MaxView

XPK Start: Tue Apr 21 14:47:53 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 14:48:18.552416: 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 14:48:18.769442 133585637513024 max_utils.py:273] Attempting to initialize the jax distributed system...
I0421 14:48:27.813090 133585637513024 distributed.py:149] Starting JAX distributed service on [::]:8482
I0421 14:48:27.815465 133585637513024 distributed.py:172] Connecting to JAX distributed service on mt-02-synthetic-ufrgb-slice-job-0-0.mt-02-synthetic-ufrgb:8482
I0421 14:48:28.937397 133585637513024 max_utils.py:284] Jax distributed system initialized!
I0421 14:48:34.973464 133585637513024 max_utils.py:800] System Information: Jax Version: 0.9.2
I0421 14:48:34.973568 133585637513024 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0421 14:48:34.973607 133585637513024 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 14:48:34.973642 133585637513024 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0421 14:48:35.665494 133585637513024 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 14:48:35.665785 133585637513024 checkpointing.py:677] Setting up checkpoint logger...
I0421 14:48:35.665841 133585637513024 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0421 14:48:35.665885 133585637513024 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0421 14:48:35.666232 133585637513024 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 0x797e1cd3ef60>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0421 14:48:38.515976 133585637513024 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0421 14:48:38.516216 133585637513024 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 0x797d108ef6e0>}, handler_registry=None
I0421 14:48:38.516453 133585637513024 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x797d108ef6e0>` 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 14:48:38.516501 133585637513024 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7969d84bb0e0>` 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 14:48:38.516538 133585637513024 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 0x797d108ef6e0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x797d108ef6e0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7969d84bb0e0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7969d84bb0e0>}).
I0421 14:48:38.516872 133585637513024 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0421 14:48:38.516952 133585637513024 async_checkpointer.py:192] [process=7][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7969d05a5580> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0421 14:48:39.711640 133585637513024 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_02_synthetic/checkpoints
I0421 14:48:39.831439 133585637513024 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_144122/linen_xpk_main_20260421_144122_02_synthetic/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7969d84bb2c0>
I0421 14:48:39.831613 133585637513024 checkpointing.py:301] Checkpoint manager created!
I0421 14:48:40.766490 133585637513024 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0421 14:48:40.766609 133585637513024 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 14:48:41.148565 133585637513024 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 14:48:41.148659 133585637513024 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 14:48:41.165541 133585637513024 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 14:48:41.165604 133585637513024 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 14:48:41.189843 133585637513024 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 14:48:41.189912 133585637513024 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 14:48:41.206927 133585637513024 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 14:48:41.206997 133585637513024 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 14:48:41.223883 133585637513024 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 14:48:41.223945 133585637513024 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 14:48:41.249097 133585637513024 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0421 14:48:41.249170 133585637513024 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 14:48:41.270726 133585637513024 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0421 14:48:41.270811 133585637513024 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0421 14:48:41.486792 133585637513024 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0421 14:48:41.486902 133585637513024 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0421 14:48:42.955872 133585637513024 maxtext_utils.py:1668]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 14:48:42.956006 133585637513024 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 14:48:42.956061 133585637513024 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 14:48:42.956121 133585637513024 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 14:48:42.956175 133585637513024 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 14:48:42.956216 133585637513024 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 14:48:42.956272 133585637513024 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 14:48:42.956326 133585637513024 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 14:48:42.956368 133585637513024 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 14:48:42.956407 133585637513024 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 14:48:42.956456 133585637513024 maxtext_utils.py:1668]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)
I0421 14:48:42.956507 133585637513024 maxtext_utils.py:1668]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')

I0421 14:48:43.462333 133585637513024 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 14:48:43.462431 133585637513024 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 14:48:43.478411 133585637513024 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 14:48:43.478472 133585637513024 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 14:48:54.335057 133585637513024 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.
I0421 14:48:54.335913 133585637513024 metric_logger.py:301] number parameters: 1.104 billion
I0421 14:49:05.929189 133585637513024 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0421 14:49:06.083487 133585637513024 checkpointing.py:776] Waited 0.15427803993225098 seconds for step 0 to finish before starting checkpointing.
I0421 14:49:06.086859 133585637513024 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 14:49:06.088403 133585637513024 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 0
I0421 14:49:06.089910 133585637513024 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_02_synthetic/checkpoints/0.
I0421 14:49:06.812948 133585637513024 signaling_client.py:364] Using JaxDistributedSignalingClient
I0421 14:49:06.813907 133585637513024 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 14:49:06.813965 133585637513024 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 14:49:07.094064 133585637513024 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.281190s
I0421 14:49:07.094244 133585637513024 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 5.380 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2867166996002197 s) (per-host)
I0421 14:49:07.094302 133585637513024 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.286786s (batch_requests_ready=0.002276s, total_serialization_initiated=0.284429s, others=0.000080s)
I0421 14:49:07.094396 133585637513024 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.290940s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.290922)
I0421 14:49:07.095265 133585637513024 event_tracking.py:125] [process=7] [async] Finished blocking save in 1.01 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_02_synthetic/checkpoints/0.
I0421 14:49:07.095583 133457045968640 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 15:09:07.095545
I0421 14:49:07.102358 133585637513024 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 14:49:07.102647 133454309549824 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save.
I0421 14:49:07.102799 133585637513024 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776782946.086841, 'wait_for_prev_duration_secs': 6.151199340820312e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776782946.0884404, 'checkpointer_blocking_duration_secs': 1.0073130130767822, 'get_old_steps_start_time': 1776782947.095785, 'get_old_steps_duration_secs': 3.147125244140625e-05, 'checkpoint_manager_blocking_start_time': 1776782946.0840657, 'checkpoint_manager_blocking_duration_secs': 1.018692970275879}
I0421 14:49:07.102957 133585637513024 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0421 14:49:07.103012 133585637513024 max_utils.py:750] 
Memstats: After params initialized:
I0421 14:49:07.103066 133585637513024 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_26(process=7,(2,6,0,0))
I0421 14:49:07.103099 133585637513024 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_27(process=7,(3,6,0,0))
I0421 14:49:07.103127 133585637513024 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_30(process=7,(2,7,0,0))
I0421 14:49:07.103156 133585637513024 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_31(process=7,(3,7,0,0))
I0421 14:49:07.418003 133585637513024 metric_logger.py:196] completed step: 0, seconds: 11.593, TFLOP/s/device: 1.172, Tokens/s/device: 176.656, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0421 14:49:07.592093 133585637513024 metric_logger.py:196] completed step: 1, seconds: 1.487, TFLOP/s/device: 9.135, Tokens/s/device: 1376.999, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0421 14:49:07.979470 133585637513024 metric_logger.py:196] completed step: 2, seconds: 0.027, TFLOP/s/device: 504.629, Tokens/s/device: 76063.138, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0421 14:49:08.126631 133585637513024 metric_logger.py:196] completed step: 3, seconds: 0.387, TFLOP/s/device: 35.083, Tokens/s/device: 5288.150, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0421 14:49:08.421584 133585637513024 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.682, Tokens/s/device: 13367.186, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0421 14:49:08.427518 133585637513024 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.331, Tokens/s/device: 13917.204, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0421 14:49:09.963426    2860 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0421 14:49:12.361339 133454317942528 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_144122/linen_xpk_main_20260421_144122_02_synthetic/checkpoints/0/items/array_metadatas/process_7
I0421 14:49:30.531263 133585637513024 metric_logger.py:196] completed step: 6, seconds: 0.295, TFLOP/s/device: 45.984, Tokens/s/device: 6931.212, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0421 14:49:30.678677 133585637513024 metric_logger.py:196] completed step: 7, seconds: 21.957, TFLOP/s/device: 0.619, Tokens/s/device: 93.273, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0421 14:49:30.826013 133585637513024 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.204, Tokens/s/device: 13445.731, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0421 14:49:30.972643 133585637513024 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0421 14:49:30.973324 133585637513024 checkpointing.py:776] Waited 0.0007007122039794922 seconds for step 9 to finish before starting checkpointing.
I0421 14:49:30.975317 133585637513024 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 14:49:43.850977 133457045968640 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 42.642 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.0434193611145 s) (per-host)
I0421 14:49:43.851097 133457045968640 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.755397s.
I0421 14:49:52.476866 133457045968640 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 45.381158s.
I0421 14:49:52.477124 133454309549824 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 14:49:52.477250 133454309549824 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save.
I0421 14:49:52.477301 133454309549824 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0421 14:49:52.479651 133454309549824 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0421 14:49:52.479832 133585637513024 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 14:49:52.479968 133585637513024 checkpoint_manager.py:1452] Waiting for previous save to complete took 21.504651 seconds. If this number is high, consider checkpointing less frequently.
I0421 14:49:52.481770 133585637513024 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 9
I0421 14:49:52.483665 133585637513024 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_02_synthetic/checkpoints/9.
I0421 14:49:53.190039 133585637513024 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0421 14:49:53.190134 133585637513024 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 14:49:53.231528 133585637513024 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.042572s
I0421 14:49:53.231693 133585637513024 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 33.003 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04674029350280762 s) (per-host)
I0421 14:49:53.231757 133585637513024 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.046813s (batch_requests_ready=0.002509s, total_serialization_initiated=0.044224s, others=0.000081s)
I0421 14:49:53.231848 133585637513024 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.051133s (all_items=0.000017s, per_item={'items': '0.00001669'}, temp_paths=0.051117)
I0421 14:49:53.232562 133585637513024 event_tracking.py:125] [process=7] [async] Finished blocking save in 0.75 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_02_synthetic/checkpoints/9.
I0421 14:49:53.232847 133454309549824 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 15:09:53.232815
I0421 14:49:53.234827 133585637513024 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 14:49:53.235066 133432050956032 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save.
I0421 14:49:53.235165 133585637513024 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_02_synthetic/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776782970.9752853, 'wait_for_prev_duration_secs': 21.504651308059692, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776782992.481809, 'checkpointer_blocking_duration_secs': 0.7511813640594482, 'get_old_steps_start_time': 1776782993.2330148, 'get_old_steps_duration_secs': 3.123283386230469e-05, 'checkpoint_manager_blocking_start_time': 1776782970.9735534, 'checkpoint_manager_blocking_duration_secs': 22.261578798294067}
I0421 14:49:53.235310 133585637513024 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0421 14:49:53.235356 133585637513024 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 14:49:59.171762 133454317942528 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_144122/linen_xpk_main_20260421_144122_02_synthetic/checkpoints/9/items/array_metadatas/process_7
I0421 14:50:34.918271 133454309549824 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 37.850 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.73327946662903 s) (per-host)
I0421 14:50:34.918401 133454309549824 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.685441s.
I0421 14:50:43.960201 133454309549824 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 50.727227s.
I0421 14:50:43.960442 133432050956032 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 14:50:43.960497 133432050956032 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save.
I0421 14:50:43.960543 133432050956032 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0421 14:50:43.967912 133432050956032 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0421 14:50:43.968089 133585637513024 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 14:50:43.968242 133585637513024 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 14:50:43.969166 133585637513024 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.353, Tokens/s/device: 13920.514, total_weights: 65536, loss: 8.179, lm_loss: 8.179, perplexity: 3564.635
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Tue Apr 21 14:50:53 UTC 2026
EXIT_CODE=0