MaxView

XPK Start: Sat Apr 25 14:10:05 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. 
2026-04-25 14:10:30.391124: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303)
I0425 14:10:30.604357 139321337673536 max_utils.py:273] Attempting to initialize the jax distributed system...
I0425 14:10:39.645333 139321337673536 distributed.py:149] Starting JAX distributed service on [::]:8482
I0425 14:10:39.647670 139321337673536 distributed.py:172] Connecting to JAX distributed service on mt-11-optimizer-offload-bal7b-slice-job-0-0.mt-11-optimizer-offload-bal7b:8482
I0425 14:10:40.674585 139321337673536 max_utils.py:284] Jax distributed system initialized!
I0425 14:10:46.897736 139321337673536 max_utils.py:800] System Information: Jax Version: 0.9.2
I0425 14:10:46.897841 139321337673536 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0425 14:10:46.897882 139321337673536 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
I0425 14:10:46.897920 139321337673536 train_utils.py:391] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0425 14:10:47.615548 139321337673536 maxtext_utils.py:1771] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0425 14:10:47.728738 139321337673536 checkpointing.py:688] Setting up checkpoint logger...
I0425 14:10:47.728859 139321337673536 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0425 14:10:47.728904 139321337673536 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0425 14:10:47.729126 139321337673536 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 0x7eb5a97e6600>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0425 14:10:50.616922 139321337673536 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0425 14:10:50.617180 139321337673536 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 0x7ea1603810d0>}, handler_registry=None
I0425 14:10:50.617429 139321337673536 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ea1603810d0>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`.
I0425 14:10:50.617480 139321337673536 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ea160380cb0>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`.
I0425 14:10:50.617516 139321337673536 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 0x7ea1603810d0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ea1603810d0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ea160380cb0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ea160380cb0>}).
I0425 14:10:50.617849 139321337673536 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0425 14:10:50.617921 139321337673536 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7ea14045b6a0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0425 14:10:51.733744 139321337673536 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260425_121355/nnx_xpk_feat_nnx_post_train_fixes_20260425_121355_11_optimizer_offload_true/checkpoints
I0425 14:10:51.743393 139321337673536 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/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260425_121355/nnx_xpk_feat_nnx_post_train_fixes_20260425_121355_11_optimizer_offload_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7eb48281e1e0>
I0425 14:10:51.743517 139321337673536 checkpointing.py:302] Checkpoint manager created!
I0425 14:10:52.716844 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='count'), GetAttrKey(name='value')) to host
I0425 14:10:52.716959 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='decoder_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to host
I0425 14:10:52.716998 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_0'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717032 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_1'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717063 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wo'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717091 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='post_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to host
I0425 14:10:52.717133 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='pre_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to host
I0425 14:10:52.717158 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='key'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717184 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='out'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717209 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='query'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717234 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='value'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717257 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='logits_dense'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717281 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='token_embedder'), DictKey(key='embedding'), GetAttrKey(name='value')) to host
I0425 14:10:52.717305 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='decoder_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to host
I0425 14:10:52.717330 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_0'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717355 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_1'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717378 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wo'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717402 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='post_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to host
I0425 14:10:52.717425 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='pre_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to host
I0425 14:10:52.717449 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='key'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717472 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='out'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717501 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='query'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717525 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='value'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717547 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='logits_dense'), DictKey(key='kernel'), GetAttrKey(name='value')) to host
I0425 14:10:52.717570 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='token_embedder'), DictKey(key='embedding'), GetAttrKey(name='value')) to host
I0425 14:10:52.717593 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=2), DictKey(key='count'), GetAttrKey(name='value')) to host
I0425 14:10:52.717615 139321337673536 maxtext_utils_nnx.py:115] max_utils.py: Moving (DictKey(key='step'), GetAttrKey(name='value')) to host
I0425 14:10:52.718747 139321337673536 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0425 14:10:52.718803 139321337673536 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0425 14:10:54.163913 139321337673536 maxtext_utils.py:1889]  decoder/decoder_norm/scale/value
    Shape:     float32[2048]
    Physical:  (None,)
I0425 14:10:54.164035 139321337673536 maxtext_utils.py:1889]  decoder/layers/mlp/wi_0/kernel/value
    Shape:     float32[2048,16,7168]
    Physical:  ('fsdp', None, None)
I0425 14:10:54.164086 139321337673536 maxtext_utils.py:1889]  decoder/layers/mlp/wi_1/kernel/value
    Shape:     float32[2048,16,7168]
    Physical:  ('fsdp', None, None)
I0425 14:10:54.164149 139321337673536 maxtext_utils.py:1889]  decoder/layers/mlp/wo/kernel/value
    Shape:     float32[7168,16,2048]
    Physical:  (None, None, 'fsdp')
I0425 14:10:54.164188 139321337673536 maxtext_utils.py:1889]  decoder/layers/post_self_attention_layer_norm/scale/value
    Shape:     float32[2048,16]
    Physical:  (None, None)
I0425 14:10:54.164222 139321337673536 maxtext_utils.py:1889]  decoder/layers/pre_self_attention_layer_norm/scale/value
    Shape:     float32[2048,16]
    Physical:  (None, None)
I0425 14:10:54.164264 139321337673536 maxtext_utils.py:1889]  decoder/layers/self_attention/key/kernel/value
    Shape:     float32[2048,16,16,128]
    Physical:  ('fsdp', None, None, None)
I0425 14:10:54.164301 139321337673536 maxtext_utils.py:1889]  decoder/layers/self_attention/out/kernel/value
    Shape:     float32[16,16,128,2048]
    Physical:  (None, None, None, 'fsdp')
I0425 14:10:54.164336 139321337673536 maxtext_utils.py:1889]  decoder/layers/self_attention/query/kernel/value
    Shape:     float32[2048,16,16,128]
    Physical:  ('fsdp', None, None, None)
I0425 14:10:54.164369 139321337673536 maxtext_utils.py:1889]  decoder/layers/self_attention/value/kernel/value
    Shape:     float32[2048,16,16,128]
    Physical:  ('fsdp', None, None, None)
I0425 14:10:54.164400 139321337673536 maxtext_utils.py:1889]  decoder/logits_dense/kernel/value
    Shape:     float32[2048,32000]
    Physical:  ('fsdp', None)

I0425 14:10:54.164430 139321337673536 maxtext_utils.py:1889]  token_embedder/embedding/value
    Shape:     float32[32000,2048]
    Physical:  (None, 'fsdp')
I0425 14:10:54.576394 139321337673536 nnx_decoders.py:465] nnx_decoders/carry Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0425 14:10:54.576490 139321337673536 nnx_decoders.py:465] nnx_decoders/carry Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 14:10:54.581828 139321337673536 nnx_decoders.py:465] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0425 14:10:54.581886 139321337673536 nnx_decoders.py:465] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 14:10:54.598259 139321337673536 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch_attn', 'activation_length_attn', 'activation_embed_attn').
I0425 14:10:54.598319 139321337673536 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 14:10:54.614039 139321337673536 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch_attn', 'activation_length_attn', 'activation_embed_attn').
I0425 14:10:54.614109 139321337673536 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 14:10:54.639230 139321337673536 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0425 14:10:54.639294 139321337673536 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 14:10:54.661707 139321337673536 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0425 14:10:54.661806 139321337673536 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 14:10:54.677877 139321337673536 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0425 14:10:54.677939 139321337673536 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 14:10:54.708040 139321337673536 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch_attn', 'activation_length_attn', 'activation_heads', 'activation_kv').
I0425 14:10:54.708132 139321337673536 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 14:10:54.728064 139321337673536 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0425 14:10:54.728137 139321337673536 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0425 14:10:54.888614 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='count'), GetAttrKey(name='value')) to device
I0425 14:10:54.888730 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='decoder_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:10:54.888773 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_0'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.888807 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_1'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.888837 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wo'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.888865 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='post_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:10:54.888892 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='pre_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:10:54.888917 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='key'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.888942 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='out'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.888967 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='query'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.888992 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='value'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.889016 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='logits_dense'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.889041 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='token_embedder'), DictKey(key='embedding'), GetAttrKey(name='value')) to device
I0425 14:10:54.889065 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='decoder_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:10:54.889089 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_0'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.889138 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_1'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.889163 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wo'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.889189 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='post_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:10:54.889213 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='pre_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:10:54.889236 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='key'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.889259 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='out'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.889286 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='query'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.889309 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='value'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.889332 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='logits_dense'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:10:54.889354 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='token_embedder'), DictKey(key='embedding'), GetAttrKey(name='value')) to device
I0425 14:10:54.889380 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=2), DictKey(key='count'), GetAttrKey(name='value')) to device
I0425 14:10:54.889405 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='step'), GetAttrKey(name='value')) to device
I0425 14:11:05.914218 139321337673536 max_utils.py:791] Total memory size: 1.3 GB, Output size: 0.1 GB, Temp size: 1.1 GB, Argument size: 0.1 GB, Host temp size: 0.0 GB.
I0425 14:11:05.916486 139321337673536 metric_logger.py:301] number parameters: 1.104 billion
I0425 14:11:06.470823 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='count'), GetAttrKey(name='value')) to device
I0425 14:11:06.470938 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='decoder_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:11:06.470979 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_0'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471012 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_1'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471042 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wo'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471071 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='post_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:11:06.471111 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='pre_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:11:06.471140 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='key'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471166 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='out'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471192 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='query'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471218 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='value'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471243 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='decoder'), DictKey(key='logits_dense'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471270 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='mu'), DictKey(key='token_embedder'), DictKey(key='embedding'), GetAttrKey(name='value')) to device
I0425 14:11:06.471300 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='decoder_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:11:06.471326 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_0'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471350 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wi_1'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471374 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='mlp'), DictKey(key='wo'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471399 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='post_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:11:06.471423 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='pre_self_attention_layer_norm'), DictKey(key='scale'), GetAttrKey(name='value')) to device
I0425 14:11:06.471452 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='key'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471477 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='out'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471501 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='query'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471527 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='layers'), DictKey(key='self_attention'), DictKey(key='value'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471550 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='decoder'), DictKey(key='logits_dense'), DictKey(key='kernel'), GetAttrKey(name='value')) to device
I0425 14:11:06.471573 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=0), DictKey(key='nu'), DictKey(key='token_embedder'), DictKey(key='embedding'), GetAttrKey(name='value')) to device
I0425 14:11:06.471596 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='opt_state'), DictKey(key=2), DictKey(key='count'), GetAttrKey(name='value')) to device
I0425 14:11:06.471618 139321337673536 maxtext_utils_nnx.py:132] max_utils.py: Moving (DictKey(key='step'), GetAttrKey(name='value')) to device
I0425 14:11:17.110632 139321337673536 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0425 14:11:17.279416 139321337673536 checkpointing.py:798] Waited 0.16876959800720215 seconds for step 0 to finish before starting checkpointing.
I0425 14:11:17.281836 139321337673536 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 14:11:17.284025 139321337673536 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0425 14:11:17.285439 139321337673536 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260425_121355/nnx_xpk_feat_nnx_post_train_fixes_20260425_121355_11_optimizer_offload_true/checkpoints/0.
I0425 14:11:17.576319 139321337673536 signaling_client.py:364] Using JaxDistributedSignalingClient
I0425 14:11:17.577384 139321337673536 jax_array_handlers.py:360] Scheduling D2H of 69 prioritized jax.Array.
I0425 14:11:17.577443 139321337673536 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False
I0425 14:11:17.849702 139321337673536 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.274180s
I0425 14:11:17.849880 139321337673536 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.431 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2840268611907959 s) (per-host)
I0425 14:11:17.849932 139321337673536 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.284090s (batch_requests_ready=0.002945s, total_serialization_initiated=0.281074s, others=0.000071s)
I0425 14:11:17.850028 139321337673536 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.288439s (all_items=0.000017s, per_item={'items': '0.00001669'}, temp_paths=0.288422)
I0425 14:11:17.850817 139321337673536 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.57 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260425_121355/nnx_xpk_feat_nnx_post_train_fixes_20260425_121355_11_optimizer_offload_true/checkpoints/0.
I0425 14:11:17.851114 139186560542464 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 14:31:17.851081
I0425 14:11:17.858655 139321337673536 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 14:11:17.858896 139186410284800 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 14:11:17.859015 139321337673536 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260425_121355/nnx_xpk_feat_nnx_post_train_fixes_20260425_121355_11_optimizer_offload_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777126277.281819, 'wait_for_prev_duration_secs': 5.888938903808594e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777126277.2840624, 'checkpointer_blocking_duration_secs': 0.5671627521514893, 'get_old_steps_start_time': 1777126277.851249, 'get_old_steps_duration_secs': 3.0279159545898438e-05, 'checkpoint_manager_blocking_start_time': 1777126277.2798119, 'checkpoint_manager_blocking_duration_secs': 0.5791656970977783}
I0425 14:11:17.859134 139321337673536 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0425 14:11:17.859205 139321337673536 max_utils.py:750] 
Memstats: After params initialized:
I0425 14:11:17.859256 139321337673536 max_utils.py:756] 	Using (GB) 0.17 / 31.25 (0.544000%) on TPU_24(process=6,(0,6,0,0))
I0425 14:11:17.859288 139321337673536 max_utils.py:756] 	Using (GB) 0.17 / 31.25 (0.544000%) on TPU_25(process=6,(1,6,0,0))
I0425 14:11:17.859314 139321337673536 max_utils.py:756] 	Using (GB) 0.17 / 31.25 (0.544000%) on TPU_28(process=6,(0,7,0,0))
I0425 14:11:17.859342 139321337673536 max_utils.py:756] 	Using (GB) 0.17 / 31.25 (0.544000%) on TPU_29(process=6,(1,7,0,0))
I0425 14:11:18.173762 139321337673536 metric_logger.py:196] completed step: 0, seconds: 11.193, TFLOP/s/device: 1.214, Tokens/s/device: 182.971, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52786.469
I0425 14:11:18.359252 139321337673536 metric_logger.py:196] completed step: 1, seconds: 1.061, TFLOP/s/device: 12.804, Tokens/s/device: 1929.953, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52786.469
I0425 14:11:18.780606 139321337673536 metric_logger.py:196] completed step: 2, seconds: 0.025, TFLOP/s/device: 551.941, Tokens/s/device: 83194.540, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28632.133
I0425 14:11:18.943156 139321337673536 metric_logger.py:196] completed step: 3, seconds: 0.410, TFLOP/s/device: 33.128, Tokens/s/device: 4993.453, total_weights: 65536, loss: 9.732, lm_loss: 9.732, perplexity: 16852.779
I0425 14:11:19.269759 139321337673536 metric_logger.py:196] completed step: 4, seconds: 0.180, TFLOP/s/device: 75.376, Tokens/s/device: 11361.430, total_weights: 65536, loss: 9.274, lm_loss: 9.274, perplexity: 10659.797
I0425 14:11:19.277522 139321337673536 metric_logger.py:196] completed step: 5, seconds: 0.162, TFLOP/s/device: 83.685, Tokens/s/device: 12613.944, total_weights: 65536, loss: 8.888, lm_loss: 8.888, perplexity: 7244.862
I0425 14:11:21.016805    2770 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0425 14:11:23.043163 139186418677504 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 69 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260425_121355/nnx_xpk_feat_nnx_post_train_fixes_20260425_121355_11_optimizer_offload_true/checkpoints/0/items/array_metadatas/process_6
I0425 14:11:41.342337 139321337673536 metric_logger.py:196] completed step: 6, seconds: 0.327, TFLOP/s/device: 41.538, Tokens/s/device: 6261.082, total_weights: 65536, loss: 8.589, lm_loss: 8.589, perplexity: 5371.227
I0425 14:11:41.505224 139321337673536 metric_logger.py:196] completed step: 7, seconds: 21.903, TFLOP/s/device: 0.620, Tokens/s/device: 93.503, total_weights: 65536, loss: 8.380, lm_loss: 8.380, perplexity: 4360.796
I0425 14:11:41.668379 139321337673536 metric_logger.py:196] completed step: 8, seconds: 0.169, TFLOP/s/device: 80.358, Tokens/s/device: 12112.395, total_weights: 65536, loss: 8.251, lm_loss: 8.251, perplexity: 3831.198
I0425 14:11:41.831920 139321337673536 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0425 14:11:41.832620 139321337673536 checkpointing.py:798] Waited 0.0007114410400390625 seconds for step 9 to finish before starting checkpointing.
I0425 14:11:41.834832 139321337673536 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 14:11:55.336807 139186560542464 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 41.819 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.770916223526 s) (per-host)
I0425 14:11:55.336931 139186560542464 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.485737s.
I0425 14:12:03.018408 139186560542464 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 45.167196s.
I0425 14:12:03.018627 139186410284800 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 14:12:03.018704 139186410284800 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 14:12:03.018757 139186410284800 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0425 14:12:03.020313 139186410284800 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0425 14:12:03.020473 139321337673536 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.
W0425 14:12:03.020607 139321337673536 checkpoint_manager.py:1452] Waiting for previous save to complete took 21.185773 seconds. If this number is high, consider checkpointing less frequently.
I0425 14:12:03.022444 139321337673536 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0425 14:12:03.024709 139321337673536 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260425_121355/nnx_xpk_feat_nnx_post_train_fixes_20260425_121355_11_optimizer_offload_true/checkpoints/9.
I0425 14:12:03.310211 139321337673536 jax_array_handlers.py:360] Scheduling D2H of 69 prioritized jax.Array.
I0425 14:12:03.310309 139321337673536 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False
I0425 14:12:03.344982 139321337673536 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.036535s
I0425 14:12:03.345136 139321337673536 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 35.503 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.043447256088256836 s) (per-host)
I0425 14:12:03.345184 139321337673536 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.043504s (batch_requests_ready=0.002646s, total_serialization_initiated=0.040795s, others=0.000064s)
I0425 14:12:03.345274 139321337673536 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.047529s (all_items=0.000017s, per_item={'items': '0.00001669'}, temp_paths=0.047513)
I0425 14:12:03.345984 139321337673536 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.32 seconds. Continuing save @ gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260425_121355/nnx_xpk_feat_nnx_post_train_fixes_20260425_121355_11_optimizer_offload_true/checkpoints/9.
I0425 14:12:03.346283 139186410284800 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 14:32:03.346261
I0425 14:12:03.348233 139321337673536 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 14:12:03.348484 139166964750080 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 14:12:03.348647 139321337673536 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260425_121355/nnx_xpk_feat_nnx_post_train_fixes_20260425_121355_11_optimizer_offload_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777126301.8348033, 'wait_for_prev_duration_secs': 21.185773372650146, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777126323.0224845, 'checkpointer_blocking_duration_secs': 0.32389092445373535, 'get_old_steps_start_time': 1777126323.3463988, 'get_old_steps_duration_secs': 2.9802322387695312e-05, 'checkpoint_manager_blocking_start_time': 1777126301.8328555, 'checkpoint_manager_blocking_duration_secs': 21.515756368637085}
I0425 14:12:03.348761 139321337673536 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0425 14:12:03.348805 139321337673536 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 14:12:08.918380 139186418677504 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 69 array_metadata.ArrayMetadata to gs://lance-maxtext/nnx_ckpt_xpk_feat_nnx_post_train_fixes_20260425_121355/nnx_xpk_feat_nnx_post_train_fixes_20260425_121355_11_optimizer_offload_true/checkpoints/9/items/array_metadatas/process_6
I0425 14:12:45.078036 139186410284800 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.809 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.776304960250854 s) (per-host)
I0425 14:12:45.078172 139186410284800 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.731817s.
I0425 14:12:53.212052 139186410284800 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 49.865681s.
I0425 14:12:53.212414 139166964750080 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 14:12:53.212540 139166964750080 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 14:12:53.212588 139166964750080 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0425 14:12:53.213828 139166964750080 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0425 14:12:53.214018 139321337673536 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0425 14:12:53.214224 139321337673536 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 14:12:53.215186 139321337673536 metric_logger.py:196] completed step: 9, seconds: 0.163, TFLOP/s/device: 83.354, Tokens/s/device: 12564.032, total_weights: 65536, loss: 8.175, lm_loss: 8.175, perplexity: 3550.804
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Sat Apr 25 14:13:05 UTC 2026
EXIT_CODE=0