XPK Start: Sat Apr 25 08:07:38 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.
2026-04-25 08:08:02.872552: 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 08:08:03.076600 135559152817984 max_utils.py:273] Attempting to initialize the jax distributed system...
I0425 08:08:12.116792 135559152817984 distributed.py:149] Starting JAX distributed service on [::]:8482
I0425 08:08:12.119438 135559152817984 distributed.py:172] Connecting to JAX distributed service on mt-15-ocdbt-true-sv-hybod-slice-job-0-0.mt-15-ocdbt-true-sv-hybod:8482
I0425 08:08:13.265152 135559152817984 max_utils.py:284] Jax distributed system initialized!
I0425 08:08:19.226932 135559152817984 max_utils.py:800] System Information: Jax Version: 0.9.2
I0425 08:08:19.227038 135559152817984 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0425 08:08:19.227079 135559152817984 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 08:08:19.227135 135559152817984 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0425 08:08:19.916470 135559152817984 maxtext_utils.py:1604] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1)
I0425 08:08:19.916758 135559152817984 checkpointing.py:677] Setting up checkpoint logger...
I0425 08:08:19.916815 135559152817984 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0425 08:08:19.916859 135559152817984 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0425 08:08:19.917217 135559152817984 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 0x7b49ba31c110>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0425 08:08:22.793377 135559152817984 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0425 08:08:22.793617 135559152817984 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 0x7b358c118d10>}, handler_registry=None
I0425 08:08:22.793854 135559152817984 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b358c118d10>` 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 08:08:22.793902 135559152817984 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b358c310c80>` 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 08:08:22.793938 135559152817984 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 0x7b358c118d10>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7b358c118d10>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b358c310c80>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7b358c310c80>}).
I0425 08:08:22.794281 135559152817984 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0425 08:08:22.794350 135559152817984 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7b346c481a80> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0425 08:08:23.873901 135559152817984 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints
I0425 08:08:24.223623 135559152817984 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=5), 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_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7b358c119670>
I0425 08:08:24.223796 135559152817984 checkpointing.py:301] Checkpoint manager created!
I0425 08:08:25.156772 135559152817984 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0425 08:08:25.156888 135559152817984 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 08:08:25.540248 135559152817984 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch_attn', 'activation_length_attn', 'activation_embed_attn').
I0425 08:08:25.540338 135559152817984 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 08:08:25.557030 135559152817984 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch_attn', 'activation_length_attn', 'activation_embed_attn').
I0425 08:08:25.557103 135559152817984 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0425 08:08:25.581136 135559152817984 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 08:08:25.581205 135559152817984 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 08:08:25.598009 135559152817984 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 08:08:25.598074 135559152817984 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 08:08:25.614883 135559152817984 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 08:08:25.614946 135559152817984 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 08:08:25.640131 135559152817984 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch_attn', 'activation_length_attn', 'activation_heads', 'activation_kv').
I0425 08:08:25.640201 135559152817984 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0425 08:08:25.661560 135559152817984 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0425 08:08:25.661625 135559152817984 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0425 08:08:25.875051 135559152817984 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0425 08:08:25.875190 135559152817984 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0425 08:08:27.346025 135559152817984 maxtext_utils.py:1707] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0425 08:08:27.346167 135559152817984 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)
I0425 08:08:27.346220 135559152817984 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)
I0425 08:08:27.346282 135559152817984 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')
I0425 08:08:27.346337 135559152817984 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)
I0425 08:08:27.346376 135559152817984 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)
I0425 08:08:27.346436 135559152817984 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)
I0425 08:08:27.346489 135559152817984 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')
I0425 08:08:27.346528 135559152817984 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)
I0425 08:08:27.346565 135559152817984 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)
I0425 08:08:27.346612 135559152817984 maxtext_utils.py:1707] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0425 08:08:27.346658 135559152817984 maxtext_utils.py:1707] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0425 08:08:27.844150 135559152817984 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0425 08:08:27.844250 135559152817984 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0425 08:08:27.859973 135559152817984 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0425 08:08:27.860033 135559152817984 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0425 08:08:38.856812 135559152817984 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.
I0425 08:08:38.857620 135559152817984 metric_logger.py:301] number parameters: 1.104 billion
I0425 08:08:50.528778 135559152817984 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0425 08:08:50.682667 135559152817984 checkpointing.py:776] Waited 0.1538715362548828 seconds for step 0 to finish before starting checkpointing.
I0425 08:08:50.686374 135559152817984 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 08:08:50.687967 135559152817984 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0425 08:08:50.689340 135559152817984 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints/0.
I0425 08:08:51.015769 135559152817984 signaling_client.py:364] Using JaxDistributedSignalingClient
I0425 08:08:51.016589 135559152817984 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0425 08:08:51.016645 135559152817984 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 08:08:51.290517 135559152817984 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.274822s
I0425 08:08:51.290699 135559152817984 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.501 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2803916931152344 s) (per-host)
I0425 08:08:51.290765 135559152817984 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.280464s (batch_requests_ready=0.002258s, total_serialization_initiated=0.278126s, others=0.000081s)
I0425 08:08:51.290883 135559152817984 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.284883s (all_items=0.000018s, per_item={'items': '0.00001764'}, temp_paths=0.284865)
I0425 08:08:51.291751 135559152817984 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.60 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints/0.
I0425 08:08:51.292087 135431344424704 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 08:28:51.292049
I0425 08:08:51.304430 135559152817984 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 08:08:51.304777 135429201540864 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 08:08:51.304952 135559152817984 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777104530.6863546, 'wait_for_prev_duration_secs': 6.556510925292969e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777104530.6880097, 'checkpointer_blocking_duration_secs': 0.6042373180389404, 'get_old_steps_start_time': 1777104531.2922752, 'get_old_steps_duration_secs': 3.24249267578125e-05, 'checkpoint_manager_blocking_start_time': 1777104530.683161, 'checkpoint_manager_blocking_duration_secs': 0.6217505931854248}
I0425 08:08:51.305056 135559152817984 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0425 08:08:51.305121 135559152817984 max_utils.py:750]
Memstats: After params initialized:
I0425 08:08:51.305172 135559152817984 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0425 08:08:51.305204 135559152817984 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0425 08:08:51.305233 135559152817984 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0425 08:08:51.305258 135559152817984 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0425 08:08:51.618162 135559152817984 metric_logger.py:196] completed step: 0, seconds: 11.671, TFLOP/s/device: 1.164, Tokens/s/device: 175.477, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0425 08:08:51.803029 135559152817984 metric_logger.py:196] completed step: 1, seconds: 1.088, TFLOP/s/device: 12.490, Tokens/s/device: 1882.644, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0425 08:08:52.232319 135559152817984 metric_logger.py:196] completed step: 2, seconds: 0.036, TFLOP/s/device: 376.740, Tokens/s/device: 56786.358, total_weights: 65536, loss: 10.560, lm_loss: 10.560, perplexity: 38547.719
I0425 08:08:52.379675 135559152817984 metric_logger.py:196] completed step: 3, seconds: 0.431, TFLOP/s/device: 31.523, Tokens/s/device: 4751.553, total_weights: 65536, loss: 9.980, lm_loss: 9.980, perplexity: 21590.506
I0425 08:08:52.384327 135559152817984 checkpointing.py:772] Waiting for step 5 to finish before checkpoint...
I0425 08:08:52.673807 135559152817984 checkpointing.py:776] Waited 0.289442777633667 seconds for step 5 to finish before starting checkpointing.
I0425 08:08:52.676901 135559152817984 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 08:08:54.129702 2802 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0425 08:08:56.163827 135429243139840 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints/0/items/array_metadatas/process_6
I0425 08:09:28.226133 135431344424704 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 42.444 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.215781927108765 s) (per-host)
I0425 08:09:28.226263 135431344424704 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.934040s.
I0425 08:09:36.791805 135431344424704 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 45.499582s.
I0425 08:09:36.792127 135429201540864 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 08:09:36.792246 135429201540864 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 08:09:36.792309 135429201540864 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0425 08:09:36.794218 135429201540864 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0425 08:09:36.794395 135559152817984 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 08:09:36.794544 135559152817984 checkpoint_manager.py:1452] Waiting for previous save to complete took 44.117640 seconds. If this number is high, consider checkpointing less frequently.
I0425 08:09:36.796434 135559152817984 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 5
I0425 08:09:36.798820 135559152817984 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints/5.
I0425 08:09:37.090182 135559152817984 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0425 08:09:37.090276 135559152817984 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 08:09:37.133239 135559152817984 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.044156s
I0425 08:09:37.133412 135559152817984 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 32.383 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04763603210449219 s) (per-host)
I0425 08:09:37.133462 135559152817984 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.047696s (batch_requests_ready=0.001792s, total_serialization_initiated=0.045837s, others=0.000067s)
I0425 08:09:37.133553 135559152817984 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.052717s (all_items=0.000014s, per_item={'items': '0.00001383'}, temp_paths=0.052703)
I0425 08:09:37.134238 135559152817984 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.34 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints/5.
I0425 08:09:37.134589 135429201540864 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 08:29:37.134551
I0425 08:09:37.136541 135559152817984 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=5] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 08:09:37.136832 135408834832128 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 08:09:37.136999 135559152817984 standard_logger.py:34] {'step': 5, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777104532.676871, 'wait_for_prev_duration_secs': 44.11764049530029, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777104576.7964742, 'checkpointer_blocking_duration_secs': 0.3382599353790283, 'get_old_steps_start_time': 1777104577.1347566, 'get_old_steps_duration_secs': 2.8371810913085938e-05, 'checkpoint_manager_blocking_start_time': 1777104532.674473, 'checkpoint_manager_blocking_duration_secs': 44.46249008178711}
I0425 08:09:37.137121 135559152817984 checkpointing.py:408] Started an asynchronous checkpoint save for step 5
I0425 08:09:37.137915 135559152817984 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.701, Tokens/s/device: 13369.979, total_weights: 65536, loss: 9.460, lm_loss: 9.460, perplexity: 12840.355
I0425 08:09:37.143872 135559152817984 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.215, Tokens/s/device: 13899.635, total_weights: 65536, loss: 8.959, lm_loss: 8.959, perplexity: 7776.467
I0425 08:09:42.690937 135429243139840 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints/5/items/array_metadatas/process_6
I0425 08:09:59.634548 135559152817984 metric_logger.py:196] completed step: 6, seconds: 44.758, TFLOP/s/device: 0.304, Tokens/s/device: 45.757, total_weights: 65536, loss: 8.469, lm_loss: 8.469, perplexity: 4762.740
I0425 08:09:59.781892 135559152817984 metric_logger.py:196] completed step: 7, seconds: 22.345, TFLOP/s/device: 0.608, Tokens/s/device: 91.655, total_weights: 65536, loss: 8.003, lm_loss: 8.003, perplexity: 2989.151
I0425 08:09:59.929207 135559152817984 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.301, Tokens/s/device: 13460.489, total_weights: 65536, loss: 7.572, lm_loss: 7.572, perplexity: 1943.098
I0425 08:10:00.075808 135559152817984 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0425 08:10:00.076479 135559152817984 checkpointing.py:776] Waited 0.0006937980651855469 seconds for step 9 to finish before starting checkpointing.
I0425 08:10:00.078481 135559152817984 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=5][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 08:10:19.685421 135429201540864 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 37.080 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.59960675239563 s) (per-host)
I0425 08:10:19.685544 135429201540864 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.550839s.
I0425 08:10:27.623064 135429201540864 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 50.488343s.
I0425 08:10:27.623373 135408834832128 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 08:10:27.623494 135408834832128 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 08:10:27.623544 135408834832128 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is syncing with other hosts...
I0425 08:10:27.624833 135408834832128 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=5] CheckpointManager Save Finalize is done on all hosts.
I0425 08:10:27.625009 135559152817984 checkpoint_manager.py:2032] [process=6][thread=MainThread][step=5][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=5.
W0425 08:10:27.625146 135559152817984 checkpoint_manager.py:1452] Waiting for previous save to complete took 27.546670 seconds. If this number is high, consider checkpointing less frequently.
I0425 08:10:27.626577 135559152817984 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0425 08:10:27.628777 135559152817984 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints/9.
I0425 08:10:27.918575 135559152817984 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0425 08:10:27.918673 135559152817984 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 08:10:27.957807 135559152817984 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.040102s
I0425 08:10:27.957972 135559152817984 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 35.395 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04358196258544922 s) (per-host)
I0425 08:10:27.958019 135559152817984 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.043639s (batch_requests_ready=0.001655s, total_serialization_initiated=0.041922s, others=0.000062s)
I0425 08:10:27.958119 135559152817984 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.047539s (all_items=0.000017s, per_item={'items': '0.00001717'}, temp_paths=0.047521)
I0425 08:10:27.958781 135559152817984 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.33 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints/9.
I0425 08:10:27.959126 135429243139840 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-25 08:30:27.959075
I0425 08:10:27.961009 135559152817984 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0425 08:10:27.961249 135408834832128 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0425 08:10:27.961375 135559152817984 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777104600.0784514, 'wait_for_prev_duration_secs': 27.546669721603394, 'time_between_consecutive_saves_sec': 23.284193992614746, 'checkpointer_blocking_start_time': 1777104627.6266162, 'checkpointer_blocking_duration_secs': 0.3326590061187744, 'get_old_steps_start_time': 1777104627.9592934, 'get_old_steps_duration_secs': 2.3365020751953125e-05, 'checkpoint_manager_blocking_start_time': 1777104600.0767086, 'checkpoint_manager_blocking_duration_secs': 27.884633541107178}
I0425 08:10:27.961480 135559152817984 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0425 08:10:27.961522 135559152817984 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0425 08:10:33.048823 135428673877760 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260425_071456/linen_xpk_main_20260425_071456_15_ocdbt_true/checkpoints/9/items/array_metadatas/process_6
I0425 08:11:09.302888 135429243139840 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 38.165 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.388458251953125 s) (per-host)
I0425 08:11:09.302993 135429243139840 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.343753s.
I0425 08:11:16.115286 135429243139840 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 48.156029s.
I0425 08:11:16.115494 135408834832128 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0425 08:11:16.115558 135408834832128 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0425 08:11:16.115597 135408834832128 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0425 08:11:16.118496 135408834832128 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0425 08:11:16.118660 135559152817984 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 08:11:16.118812 135559152817984 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0425 08:11:16.119787 135559152817984 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.312, Tokens/s/device: 13914.272, total_weights: 65536, loss: 7.181, lm_loss: 7.181, perplexity: 1314.495
Per train step:
Total TFLOPs: 13.59
split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Sat Apr 25 08:11:28 UTC 2026
EXIT_CODE=0