MaxView

XPK Start: Thu Apr 23 07:54:16 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. 
2026-04-23 07:54:41.052619: 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)
I0423 07:54:41.263463 134937117493056 max_utils.py:273] Attempting to initialize the jax distributed system...
I0423 07:54:50.304568 134937117493056 distributed.py:149] Starting JAX distributed service on [::]:8482
I0423 07:54:50.306813 134937117493056 distributed.py:172] Connecting to JAX distributed service on mt-10-shardy-true-8riju-slice-job-0-0.mt-10-shardy-true-8riju:8482
I0423 07:54:51.681029 134937117493056 max_utils.py:284] Jax distributed system initialized!
I0423 07:54:57.945348 134937117493056 max_utils.py:800] System Information: Jax Version: 0.9.2
I0423 07:54:57.945455 134937117493056 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0423 07:54:57.945494 134937117493056 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
I0423 07:54:57.945535 134937117493056 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0423 07:54:58.653835 134937117493056 maxtext_utils.py:1604] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0423 07:54:58.654127 134937117493056 checkpointing.py:677] Setting up checkpoint logger...
I0423 07:54:58.654183 134937117493056 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0423 07:54:58.654227 134937117493056 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0423 07:54:58.654578 134937117493056 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 0x7ab8c7348a40>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0423 07:55:01.802383 134937117493056 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0423 07:55:01.802626 134937117493056 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 0x7aa47432c3b0>}, handler_registry=None
I0423 07:55:01.802865 134937117493056 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7aa47432c3b0>` 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`.
I0423 07:55:01.802912 134937117493056 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7aa474118b90>` 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`.
I0423 07:55:01.802948 134937117493056 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 0x7aa47432c3b0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7aa47432c3b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7aa474118b90>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7aa474118b90>}).
I0423 07:55:01.803274 134937117493056 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0423 07:55:01.803348 134937117493056 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7aa39047da80> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0423 07:55:03.561692 134937117493056 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_10_shardy_true/checkpoints
I0423 07:55:03.596245 134937117493056 checkpoint_manager.py:929] [process=6][thread=MainThread] CheckpointManager created,  primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_10_shardy_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7aa474118620>
I0423 07:55:03.596381 134937117493056 checkpointing.py:301] Checkpoint manager created!
I0423 07:55:04.523080 134937117493056 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0423 07:55:04.523214 134937117493056 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 07:55:04.904983 134937117493056 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch_attn', 'activation_length_attn', 'activation_embed_attn').
I0423 07:55:04.905079 134937117493056 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 07:55:04.922027 134937117493056 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch_attn', 'activation_length_attn', 'activation_embed_attn').
I0423 07:55:04.922087 134937117493056 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0423 07:55:04.946365 134937117493056 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 07:55:04.946434 134937117493056 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 07:55:04.963525 134937117493056 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 07:55:04.963596 134937117493056 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 07:55:04.980572 134937117493056 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_length_attn', 'activation_kv_heads', 'activation_kv_head_dim').
I0423 07:55:04.980639 134937117493056 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 07:55:05.006105 134937117493056 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch_attn', 'activation_length_attn', 'activation_heads', 'activation_kv').
I0423 07:55:05.006174 134937117493056 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0423 07:55:05.027746 134937117493056 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0423 07:55:05.027814 134937117493056 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0423 07:55:05.241860 134937117493056 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0423 07:55:05.241975 134937117493056 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0423 07:55:06.671266 134937117493056 maxtext_utils.py:1707]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0423 07:55:06.671398 134937117493056 maxtext_utils.py:1707]  params/params/decoder/layers/mlp/wi_0/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0423 07:55:06.671452 134937117493056 maxtext_utils.py:1707]  params/params/decoder/layers/mlp/wi_1/kernel
    Shape:     float32[2048,16,7168]
    Logical:   P('embed', 'layers', 'mlp')
    Physical:  ('fsdp', None, None)
I0423 07:55:06.671509 134937117493056 maxtext_utils.py:1707]  params/params/decoder/layers/mlp/wo/kernel
    Shape:     float32[7168,16,2048]
    Logical:   P('mlp', 'layers', 'embed')
    Physical:  (None, None, 'fsdp')
I0423 07:55:06.671571 134937117493056 maxtext_utils.py:1707]  params/params/decoder/layers/post_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0423 07:55:06.671612 134937117493056 maxtext_utils.py:1707]  params/params/decoder/layers/pre_self_attention_layer_norm/scale
    Shape:     float32[2048,16]
    Logical:   P('norm', 'layers')
    Physical:  (None, None)
I0423 07:55:06.671665 134937117493056 maxtext_utils.py:1707]  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)
I0423 07:55:06.671716 134937117493056 maxtext_utils.py:1707]  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')
I0423 07:55:06.671755 134937117493056 maxtext_utils.py:1707]  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)
I0423 07:55:06.671792 134937117493056 maxtext_utils.py:1707]  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)
I0423 07:55:06.671839 134937117493056 maxtext_utils.py:1707]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)

I0423 07:55:06.671887 134937117493056 maxtext_utils.py:1707]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0423 07:55:07.171033 134937117493056 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0423 07:55:07.171148 134937117493056 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0423 07:55:07.187032 134937117493056 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0423 07:55:07.187105 134937117493056 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0423 07:55:18.230356 134937117493056 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.
I0423 07:55:18.231228 134937117493056 metric_logger.py:301] number parameters: 1.104 billion
I0423 07:55:30.375482 134937117493056 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0423 07:55:30.531723 134937117493056 checkpointing.py:776] Waited 0.1562199592590332 seconds for step 0 to finish before starting checkpointing.
I0423 07:55:30.534090 134937117493056 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 07:55:30.535690 134937117493056 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0423 07:55:30.537041 134937117493056 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_10_shardy_true/checkpoints/0.
I0423 07:55:31.296318 134937117493056 signaling_client.py:364] Using JaxDistributedSignalingClient
I0423 07:55:31.298907 134937117493056 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0423 07:55:31.298964 134937117493056 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
I0423 07:55:31.577702 134937117493056 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.281462s
I0423 07:55:31.577886 134937117493056 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.375 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2869696617126465 s) (per-host)
I0423 07:55:31.577942 134937117493056 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.287046s (batch_requests_ready=0.002237s, total_serialization_initiated=0.284725s, others=0.000085s)
I0423 07:55:31.578048 134937117493056 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.291074s (all_items=0.000018s, per_item={'items': '0.00001836'}, temp_paths=0.291056)
I0423 07:55:31.578890 134937117493056 event_tracking.py:125] [process=6] [async] Finished blocking save in 1.04 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_10_shardy_true/checkpoints/0.
I0423 07:55:31.579238 134807935178496 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 08:15:31.579198
I0423 07:55:31.587080 134937117493056 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 07:55:31.587384 134804162995968 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0423 07:55:31.587539 134937117493056 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776930930.5340724, 'wait_for_prev_duration_secs': 7.200241088867188e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776930930.5357287, 'checkpointer_blocking_duration_secs': 1.0436692237854004, 'get_old_steps_start_time': 1776930931.5794241, 'get_old_steps_duration_secs': 3.0994415283203125e-05, 'checkpoint_manager_blocking_start_time': 1776930930.532253, 'checkpoint_manager_blocking_duration_secs': 1.0552465915679932}
I0423 07:55:31.587644 134937117493056 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0423 07:55:31.587693 134937117493056 max_utils.py:750] 
Memstats: After params initialized:
I0423 07:55:31.587743 134937117493056 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0423 07:55:31.587776 134937117493056 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0423 07:55:31.587801 134937117493056 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0423 07:55:31.587827 134937117493056 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0423 07:55:31.904980 134937117493056 metric_logger.py:196] completed step: 0, seconds: 12.144, TFLOP/s/device: 1.119, Tokens/s/device: 168.641, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0423 07:55:32.078904 134937117493056 metric_logger.py:196] completed step: 1, seconds: 1.528, TFLOP/s/device: 8.892, Tokens/s/device: 1340.372, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0423 07:55:32.505897 134937117493056 metric_logger.py:196] completed step: 2, seconds: 0.027, TFLOP/s/device: 509.148, Tokens/s/device: 76744.360, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0423 07:55:32.653220 134937117493056 metric_logger.py:196] completed step: 3, seconds: 0.427, TFLOP/s/device: 31.793, Tokens/s/device: 4792.224, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0423 07:55:32.948232 134937117493056 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.886, Tokens/s/device: 13397.880, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0423 07:55:32.953931 134937117493056 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.291, Tokens/s/device: 13911.059, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0423 07:55:36.576321    2858 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0423 07:55:39.120030 134805212292864 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_10_shardy_true/checkpoints/0/items/array_metadatas/process_6
I0423 07:55:55.565791 134937117493056 metric_logger.py:196] completed step: 6, seconds: 0.295, TFLOP/s/device: 45.998, Tokens/s/device: 6933.348, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0423 07:55:55.713122 134937117493056 metric_logger.py:196] completed step: 7, seconds: 22.465, TFLOP/s/device: 0.605, Tokens/s/device: 91.163, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0423 07:55:55.860363 134937117493056 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.173, Tokens/s/device: 13441.142, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0423 07:55:56.006937 134937117493056 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0423 07:55:56.007560 134937117493056 checkpointing.py:776] Waited 0.0006430149078369141 seconds for step 9 to finish before starting checkpointing.
I0423 07:55:56.010975 134937117493056 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 07:56:09.831361 134807935178496 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 40.985 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 38.54042100906372 s) (per-host)
I0423 07:56:09.831467 134807935178496 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 38.252110s.
I0423 07:56:19.445087 134807935178496 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 47.865712s.
I0423 07:56:19.445423 134804162995968 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 07:56:19.445561 134804162995968 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0423 07:56:19.445611 134804162995968 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0423 07:56:19.447243 134804162995968 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0423 07:56:19.447414 134937117493056 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.
W0423 07:56:19.447543 134937117493056 checkpoint_manager.py:1452] Waiting for previous save to complete took 23.436569 seconds. If this number is high, consider checkpointing less frequently.
I0423 07:56:19.449466 134937117493056 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0423 07:56:19.451787 134937117493056 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_10_shardy_true/checkpoints/9.
I0423 07:56:20.145057 134937117493056 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0423 07:56:20.145166 134937117493056 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
I0423 07:56:20.185118 134937117493056 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.041193s
I0423 07:56:20.185293 134937117493056 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 34.490 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04472541809082031 s) (per-host)
I0423 07:56:20.185344 134937117493056 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.044787s (batch_requests_ready=0.001826s, total_serialization_initiated=0.042891s, others=0.000070s)
I0423 07:56:20.185439 134937117493056 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.048857s (all_items=0.000016s, per_item={'items': '0.00001645'}, temp_paths=0.048840)
I0423 07:56:20.186110 134937117493056 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.74 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_10_shardy_true/checkpoints/9.
I0423 07:56:20.186463 134804162995968 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-23 08:16:20.186424
I0423 07:56:20.188544 134937117493056 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0423 07:56:20.188867 134805212292864 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0423 07:56:20.189030 134937117493056 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_10_shardy_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776930956.010944, 'wait_for_prev_duration_secs': 23.436569452285767, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776930979.4495053, 'checkpointer_blocking_duration_secs': 0.7371017932891846, 'get_old_steps_start_time': 1776930980.1866317, 'get_old_steps_duration_secs': 2.8848648071289062e-05, 'checkpoint_manager_blocking_start_time': 1776930956.007788, 'checkpoint_manager_blocking_duration_secs': 24.18120574951172}
I0423 07:56:20.189172 134937117493056 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0423 07:56:20.189217 134937117493056 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0423 07:56:26.156921 134786286847744 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260423_071538/linen_xpk_main_20260423_071538_10_shardy_true/checkpoints/9/items/array_metadatas/process_6
I0423 07:57:02.288498 134804162995968 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.477 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.147905111312866 s) (per-host)
I0423 07:57:02.288614 134804162995968 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.102040s.
I0423 07:57:11.166948 134804162995968 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 50.980358s.
I0423 07:57:11.167242 134805212292864 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0423 07:57:11.167365 134805212292864 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0423 07:57:11.167413 134805212292864 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0423 07:57:11.169355 134805212292864 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0423 07:57:11.169580 134937117493056 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.
I0423 07:57:11.169735 134937117493056 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0423 07:57:11.170693 134937117493056 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.254, Tokens/s/device: 13905.486, 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: Thu Apr 23 07:57:21 UTC 2026
EXIT_CODE=0