MaxView

XPK Start: Tue Apr 21 06:55:38 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. 
2026-04-21 06:56:03.406619: 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)
I0421 06:56:03.619498 137223643490112 max_utils.py:273] Attempting to initialize the jax distributed system...
I0421 06:56:12.661905 137223643490112 distributed.py:149] Starting JAX distributed service on [::]:8482
I0421 06:56:12.664291 137223643490112 distributed.py:172] Connecting to JAX distributed service on mt-11-optimizer-offload-5oq4c-slice-job-0-0.mt-11-optimizer-offload-5oq4c:8482
I0421 06:56:13.723879 137223643490112 max_utils.py:284] Jax distributed system initialized!
I0421 06:56:19.716351 137223643490112 max_utils.py:800] System Information: Jax Version: 0.9.2
I0421 06:56:19.716460 137223643490112 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0421 06:56:19.716501 137223643490112 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
I0421 06:56:19.716537 137223643490112 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0421 06:56:20.417844 137223643490112 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 06:56:20.418131 137223643490112 checkpointing.py:677] Setting up checkpoint logger...
I0421 06:56:20.418185 137223643490112 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0421 06:56:20.418229 137223643490112 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0421 06:56:20.418563 137223643490112 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 0x7ccd2a1366c0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0421 06:56:23.415407 137223643490112 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0421 06:56:23.415651 137223643490112 checkpoint_manager.py:708] [process=7][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7cb9080d5f40>}, handler_registry=None
I0421 06:56:23.415904 137223643490112 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7cb9080d5f40>` 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`.
I0421 06:56:23.415955 137223643490112 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7cb9080d53d0>` 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`.
I0421 06:56:23.415992 137223643490112 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 0x7cb9080d5f40>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7cb9080d5f40>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7cb9080d53d0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7cb9080d53d0>}).
I0421 06:56:23.416353 137223643490112 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0421 06:56:23.416432 137223643490112 async_checkpointer.py:192] [process=7][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7cb8d86d1440> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0421 06:56:24.880344 137223643490112 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_11_optimizer_offload_true/checkpoints
I0421 06:56:24.967668 137223643490112 checkpoint_manager.py:929] [process=7][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_main_20260421_061359/linen_xpk_main_20260421_061359_11_optimizer_offload_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7cb9080d4e90>
I0421 06:56:24.967863 137223643490112 checkpointing.py:301] Checkpoint manager created!
I0421 06:56:25.893095 137223643490112 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0421 06:56:25.893215 137223643490112 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 06:56:26.273572 137223643490112 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 06:56:26.273667 137223643490112 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 06:56:26.290310 137223643490112 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 06:56:26.290374 137223643490112 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 06:56:26.314515 137223643490112 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 06:56:26.314582 137223643490112 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 06:56:26.331070 137223643490112 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 06:56:26.331133 137223643490112 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 06:56:26.347763 137223643490112 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 06:56:26.347824 137223643490112 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 06:56:26.372965 137223643490112 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0421 06:56:26.373037 137223643490112 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 06:56:26.394144 137223643490112 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0421 06:56:26.394209 137223643490112 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0421 06:56:26.606376 137223643490112 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0421 06:56:26.606487 137223643490112 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0421 06:56:28.212905 137223643490112 maxtext_utils.py:1668]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 06:56:28.213039 137223643490112 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)
I0421 06:56:28.213095 137223643490112 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)
I0421 06:56:28.213153 137223643490112 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')
I0421 06:56:28.213205 137223643490112 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)
I0421 06:56:28.213242 137223643490112 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)
I0421 06:56:28.213297 137223643490112 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)
I0421 06:56:28.213347 137223643490112 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')
I0421 06:56:28.213387 137223643490112 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)
I0421 06:56:28.213423 137223643490112 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)
I0421 06:56:28.213469 137223643490112 maxtext_utils.py:1668]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0421 06:56:28.213514 137223643490112 maxtext_utils.py:1668]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0421 06:56:28.714046 137223643490112 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 06:56:28.714143 137223643490112 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 06:56:28.729505 137223643490112 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 06:56:28.729567 137223643490112 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 06:56:39.585554 137223643490112 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.
I0421 06:56:39.586382 137223643490112 metric_logger.py:301] number parameters: 1.104 billion
I0421 06:56:51.360411 137223643490112 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0421 06:56:51.528840 137223643490112 checkpointing.py:776] Waited 0.1684110164642334 seconds for step 0 to finish before starting checkpointing.
I0421 06:56:51.531227 137223643490112 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 06:56:51.532899 137223643490112 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 0
I0421 06:56:51.534296 137223643490112 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_11_optimizer_offload_true/checkpoints/0.
I0421 06:56:52.303595 137223643490112 signaling_client.py:364] Using JaxDistributedSignalingClient
I0421 06:56:52.304692 137223643490112 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 06:56:52.304765 137223643490112 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
I0421 06:56:52.575052 137223643490112 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.271531s
I0421 06:56:52.575230 137223643490112 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 5.548 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2780416011810303 s) (per-host)
I0421 06:56:52.575284 137223643490112 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.278106s (batch_requests_ready=0.002215s, total_serialization_initiated=0.275817s, others=0.000073s)
I0421 06:56:52.575381 137223643490112 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.282200s (all_items=0.000018s, per_item={'items': '0.00001788'}, temp_paths=0.282182)
I0421 06:56:52.576186 137223643490112 event_tracking.py:125] [process=7] [async] Finished blocking save in 1.04 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_11_optimizer_offload_true/checkpoints/0.
I0421 06:56:52.576518 137088670344960 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 07:16:52.576479
I0421 06:56:52.583221 137223643490112 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 06:56:52.583470 137088260245248 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save.
I0421 06:56:52.583621 137223643490112 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_11_optimizer_offload_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776754611.5312085, 'wait_for_prev_duration_secs': 6.198883056640625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776754611.5329382, 'checkpointer_blocking_duration_secs': 1.0437374114990234, 'get_old_steps_start_time': 1776754612.5767012, 'get_old_steps_duration_secs': 4.2438507080078125e-05, 'checkpoint_manager_blocking_start_time': 1776754611.5293765, 'checkpoint_manager_blocking_duration_secs': 1.0542078018188477}
I0421 06:56:52.583727 137223643490112 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0421 06:56:52.583795 137223643490112 max_utils.py:750] 
Memstats: After params initialized:
I0421 06:56:52.583845 137223643490112 max_utils.py:756] 	Using (GB) 0.17 / 31.25 (0.544000%) on TPU_26(process=7,(2,6,0,0))
I0421 06:56:52.583878 137223643490112 max_utils.py:756] 	Using (GB) 0.17 / 31.25 (0.544000%) on TPU_27(process=7,(3,6,0,0))
I0421 06:56:52.583905 137223643490112 max_utils.py:756] 	Using (GB) 0.17 / 31.25 (0.544000%) on TPU_30(process=7,(2,7,0,0))
I0421 06:56:52.583931 137223643490112 max_utils.py:756] 	Using (GB) 0.17 / 31.25 (0.544000%) on TPU_31(process=7,(3,7,0,0))
I0421 06:56:52.893120 137223643490112 metric_logger.py:196] completed step: 0, seconds: 11.774, TFLOP/s/device: 1.154, Tokens/s/device: 173.944, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0421 06:56:53.100726 137223643490112 metric_logger.py:196] completed step: 1, seconds: 1.531, TFLOP/s/device: 8.874, Tokens/s/device: 1337.513, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0421 06:56:53.522831 137223643490112 metric_logger.py:196] completed step: 2, seconds: 0.045, TFLOP/s/device: 302.892, Tokens/s/device: 45655.179, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0421 06:56:53.685482 137223643490112 metric_logger.py:196] completed step: 3, seconds: 0.422, TFLOP/s/device: 32.175, Tokens/s/device: 4849.805, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.047
I0421 06:56:54.011580 137223643490112 metric_logger.py:196] completed step: 4, seconds: 0.169, TFLOP/s/device: 80.607, Tokens/s/device: 12149.976, total_weights: 65536, loss: 9.278, lm_loss: 9.278, perplexity: 10694.968
I0421 06:56:54.017573 137223643490112 metric_logger.py:196] completed step: 5, seconds: 0.162, TFLOP/s/device: 83.618, Tokens/s/device: 12603.775, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.520
I0421 06:56:56.275436    2782 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0421 06:56:58.293253 137088268637952 array_metadata_store.py:203] [process=7][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_11_optimizer_offload_true/checkpoints/0/items/array_metadatas/process_7
I0421 06:57:16.068508 137223643490112 metric_logger.py:196] completed step: 6, seconds: 0.327, TFLOP/s/device: 41.582, Tokens/s/device: 6267.693, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.688
I0421 06:57:16.231259 137223643490112 metric_logger.py:196] completed step: 7, seconds: 21.889, TFLOP/s/device: 0.621, Tokens/s/device: 93.564, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4377.010
I0421 06:57:16.394099 137223643490112 metric_logger.py:196] completed step: 8, seconds: 0.168, TFLOP/s/device: 81.007, Tokens/s/device: 12210.318, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3845.943
I0421 06:57:16.556265 137223643490112 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0421 06:57:16.556970 137223643490112 checkpointing.py:776] Waited 0.0007271766662597656 seconds for step 9 to finish before starting checkpointing.
I0421 06:57:16.558944 137223643490112 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 06:57:30.279133 137088670344960 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 41.586 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.98190689086914 s) (per-host)
I0421 06:57:30.279263 137088670344960 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.702631s.
I0421 06:57:38.986404 137088670344960 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 46.409756s.
I0421 06:57:38.986695 137088260245248 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 06:57:38.986859 137088260245248 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save.
I0421 06:57:38.986914 137088260245248 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0421 06:57:38.989192 137088260245248 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0421 06:57:38.989375 137223643490112 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0421 06:57:38.989514 137223643490112 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.430570 seconds. If this number is high, consider checkpointing less frequently.
I0421 06:57:38.991424 137223643490112 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 9
I0421 06:57:38.993618 137223643490112 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_11_optimizer_offload_true/checkpoints/9.
I0421 06:57:39.708776 137223643490112 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 06:57:39.708870 137223643490112 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
I0421 06:57:39.747412 137223643490112 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.039615s
I0421 06:57:39.747580 137223643490112 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 35.864 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04301047325134277 s) (per-host)
I0421 06:57:39.747627 137223643490112 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.043067s (batch_requests_ready=0.001814s, total_serialization_initiated=0.041189s, others=0.000064s)
I0421 06:57:39.747712 137223643490112 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.047303s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.047288)
I0421 06:57:39.748452 137223643490112 event_tracking.py:125] [process=7] [async] Finished blocking save in 0.76 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_11_optimizer_offload_true/checkpoints/9.
I0421 06:57:39.748763 137088260245248 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 07:17:39.748714
I0421 06:57:39.750850 137223643490112 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 06:57:39.751058 137069041018624 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save.
I0421 06:57:39.751175 137223643490112 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_11_optimizer_offload_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776754636.558914, 'wait_for_prev_duration_secs': 22.430570363998413, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776754658.9914637, 'checkpointer_blocking_duration_secs': 0.7574396133422852, 'get_old_steps_start_time': 1776754659.748926, 'get_old_steps_duration_secs': 2.8848648071289062e-05, 'checkpoint_manager_blocking_start_time': 1776754636.5572085, 'checkpoint_manager_blocking_duration_secs': 23.193934440612793}
I0421 06:57:39.751277 137223643490112 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0421 06:57:39.751321 137223643490112 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 06:57:45.678504 137088268637952 array_metadata_store.py:203] [process=7][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260421_061359/linen_xpk_main_20260421_061359_11_optimizer_offload_true/checkpoints/9/items/array_metadatas/process_7
I0421 06:58:21.396255 137088260245248 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 37.886 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.69164681434631 s) (per-host)
I0421 06:58:21.396386 137088260245248 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.647508s.
I0421 06:58:29.999509 137088260245248 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 50.250617s.
I0421 06:58:29.999723 137069041018624 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 06:58:29.999807 137069041018624 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save.
I0421 06:58:29.999863 137069041018624 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0421 06:58:30.002551 137069041018624 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0421 06:58:30.002798 137223643490112 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0421 06:58:30.002943 137223643490112 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 06:58:30.003921 137223643490112 metric_logger.py:196] completed step: 9, seconds: 0.163, TFLOP/s/device: 83.410, Tokens/s/device: 12572.439, total_weights: 65536, loss: 8.179, lm_loss: 8.179, perplexity: 3564.739
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Tue Apr 21 06:58:38 UTC 2026
EXIT_CODE=0