XPK Start: Fri Apr 24 21:20:00 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.
2026-04-24 21:20:25.019086: 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)
I0424 21:20:25.228130 139037601449792 max_utils.py:273] Attempting to initialize the jax distributed system...
I0424 21:20:34.268210 139037601449792 distributed.py:149] Starting JAX distributed service on [::]:8482
I0424 21:20:34.270597 139037601449792 distributed.py:172] Connecting to JAX distributed service on mt-16-zarr3-true-sv-pieic-slice-job-0-0.mt-16-zarr3-true-sv-pieic:8482
I0424 21:20:35.485309 139037601449792 max_utils.py:284] Jax distributed system initialized!
I0424 21:20:41.823555 139037601449792 max_utils.py:800] System Information: Jax Version: 0.9.2
I0424 21:20:41.823662 139037601449792 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0424 21:20:41.823702 139037601449792 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
I0424 21:20:41.823738 139037601449792 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0424 21:20:42.522906 139037601449792 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0424 21:20:42.523209 139037601449792 checkpointing.py:677] Setting up checkpoint logger...
I0424 21:20:42.523266 139037601449792 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0424 21:20:42.523312 139037601449792 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0424 21:20:42.523669 139037601449792 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 0x7e737f04c380>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0424 21:20:45.472619 139037601449792 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0424 21:20:45.472856 139037601449792 checkpoint_manager.py:708] [process=3][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e5f50787470>}, handler_registry=None
I0424 21:20:45.473110 139037601449792 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e5f50787470>` 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`.
I0424 21:20:45.473162 139037601449792 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e5f50788650>` 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`.
I0424 21:20:45.473200 139037601449792 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 0x7e5f50787470>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e5f50787470>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e5f50788650>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e5f50788650>}).
I0424 21:20:45.473546 139037601449792 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0424 21:20:45.473618 139037601449792 async_checkpointer.py:192] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7e5f5028f420> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0424 21:20:46.514163 139037601449792 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints
I0424 21:20:46.522583 139037601449792 checkpoint_manager.py:929] [process=3][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_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7e5f5078aea0>
I0424 21:20:46.522698 139037601449792 checkpointing.py:301] Checkpoint manager created!
I0424 21:20:47.453601 139037601449792 nnx_wrappers.py:453] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0424 21:20:47.453726 139037601449792 nnx_wrappers.py:453] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 21:20:47.834321 139037601449792 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0424 21:20:47.834419 139037601449792 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 21:20:47.855912 139037601449792 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0424 21:20:47.856037 139037601449792 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0424 21:20:47.879762 139037601449792 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 21:20:47.879830 139037601449792 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 21:20:47.896215 139037601449792 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 21:20:47.896277 139037601449792 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 21:20:47.912560 139037601449792 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0424 21:20:47.912624 139037601449792 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 21:20:47.937351 139037601449792 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0424 21:20:47.937423 139037601449792 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0424 21:20:47.958579 139037601449792 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0424 21:20:47.958648 139037601449792 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0424 21:20:48.177412 139037601449792 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0424 21:20:48.177530 139037601449792 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0424 21:20:49.616864 139037601449792 maxtext_utils.py:1668] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0424 21:20:49.617005 139037601449792 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)
I0424 21:20:49.617059 139037601449792 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)
I0424 21:20:49.617116 139037601449792 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')
I0424 21:20:49.617172 139037601449792 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)
I0424 21:20:49.617210 139037601449792 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)
I0424 21:20:49.617263 139037601449792 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)
I0424 21:20:49.617315 139037601449792 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')
I0424 21:20:49.617355 139037601449792 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)
I0424 21:20:49.617390 139037601449792 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)
I0424 21:20:49.617437 139037601449792 maxtext_utils.py:1668] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0424 21:20:49.617483 139037601449792 maxtext_utils.py:1668] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0424 21:20:50.106609 139037601449792 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0424 21:20:50.106707 139037601449792 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0424 21:20:50.122215 139037601449792 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0424 21:20:50.122274 139037601449792 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0424 21:21:00.789961 139037601449792 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.
I0424 21:21:00.790663 139037601449792 metric_logger.py:301] number parameters: 1.104 billion
I0424 21:21:12.074747 139037601449792 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0424 21:21:12.651993 139037601449792 checkpointing.py:776] Waited 0.5771982669830322 seconds for step 0 to finish before starting checkpointing.
I0424 21:21:12.654357 139037601449792 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0424 21:21:12.656117 139037601449792 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 0
I0424 21:21:12.657540 139037601449792 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints/0.
I0424 21:21:12.959691 139037601449792 signaling_client.py:364] Using JaxDistributedSignalingClient
I0424 21:21:12.960525 139037601449792 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0424 21:21:12.960581 139037601449792 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
I0424 21:21:13.237327 139037601449792 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.277708s
I0424 21:21:13.237492 139037601449792 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 5.442 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2834491729736328 s) (per-host)
I0424 21:21:13.237543 139037601449792 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.283510s (batch_requests_ready=0.002572s, total_serialization_initiated=0.280871s, others=0.000068s)
I0424 21:21:13.237634 139037601449792 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.287963s (all_items=0.000019s, per_item={'items': '0.00001860'}, temp_paths=0.287944)
I0424 21:21:13.238356 139037601449792 event_tracking.py:125] [process=3] [async] Finished blocking save in 0.58 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints/0.
I0424 21:21:13.238636 138907891001088 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 21:41:13.238596
I0424 21:21:13.265592 139037601449792 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 21:21:13.265884 138909462628096 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0424 21:21:13.266065 139037601449792 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777065672.65434, 'wait_for_prev_duration_secs': 5.984306335449219e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777065672.6561549, 'checkpointer_blocking_duration_secs': 0.5826413631439209, 'get_old_steps_start_time': 1777065673.238818, 'get_old_steps_duration_secs': 2.5510787963867188e-05, 'checkpoint_manager_blocking_start_time': 1777065672.6525254, 'checkpoint_manager_blocking_duration_secs': 0.6135029792785645}
I0424 21:21:13.266164 139037601449792 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0424 21:21:13.266212 139037601449792 max_utils.py:750]
Memstats: After params initialized:
I0424 21:21:13.266264 139037601449792 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_10(process=3,(2,2,0,0))
I0424 21:21:13.266298 139037601449792 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_11(process=3,(3,2,0,0))
I0424 21:21:13.266326 139037601449792 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_14(process=3,(2,3,0,0))
I0424 21:21:13.266353 139037601449792 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_15(process=3,(3,3,0,0))
I0424 21:21:13.578071 139037601449792 metric_logger.py:196] completed step: 0, seconds: 11.284, TFLOP/s/device: 1.204, Tokens/s/device: 181.497, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0424 21:21:13.760896 139037601449792 metric_logger.py:196] completed step: 1, seconds: 1.502, TFLOP/s/device: 9.047, Tokens/s/device: 1363.689, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0424 21:21:14.204663 139037601449792 metric_logger.py:196] completed step: 2, seconds: 0.035, TFLOP/s/device: 384.295, Tokens/s/device: 57925.105, total_weights: 65536, loss: 10.560, lm_loss: 10.560, perplexity: 38547.719
I0424 21:21:14.352134 139037601449792 metric_logger.py:196] completed step: 3, seconds: 0.444, TFLOP/s/device: 30.600, Tokens/s/device: 4612.343, total_weights: 65536, loss: 9.980, lm_loss: 9.980, perplexity: 21590.506
I0424 21:21:14.356643 139037601449792 checkpointing.py:772] Waiting for step 5 to finish before checkpoint...
I0424 21:21:14.646479 139037601449792 checkpointing.py:776] Waited 0.28979969024658203 seconds for step 5 to finish before starting checkpointing.
I0424 21:21:14.649646 139037601449792 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 21:21:16.428999 2798 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0424 21:21:18.652680 138907899393792 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints/0/items/array_metadatas/process_3
I0424 21:21:49.859172 138907891001088 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 42.802 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.90508770942688 s) (per-host)
I0424 21:21:49.859294 138907891001088 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.620543s.
I0424 21:21:58.364051 138907891001088 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 45.125282s.
I0424 21:21:58.364359 138909462628096 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 21:21:58.364497 138909462628096 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0424 21:21:58.364564 138909462628096 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0424 21:21:58.366109 138909462628096 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0424 21:21:58.366296 139037601449792 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0424 21:21:58.366429 139037601449792 checkpoint_manager.py:1452] Waiting for previous save to complete took 43.716781 seconds. If this number is high, consider checkpointing less frequently.
I0424 21:21:58.368363 139037601449792 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 5
I0424 21:21:58.371413 139037601449792 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints/5.
I0424 21:21:59.097821 139037601449792 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0424 21:21:59.097913 139037601449792 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
I0424 21:21:59.134459 139037601449792 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.037500s
I0424 21:21:59.134621 139037601449792 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 37.504 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04113125801086426 s) (per-host)
I0424 21:21:59.134671 139037601449792 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.041190s (batch_requests_ready=0.001893s, total_serialization_initiated=0.039231s, others=0.000066s)
I0424 21:21:59.134757 139037601449792 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.046333s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.046315)
I0424 21:21:59.135474 139037601449792 event_tracking.py:125] [process=3] [async] Finished blocking save in 0.77 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints/5.
I0424 21:21:59.135821 138909462628096 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 21:41:59.135784
I0424 21:21:59.142199 139037601449792 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=5] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 21:21:59.142505 138904625133312 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0424 21:21:59.142669 139037601449792 standard_logger.py:34] {'step': 5, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777065674.6496184, 'wait_for_prev_duration_secs': 43.71678113937378, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1777065718.368402, 'checkpointer_blocking_duration_secs': 0.767575740814209, 'get_old_steps_start_time': 1777065719.1360004, 'get_old_steps_duration_secs': 3.0279159545898438e-05, 'checkpoint_manager_blocking_start_time': 1777065674.6471424, 'checkpoint_manager_blocking_duration_secs': 44.49549317359924}
I0424 21:21:59.142773 139037601449792 checkpointing.py:408] Started an asynchronous checkpoint save for step 5
I0424 21:21:59.143615 139037601449792 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.643, Tokens/s/device: 13361.256, total_weights: 65536, loss: 9.460, lm_loss: 9.460, perplexity: 12840.355
I0424 21:21:59.149677 139037601449792 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.231, Tokens/s/device: 13901.994, total_weights: 65536, loss: 8.959, lm_loss: 8.959, perplexity: 7776.467
I0424 21:22:04.263815 138907899393792 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints/5/items/array_metadatas/process_3
I0424 21:22:21.784993 139037601449792 metric_logger.py:196] completed step: 6, seconds: 44.792, TFLOP/s/device: 0.303, Tokens/s/device: 45.723, total_weights: 65536, loss: 8.469, lm_loss: 8.469, perplexity: 4762.740
I0424 21:22:21.932144 139037601449792 metric_logger.py:196] completed step: 7, seconds: 22.489, TFLOP/s/device: 0.604, Tokens/s/device: 91.066, total_weights: 65536, loss: 8.003, lm_loss: 8.003, perplexity: 2989.151
I0424 21:22:22.079352 139037601449792 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.296, Tokens/s/device: 13459.693, total_weights: 65536, loss: 7.572, lm_loss: 7.572, perplexity: 1943.098
I0424 21:22:22.226311 139037601449792 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0424 21:22:22.227022 139037601449792 checkpointing.py:776] Waited 0.0007314682006835938 seconds for step 9 to finish before starting checkpointing.
I0424 21:22:22.229510 139037601449792 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=5][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 21:22:40.873837 138909462628096 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 37.807 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.78030705451965 s) (per-host)
I0424 21:22:40.873964 138909462628096 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.738034s.
I0424 21:22:49.600563 138909462628096 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 50.464619s.
I0424 21:22:49.600861 138904625133312 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 21:22:49.601010 138904625133312 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0424 21:22:49.601064 138904625133312 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=5] CheckpointManager Save Finalize is syncing with other hosts...
I0424 21:22:49.602395 138904625133312 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=5] CheckpointManager Save Finalize is done on all hosts.
I0424 21:22:49.602579 139037601449792 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=5][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=5.
W0424 21:22:49.602703 139037601449792 checkpoint_manager.py:1452] Waiting for previous save to complete took 27.373201 seconds. If this number is high, consider checkpointing less frequently.
I0424 21:22:49.605142 139037601449792 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 9
I0424 21:22:49.607230 139037601449792 event_tracking.py:70] [process=3] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints/9.
I0424 21:22:50.192823 139037601449792 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0424 21:22:50.192918 139037601449792 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
I0424 21:22:50.228371 139037601449792 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.036461s
I0424 21:22:50.228517 139037601449792 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 38.483 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.0400846004486084 s) (per-host)
I0424 21:22:50.228563 139037601449792 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.040140s (batch_requests_ready=0.001786s, total_serialization_initiated=0.038295s, others=0.000060s)
I0424 21:22:50.228644 139037601449792 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.044529s (all_items=0.000011s, per_item={'items': '0.00001097'}, temp_paths=0.044518)
I0424 21:22:50.229342 139037601449792 event_tracking.py:125] [process=3] [async] Finished blocking save in 0.62 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints/9.
I0424 21:22:50.229672 138904625133312 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-24 21:42:50.229634
I0424 21:22:50.231680 139037601449792 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0424 21:22:50.231947 138907899393792 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0424 21:22:50.232106 139037601449792 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1777065742.2294798, 'wait_for_prev_duration_secs': 27.37320065498352, 'time_between_consecutive_saves_sec': 23.863330602645874, 'checkpointer_blocking_start_time': 1777065769.605181, 'checkpointer_blocking_duration_secs': 0.6246368885040283, 'get_old_steps_start_time': 1777065770.229836, 'get_old_steps_duration_secs': 2.4557113647460938e-05, 'checkpoint_manager_blocking_start_time': 1777065742.2272513, 'checkpoint_manager_blocking_duration_secs': 28.004822731018066}
I0424 21:22:50.232221 139037601449792 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0424 21:22:50.232263 139037601449792 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0424 21:22:55.825222 138908419479296 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_test_pipeline_scan_nnx_20260424_200834/linen_xpk_test_pipeline_scan_nnx_20260424_200834_16_zarr3_true/checkpoints/9/items/array_metadatas/process_3
I0424 21:23:32.576981 138904625133312 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 37.265 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.38849759101868 s) (per-host)
I0424 21:23:32.577103 138904625133312 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 42.347319s.
I0424 21:23:39.964123 138904625133312 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 49.734324s.
I0424 21:23:39.964411 138907899393792 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0424 21:23:39.964535 138907899393792 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0424 21:23:39.964576 138907899393792 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0424 21:23:39.966039 138907899393792 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0424 21:23:39.966203 139037601449792 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0424 21:23:39.966383 139037601449792 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0424 21:23:39.967299 139037601449792 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.386, Tokens/s/device: 13925.436, 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: Fri Apr 24 21:23:50 UTC 2026
EXIT_CODE=0