 distributed.py:172] Connecting to JAX distributed service on mt-05-fp8-mul1n-slice-job-0-0.mt-05-fp8-mul1n:8482
I0424 20:21:19.483538 133399450851136 max_utils.py:284] Jax distributed system initialized!
I0424 20:21:25.568406 133399450851136 max_utils.py:800] System Information: Jax Version: 0.9.2
I0424 20:21:25.568511 133399450851136 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0424 20:21:25.568554 133399450851136 max_utils.py:802] System Information: Jax Backend: PJRT C API
TFRT TPU v6 lite
Built on Mar 4 2026 11:32:08 (1772652728) cl/878335365
I0424 20:21:25.568590 133399450851136 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0424 20:21:26.279306 133399450851136 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0424 20:21:26.279591 133399450851136 checkpointing.py:677] Setting up checkpoint logger...
I0424 20:21:26.279647 133399450851136 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0424 20:21:26.279691 133399450851136 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0424 20:21:26.280034 133399450851136 base_pytree_checkpoint_handler.py:441] Created BasePyTreeCheckpointHandler: use_ocdbt=True, use_zarr3=True, pytree_metadata_options=PyTreeMetadataOptions(support_rich_types=False), array_metadata_store=<orbax.checkpoint._src.metadata.array_metadata_store.Store object at 0x7952c363c1d0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0424 20:21:29.279608 133399450851136 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0424 20:21:29.279835 133399450851136 checkpoint_manager.py:708] [process=6][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x793ec86ac380>}, handler_registry=None
I0424 20:21:29.280069 133399450851136 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x793ec86ac380>` 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`.
I0424 20:21:29.280134 133399450851136 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x793ec86af860>` 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`.
I0424 20:21:29.280170 133399450851136 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 0x793ec86ac380>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x793ec86ac380>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x793ec86af860>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x793ec86af860>}).
I0424 20:21:29.280487 133399450851136 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0424 20:21:29.280558 133399450851136 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x793ea07a32e0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0424 20:21:30.460777 133399450851136 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints
I0424 20:21:30.921244 133399450851136 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_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x793ef00f70e0>
I0424 20:21:30.921412 133399450851136 checkpointing.py:301] Checkpoint manager created!
I0424 20:21:32.036186 133399450851136 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0424 20:21:32.036309 133399450851136 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 20:21:32.427778 133399450851136 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0424 20:21:32.427870 133399450851136 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 20:21:32.444793 133399450851136 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0424 20:21:32.444854 133399450851136 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 20:21:32.498048 133399450851136 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 20:21:32.498146 133399450851136 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 20:21:32.515058 133399450851136 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 20:21:32.515135 133399450851136 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 20:21:32.531884 133399450851136 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 20:21:32.531945 133399450851136 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 20:21:32.557740 133399450851136 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0424 20:21:32.557810 133399450851136 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 20:21:32.609912 133399450851136 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0424 20:21:32.609992 133399450851136 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0424 20:21:33.047421 133399450851136 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0424 20:21:33.047540 133399450851136 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0424 20:21:34.995794 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.995916 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.995964 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996002 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996042 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996076 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_0/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996119 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996153 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996184 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996212 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996241 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996269 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wi_1/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996301 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996338 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996368 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996395 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996421 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996450 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/mlp/wo/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996477 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996510 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996538 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996565 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996592 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996619 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/key/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996645 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996671 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996697 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996723 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996749 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996775 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/out/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996801 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996828 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996854 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996880 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996906 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996931 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/query/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996957 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/input_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.996983 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/input_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.997009 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/kernel_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.997035 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/kernel_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.997062 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/output_grad_amax_history
    Shape:     float32[16,1024]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.997090 133399450851136 maxtext_utils.py:1668]  params/_overwrite_with_gradient/decoder/layers/self_attention/value/Fp8DirectDotGeneralOp_0/output_grad_scale
    Shape:     float32[16,1]
    Logical:   P()
    Physical:  ()
I0424 20:21:34.997178 133399450851136 maxtext_utils.py:1668]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0424 20:21:34.997243 133399450851136 maxtext_utils.py:1668]  params/params/decoder/layers/mlp/wi_0/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0424 20:21:34.997284 133399450851136 maxtext_utils.py:1668]  params/params/decoder/layers/mlp/wi_1/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0424 20:21:34.997332 133399450851136 maxtext_utils.py:1668]  params/params/decoder/layers/mlp/wo/kernel
    Shape:     float32[7168,16,2048]
    Logical:   P('mlp', 'layers', 'embed')
    Physical:  (None, None, 'fsdp')
I0424 20:21:34.997380 133399450851136 maxtext_utils.py:1668]  params/params/decoder/layers/post_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0424 20:21:34.997414 133399450851136 maxtext_utils.py:1668]  params/params/decoder/layers/pre_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0424 20:21:34.997461 133399450851136 maxtext_utils.py:1668]  params/params/decoder/layers/self_attention/key/kernel
    Shape:     float32[2048,16,16,128]
    Logical:   P('embed', 'layers', 'kv_heads', 'kv_head_dim')
    Physical:  ('fsdp', None, None, None)
I0424 20:21:34.997513 133399450851136 maxtext_utils.py:1668]  params/params/decoder/layers/self_attention/out/kernel
    Shape:     float32[16,16,128,2048]
    Logical:   P('heads', 'layers', 'kv', 'embed')
    Physical:  (None, None, None, 'fsdp')
I0424 20:21:34.997550 133399450851136 maxtext_utils.py:1668]  params/params/decoder/layers/self_attention/query/kernel
    Shape:     float32[2048,16,16,128]
    Logical:   P('embed', 'layers', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None, None)
I0424 20:21:34.997583 133399450851136 maxtext_utils.py:1668]  params/params/decoder/layers/self_attention/value/kernel
    Shape:     float32[2048,16,16,128]
    Logical:   P('embed', 'layers', 'kv_heads', 'kv_head_dim')
    Physical:  ('fsdp', None, None, None)
I0424 20:21:34.997628 133399450851136 maxtext_utils.py:1668]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0424 20:21:34.997671 133399450851136 maxtext_utils.py:1668]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0424 20:21:36.624632 133399450851136 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0424 20:21:36.624725 133399450851136 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0424 20:21:36.640441 133399450851136 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0424 20:21:36.640505 133399450851136 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0424 20:21:50.465475 133399450851136 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.
I0424 20:21:50.466376 133399450851136 metric_logger.py:301] number parameters: 1.105 billion
I0424 20:22:05.618892 133399450851136 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0424 20:22:05.789836 133399450851136 checkpointing.py:776] Waited 0.17092680931091309 seconds for step 0 to finish before starting checkpointing.
I0424 20:22:05.792761 133399450851136 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0424 20:22:05.794667 133399450851136 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0424 20:22:05.796004 133399450851136 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints/0.
I0424 20:22:06.533004 133399450851136 signaling_client.py:364] Using JaxDistributedSignalingClient
I0424 20:22:06.534060 133399450851136 jax_array_handlers.py:360] Scheduling D2H of 81 prioritized jax.Array.
I0424 20:22:06.534132 133399450851136 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
I0424 20:22:06.960053 133399450851136 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.427168s
I0424 20:22:06.960244 133399450851136 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 3.542 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.4355638027191162 s) (per-host)
I0424 20:22:06.960299 133399450851136 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.435630s (batch_requests_ready=0.003590s, total_serialization_initiated=0.431965s, others=0.000075s)
I0424 20:22:06.960408 133399450851136 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.439687s (all_items=0.000018s, per_item={'items': '0.00001812'}, temp_paths=0.439669)
I0424 20:22:06.961222 133399450851136 event_tracking.py:125] [process=6] [async] Finished blocking save in 1.17 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints/0.
I0424 20:22:06.961499 133269461415680 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 20:42:06.961474
I0424 20:22:06.963517 133399450851136 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 20:22:06.963827 133268916152064 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0424 20:22:06.963983 133399450851136 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777062125.7927318, 'wait_for_prev_duration_secs': 0.00011181831359863281, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777062125.7947063, 'checkpointer_blocking_duration_secs': 1.166905164718628, 'get_old_steps_start_time': 1777062126.961635, 'get_old_steps_duration_secs': 3.170967102050781e-05, 'checkpoint_manager_blocking_start_time': 1777062125.7904503, 'checkpoint_manager_blocking_duration_secs': 1.1734938621520996}
I0424 20:22:06.964124 133399450851136 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0424 20:22:06.964180 133399450851136 max_utils.py:750] 
Memstats: After params initialized:
I0424 20:22:06.964232 133399450851136 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_24(process=6,(0,6,0,0))
I0424 20:22:06.964268 133399450851136 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_25(process=6,(1,6,0,0))
I0424 20:22:06.964294 133399450851136 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_28(process=6,(0,7,0,0))
I0424 20:22:06.964321 133399450851136 max_utils.py:756] 	Using (GB) 0.45 / 31.25 (1.440000%) on TPU_29(process=6,(1,7,0,0))
I0424 20:22:07.277013 133399450851136 metric_logger.py:196] completed step: 0, seconds: 15.152, TFLOP/s/device: 0.897, Tokens/s/device: 135.160, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52776.805
I0424 20:22:07.466977 133399450851136 metric_logger.py:196] completed step: 1, seconds: 1.656, TFLOP/s/device: 8.204, Tokens/s/device: 1236.652, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52761.840
I0424 20:22:07.871748 133399450851136 metric_logger.py:196] completed step: 2, seconds: 0.033, TFLOP/s/device: 416.706, Tokens/s/device: 62810.526, total_weights: 65536, loss: 10.816, lm_loss: 10.816, perplexity: 49832.398
I0424 20:22:08.029888 133399450851136 metric_logger.py:196] completed step: 3, seconds: 0.405, TFLOP/s/device: 33.546, Tokens/s/device: 5056.466, total_weights: 65536, loss: 10.431, lm_loss: 10.431, perplexity: 33901.820
I0424 20:22:08.347088 133399450851136 metric_logger.py:196] completed step: 4, seconds: 0.164, TFLOP/s/device: 83.004, Tokens/s/device: 12511.302, total_weights: 65536, loss: 9.992, lm_loss: 9.992, perplexity: 21847.639
I0424 20:22:08.353549 133399450851136 metric_logger.py:196] completed step: 5, seconds: 0.158, TFLOP/s/device: 85.906, Tokens/s/device: 12948.667, total_weights: 65536, loss: 9.549, lm_loss: 9.549, perplexity: 14035.006
I0424 20:22:10.978151    2827 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0424 20:22:13.910620 133268924544768 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints/0/items/array_metadatas/process_6
I0424 20:22:27.391769 133269461415680 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 75.706 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 20.867056608200073 s) (per-host)
I0424 20:22:27.391876 133269461415680 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 20.430296s.
I0424 20:22:32.790662 133399450851136 metric_logger.py:196] completed step: 6, seconds: 0.318, TFLOP/s/device: 42.750, Tokens/s/device: 6443.717, total_weights: 65536, loss: 9.111, lm_loss: 9.111, perplexity: 9058.688
I0424 20:22:32.949203 133399450851136 metric_logger.py:196] completed step: 7, seconds: 24.280, TFLOP/s/device: 0.560, Tokens/s/device: 84.348, total_weights: 65536, loss: 8.685, lm_loss: 8.685, perplexity: 5914.720
I0424 20:22:33.107906 133399450851136 metric_logger.py:196] completed step: 8, seconds: 0.163, TFLOP/s/device: 83.238, Tokens/s/device: 12546.559, total_weights: 65536, loss: 8.281, lm_loss: 8.281, perplexity: 3950.010
I0424 20:22:33.112963 133399450851136 checkpointing.py:772] Waiting for step 10 to finish before checkpoint...
I0424 20:22:33.424191 133399450851136 checkpointing.py:776] Waited 0.3111910820007324 seconds for step 10 to finish before starting checkpointing.
I0424 20:22:33.426794 133399450851136 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 20:22:34.462032 133269461415680 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 27.500435s.
I0424 20:22:34.462328 133268916152064 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 20:22:34.462457 133268916152064 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0424 20:22:34.462511 133268916152064 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0424 20:22:34.465119 133268916152064 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0424 20:22:34.465288 133399450851136 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0424 20:22:34.465418 133399450851136 checkpoint_manager.py:1452] Waiting for previous save to complete took 1.038623 seconds. If this number is high, consider checkpointing less frequently.
I0424 20:22:34.467226 133399450851136 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 10
I0424 20:22:34.469269 133399450851136 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints/10.
I0424 20:22:34.761580 133399450851136 jax_array_handlers.py:360] Scheduling D2H of 81 prioritized jax.Array.
I0424 20:22:34.761673 133399450851136 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
I0424 20:22:34.815748 133399450851136 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.055169s
I0424 20:22:34.815921 133399450851136 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 25.148 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.06134605407714844 s) (per-host)
I0424 20:22:34.815972 133399450851136 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.061407s (batch_requests_ready=0.003045s, total_serialization_initiated=0.058294s, others=0.000068s)
I0424 20:22:34.816079 133399450851136 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.065532s (all_items=0.000015s, per_item={'items': '0.00001526'}, temp_paths=0.065517)
I0424 20:22:34.816762 133399450851136 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.35 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints/10.
I0424 20:22:34.817065 133266761824000 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 20:42:34.817029
I0424 20:22:34.819124 133399450851136 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=10] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 20:22:34.819450 133268916152064 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0424 20:22:34.819614 133399450851136 standard_logger.py:34] {'step': 10, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777062153.426764, 'wait_for_prev_duration_secs': 1.038623332977295, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777062154.4672642, 'checkpointer_blocking_duration_secs': 0.3499631881713867, 'get_old_steps_start_time': 1777062154.8172498, 'get_old_steps_duration_secs': 3.314018249511719e-05, 'checkpoint_manager_blocking_start_time': 1777062153.424947, 'checkpoint_manager_blocking_duration_secs': 1.394629716873169}
I0424 20:22:34.819731 133399450851136 checkpointing.py:408] Started an asynchronous checkpoint save for step 10
I0424 20:22:34.820407 133399450851136 metric_logger.py:196] completed step: 9, seconds: 0.158, TFLOP/s/device: 85.854, Tokens/s/device: 12940.894, total_weights: 65536, loss: 7.908, lm_loss: 7.908, perplexity: 2720.116
I0424 20:22:34.827937 133399450851136 metric_logger.py:196] completed step: 10, seconds: 0.159, TFLOP/s/device: 85.593, Tokens/s/device: 12901.438, total_weights: 65536, loss: 7.568, lm_loss: 7.568, perplexity: 1936.198
I0424 20:22:34.985662 133399450851136 metric_logger.py:196] completed step: 11, seconds: 1.713, TFLOP/s/device: 7.933, Tokens/s/device: 1195.813, total_weights: 65536, loss: 7.288, lm_loss: 7.288, perplexity: 1463.272
I0424 20:22:35.549528 133399450851136 metric_logger.py:196] completed step: 12, seconds: 0.006, TFLOP/s/device: 2098.074, Tokens/s/device: 316244.595, total_weights: 65536, loss: 7.032, lm_loss: 7.032, perplexity: 1132.061
I0424 20:22:35.707961 133399450851136 metric_logger.py:196] completed step: 13, seconds: 0.559, TFLOP/s/device: 24.289, Tokens/s/device: 3661.137, total_weights: 65536, loss: 6.815, lm_loss: 6.815, perplexity: 911.031
I0424 20:22:35.866212 133399450851136 metric_logger.py:196] completed step: 14, seconds: 0.163, TFLOP/s/device: 83.261, Tokens/s/device: 12549.942, total_weights: 65536, loss: 6.635, lm_loss: 6.635, perplexity: 761.080
I0424 20:22:36.024532 133399450851136 metric_logger.py:196] completed step: 15, seconds: 0.158, TFLOP/s/device: 85.850, Tokens/s/device: 12940.240, total_weights: 65536, loss: 6.492, lm_loss: 6.492, perplexity: 659.915
I0424 20:22:36.182788 133399450851136 metric_logger.py:196] completed step: 16, seconds: 0.158, TFLOP/s/device: 85.766, Tokens/s/device: 12927.579, total_weights: 65536, loss: 6.380, lm_loss: 6.380, perplexity: 589.948
I0424 20:22:36.341303 133399450851136 metric_logger.py:196] completed step: 17, seconds: 0.158, TFLOP/s/device: 85.816, Tokens/s/device: 12935.091, total_weights: 65536, loss: 6.293, lm_loss: 6.293, perplexity: 540.941
I0424 20:22:36.499318 133399450851136 metric_logger.py:196] completed step: 18, seconds: 0.158, TFLOP/s/device: 85.891, Tokens/s/device: 12946.375, total_weights: 65536, loss: 6.223, lm_loss: 6.223, perplexity: 504.108
I0424 20:22:36.656977 133399450851136 checkpointing.py:772] Waiting for step 19 to finish before checkpoint...
I0424 20:22:36.658023 133399450851136 checkpointing.py:776] Waited 0.0010640621185302734 seconds for step 19 to finish before starting checkpointing.
I0424 20:22:36.660228 133399450851136 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=10][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 20:22:41.739168 133266258523904 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints/10/items/array_metadatas/process_6
I0424 20:23:17.802677 133266761824000 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 36.698 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.04805135726929 s) (per-host)
I0424 20:23:17.802803 133266761824000 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.985609s.
I0424 20:23:26.359084 133266761824000 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 51.541873s.
I0424 20:23:26.359377 133268916152064 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 20:23:26.359494 133268916152064 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0424 20:23:26.359542 133268916152064 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=10] CheckpointManager Save Finalize is syncing with other hosts...
I0424 20:23:26.360976 133268916152064 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=10] CheckpointManager Save Finalize is done on all hosts.
I0424 20:23:26.361184 133399450851136 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=10][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=10.
W0424 20:23:26.361313 133399450851136 checkpoint_manager.py:1452] Waiting for previous save to complete took 49.701088 seconds. If this number is high, consider checkpointing less frequently.
I0424 20:23:26.362888 133399450851136 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 19
I0424 20:23:26.364948 133399450851136 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints/19.
I0424 20:23:26.653214 133399450851136 jax_array_handlers.py:360] Scheduling D2H of 81 prioritized jax.Array.
I0424 20:23:26.653314 133399450851136 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
I0424 20:23:26.708698 133399450851136 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.056554s
I0424 20:23:26.708868 133399450851136 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 24.636 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.0626215934753418 s) (per-host)
I0424 20:23:26.708919 133399450851136 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.062683s (batch_requests_ready=0.002990s, total_serialization_initiated=0.059627s, others=0.000066s)
I0424 20:23:26.709023 133399450851136 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.066804s (all_items=0.000012s, per_item={'items': '0.00001192'}, temp_paths=0.066792)
I0424 20:23:26.709760 133399450851136 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.35 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints/19.
I0424 20:23:26.710039 133268916152064 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 20:43:26.710007
I0424 20:23:26.711982 133399450851136 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=19] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 20:23:26.712252 133266258523904 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0424 20:23:26.712393 133399450851136 standard_logger.py:34] {'step': 19, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777062156.6601987, 'wait_for_prev_duration_secs': 49.701087951660156, 'time_between_consecutive_saves_sec': 2.1950347423553467, 'checkpointer_blocking_start_time': 1777062206.3629305, 'checkpointer_blocking_duration_secs': 0.3472270965576172, 'get_old_steps_start_time': 1777062206.7101781, 'get_old_steps_duration_secs': 2.5510787963867188e-05, 'checkpoint_manager_blocking_start_time': 1777062156.6583219, 'checkpoint_manager_blocking_duration_secs': 50.05404186248779}
I0424 20:23:26.712555 133399450851136 checkpointing.py:408] Started an asynchronous checkpoint save for step 19
I0424 20:23:26.712601 133399450851136 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=19][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 20:23:31.792961 133267274209024 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 81 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_05_fp8/checkpoints/19/items/array_metadatas/process_6
I0424 20:24:08.505091 133268916152064 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.740 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.85879850387573 s) (per-host)
I0424 20:24:08.505235 133268916152064 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.794531s.
I0424 20:24:16.251711 133268916152064 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 49.540988s.
I0424 20:24:16.252019 133266258523904 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 20:24:16.252172 133266258523904 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0424 20:24:16.252217 133266258523904 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=19] CheckpointManager Save Finalize is syncing with other hosts...
I0424 20:24:16.254501 133266258523904 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=19] CheckpointManager Save Finalize is done on all hosts.
I0424 20:24:16.254697 133399450851136 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=19][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=19.
I0424 20:24:16.254844 133399450851136 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0424 20:24:16.255947 133399450851136 metric_logger.py:196] completed step: 19, seconds: 0.159, TFLOP/s/device: 85.673, Tokens/s/device: 12913.640, total_weights: 65536, loss: 6.167, lm_loss: 6.167, perplexity: 476.947
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Fri Apr 24 20:24:29 UTC 2026
EXIT_CODE=0