XPK Start: Tue Apr 21 12:22:20 UTC 2026 PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. 2026-04-21 12:22:44.251761: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303) I0421 12:22:44.784802 138650933319488 max_utils.py:273] Attempting to initialize the jax distributed system... I0421 12:22:53.826854 138650933319488 distributed.py:149] Starting JAX distributed service on [::]:8482 I0421 12:22:53.829236 138650933319488 distributed.py:172] Connecting to JAX distributed service on mt-11-optimizer-offload-i5gnn-slice-job-0-0.mt-11-optimizer-offload-i5gnn:8482 I0421 12:22:54.932614 138650933319488 max_utils.py:284] Jax distributed system initialized! I0421 12:23:00.988574 138650933319488 max_utils.py:800] System Information: Jax Version: 0.9.2 I0421 12:23:00.988673 138650933319488 max_utils.py:801] System Information: Jaxlib Version: 0.9.2 I0421 12:23:00.988712 138650933319488 max_utils.py:802] System Information: Jax Backend: PJRT C API TFRT TPU v6 lite Built on Mar 4 2026 11:32:08 (1772652728) cl/878335365 I0421 12:23:00.988768 138650933319488 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0421 12:23:01.695586 138650933319488 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0421 12:23:01.696182 138650933319488 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0421 12:23:01.696355 138650933319488 checkpointing.py:688] Setting up checkpoint logger... I0421 12:23:01.696404 138650933319488 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0421 12:23:01.696456 138650933319488 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None I0421 12:23:01.696792 138650933319488 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 0x7e19a351de20>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0421 12:23:04.575530 138650933319488 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0421 12:23:04.575784 138650933319488 checkpoint_manager.py:708] [process=7][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e0610498ef0>}, handler_registry=None I0421 12:23:04.576021 138650933319488 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e0610498ef0>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`. I0421 12:23:04.576074 138650933319488 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e061033e450>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`. I0421 12:23:04.576110 138650933319488 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 0x7e0610498ef0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7e0610498ef0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e061033e450>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7e061033e450>}). I0421 12:23:04.576415 138650933319488 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34 I0421 12:23:04.576484 138650933319488 async_checkpointer.py:192] [process=7][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7e05f07d4180> timeout: 1200 secs and primary_host=0 for async checkpoint writes I0421 12:23:05.691682 138650933319488 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260421_114057/linen_xpk_feat_nnx_post_train_fixes_20260421_114057_11_optimizer_offload_true/checkpoints I0421 12:23:06.117351 138650933319488 checkpoint_manager.py:929] [process=7][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260421_114057/linen_xpk_feat_nnx_post_train_fixes_20260421_114057_11_optimizer_offload_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7e06103409e0> I0421 12:23:06.117517 138650933319488 checkpointing.py:302] Checkpoint manager created! I0421 12:23:07.044650 138650933319488 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0421 12:23:07.044765 138650933319488 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0421 12:23:07.424973 138650933319488 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0421 12:23:07.425061 138650933319488 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0421 12:23:07.441278 138650933319488 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0421 12:23:07.441334 138650933319488 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0421 12:23:07.465159 138650933319488 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0421 12:23:07.465227 138650933319488 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0421 12:23:07.481573 138650933319488 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0421 12:23:07.481631 138650933319488 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0421 12:23:07.497958 138650933319488 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0421 12:23:07.498020 138650933319488 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0421 12:23:07.523516 138650933319488 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0421 12:23:07.523584 138650933319488 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0421 12:23:07.544499 138650933319488 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0421 12:23:07.544565 138650933319488 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0421 12:23:07.760841 138650933319488 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0421 12:23:07.760949 138650933319488 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0421 12:23:09.413245 138650933319488 maxtext_utils.py:1827] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0421 12:23:09.413369 138650933319488 maxtext_utils.py:1827] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: P('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0421 12:23:09.413423 138650933319488 maxtext_utils.py:1827] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: P('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0421 12:23:09.413480 138650933319488 maxtext_utils.py:1827] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: P('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0421 12:23:09.413533 138650933319488 maxtext_utils.py:1827] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: P('norm', 'layers') Physical: (None, None) I0421 12:23:09.413571 138650933319488 maxtext_utils.py:1827] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: P('norm', 'layers') Physical: (None, None) I0421 12:23:09.413621 138650933319488 maxtext_utils.py:1827] params/params/decoder/layers/self_attention/key/kernel Shape: float32[2048,16,16,128] Logical: P('embed', 'layers', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None, None) I0421 12:23:09.413673 138650933319488 maxtext_utils.py:1827] params/params/decoder/layers/self_attention/out/kernel Shape: float32[16,16,128,2048] Logical: P('heads', 'layers', 'kv', 'embed') Physical: (None, None, None, 'fsdp') I0421 12:23:09.413712 138650933319488 maxtext_utils.py:1827] params/params/decoder/layers/self_attention/query/kernel Shape: float32[2048,16,16,128] Logical: P('embed', 'layers', 'q_heads', 'kv') Physical: ('fsdp', None, None, None) I0421 12:23:09.413831 138650933319488 maxtext_utils.py:1827] params/params/decoder/layers/self_attention/value/kernel Shape: float32[2048,16,16,128] Logical: P('embed', 'layers', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None, None) I0421 12:23:09.413890 138650933319488 maxtext_utils.py:1827] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: P('embed_vocab', 'vocab') Physical: ('fsdp', None) I0421 12:23:09.413940 138650933319488 maxtext_utils.py:1827] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: P('vocab', 'embed_vocab') Physical: (None, 'fsdp') I0421 12:23:09.907640 138650933319488 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0421 12:23:09.907744 138650933319488 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0421 12:23:09.923088 138650933319488 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0421 12:23:09.923150 138650933319488 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0421 12:23:21.070577 138650933319488 max_utils.py:791] Total memory size: 1.3 GB, Output size: 0.1 GB, Temp size: 1.1 GB, Argument size: 0.1 GB, Host temp size: 0.0 GB. I0421 12:23:21.071413 138650933319488 metric_logger.py:301] number parameters: 1.104 billion I0421 12:23:32.562522 138650933319488 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0421 12:23:32.730263 138650933319488 checkpointing.py:798] Waited 0.16772794723510742 seconds for step 0 to finish before starting checkpointing. I0421 12:23:32.733342 138650933319488 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0421 12:23:32.734987 138650933319488 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 0 I0421 12:23:32.736367 138650933319488 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260421_114057/linen_xpk_feat_nnx_post_train_fixes_20260421_114057_11_optimizer_offload_true/checkpoints/0. I0421 12:23:33.085914 138650933319488 signaling_client.py:364] Using JaxDistributedSignalingClient I0421 12:23:33.086754 138650933319488 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0421 12:23:33.086809 138650933319488 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0421 12:23:33.360873 138650933319488 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.275037s I0421 12:23:33.361044 138650933319488 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 5.480 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28147172927856445 s) (per-host) I0421 12:23:33.361101 138650933319488 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.281538s (batch_requests_ready=0.002260s, total_serialization_initiated=0.279204s, others=0.000074s) I0421 12:23:33.361195 138650933319488 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.285701s (all_items=0.000025s, per_item={'items': '0.00002456'}, temp_paths=0.285676) I0421 12:23:33.362051 138650933319488 event_tracking.py:125] [process=7] [async] Finished blocking save in 0.63 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260421_114057/linen_xpk_feat_nnx_post_train_fixes_20260421_114057_11_optimizer_offload_true/checkpoints/0. I0421 12:23:33.362380 138520328918784 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 12:43:33.362342 I0421 12:23:33.371726 138650933319488 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0421 12:23:33.371958 138519759423232 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save. I0421 12:23:33.372083 138650933319488 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260421_114057/linen_xpk_feat_nnx_post_train_fixes_20260421_114057_11_optimizer_offload_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776774212.7333238, 'wait_for_prev_duration_secs': 6.580352783203125e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776774212.7350254, 'checkpointer_blocking_duration_secs': 0.6275146007537842, 'get_old_steps_start_time': 1776774213.3625653, 'get_old_steps_duration_secs': 2.9087066650390625e-05, 'checkpoint_manager_blocking_start_time': 1776774212.7307863, 'checkpoint_manager_blocking_duration_secs': 0.6412546634674072} I0421 12:23:33.372186 138650933319488 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0421 12:23:33.372231 138650933319488 max_utils.py:750] Memstats: After params initialized: I0421 12:23:33.372278 138650933319488 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_26(process=7,(2,6,0,0)) I0421 12:23:33.372314 138650933319488 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_27(process=7,(3,6,0,0)) I0421 12:23:33.372340 138650933319488 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_30(process=7,(2,7,0,0)) I0421 12:23:33.372364 138650933319488 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_31(process=7,(3,7,0,0)) I0421 12:23:33.687974 138650933319488 metric_logger.py:196] completed step: 0, seconds: 11.491, TFLOP/s/device: 1.182, Tokens/s/device: 178.227, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0421 12:23:33.878324 138650933319488 metric_logger.py:196] completed step: 1, seconds: 1.124, TFLOP/s/device: 12.088, Tokens/s/device: 1822.095, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0421 12:23:34.293577 138650933319488 metric_logger.py:196] completed step: 2, seconds: 0.028, TFLOP/s/device: 490.705, Tokens/s/device: 73964.390, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129 I0421 12:23:34.456562 138650933319488 metric_logger.py:196] completed step: 3, seconds: 0.415, TFLOP/s/device: 32.717, Tokens/s/device: 4931.422, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.047 I0421 12:23:34.782386 138650933319488 metric_logger.py:196] completed step: 4, seconds: 0.169, TFLOP/s/device: 80.607, Tokens/s/device: 12149.976, total_weights: 65536, loss: 9.278, lm_loss: 9.278, perplexity: 10694.968 I0421 12:23:34.788381 138650933319488 metric_logger.py:196] completed step: 5, seconds: 0.163, TFLOP/s/device: 83.369, Tokens/s/device: 12566.345, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.520 I0421 12:23:36.287769 2726 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0421 12:23:38.322190 138519767815936 array_metadata_store.py:203] [process=7][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260421_114057/linen_xpk_feat_nnx_post_train_fixes_20260421_114057_11_optimizer_offload_true/checkpoints/0/items/array_metadatas/process_7 I0421 12:23:56.645029 138650933319488 metric_logger.py:196] completed step: 6, seconds: 0.326, TFLOP/s/device: 41.629, Tokens/s/device: 6274.798, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.688 I0421 12:23:56.807698 138650933319488 metric_logger.py:196] completed step: 7, seconds: 21.695, TFLOP/s/device: 0.626, Tokens/s/device: 94.402, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4377.010 I0421 12:23:56.970483 138650933319488 metric_logger.py:196] completed step: 8, seconds: 0.168, TFLOP/s/device: 80.987, Tokens/s/device: 12207.261, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3845.943 I0421 12:23:57.132555 138650933319488 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0421 12:23:57.133229 138650933319488 checkpointing.py:798] Waited 0.0006978511810302734 seconds for step 9 to finish before starting checkpointing. I0421 12:23:57.135233 138650933319488 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0421 12:24:09.407642 138520328918784 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 43.480 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.32803249359131 s) (per-host) I0421 12:24:09.407764 138520328918784 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.045265s. I0421 12:24:20.411900 138520328918784 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 47.049384s. I0421 12:24:20.412128 138519759423232 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save. I0421 12:24:20.412186 138519759423232 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save. I0421 12:24:20.412245 138519759423232 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0421 12:24:20.414995 138519759423232 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0421 12:24:20.415162 138650933319488 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0421 12:24:20.415295 138650933319488 checkpoint_manager.py:1452] Waiting for previous save to complete took 23.280062 seconds. If this number is high, consider checkpointing less frequently. I0421 12:24:20.417093 138650933319488 checkpoint_manager.py:1512] [process=7] Saving checkpoint at step 9 I0421 12:24:20.419234 138650933319488 event_tracking.py:70] [process=7] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260421_114057/linen_xpk_feat_nnx_post_train_fixes_20260421_114057_11_optimizer_offload_true/checkpoints/9. I0421 12:24:20.717227 138650933319488 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0421 12:24:20.717319 138650933319488 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0421 12:24:20.751212 138650933319488 base_pytree_checkpoint_handler.py:154] [process=7][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.034944s I0421 12:24:20.751379 138650933319488 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/blocking_gbytes_per_sec: 40.340 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.0382380485534668 s) (per-host) I0421 12:24:20.751428 138650933319488 base_pytree_checkpoint_handler.py:768] [process=7][thread=MainThread] Initiated Pytree async_save. Time taken: 0.038296s (batch_requests_ready=0.001760s, total_serialization_initiated=0.036470s, others=0.000066s) I0421 12:24:20.751515 138650933319488 composite_checkpoint_handler.py:715] [process=7][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.042425s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=0.042409) I0421 12:24:20.752239 138650933319488 event_tracking.py:125] [process=7] [async] Finished blocking save in 0.34 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260421_114057/linen_xpk_feat_nnx_post_train_fixes_20260421_114057_11_optimizer_offload_true/checkpoints/9. I0421 12:24:20.752579 138519759423232 async_checkpointer.py:76] [process=7][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 12:44:20.752544 I0421 12:24:20.754959 138650933319488 checkpoint_manager.py:1560] [process=7][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0421 12:24:20.755239 138519767815936 async_checkpointer.py:280] [process=7][thread=save_finalize] Waiting for background save thread=async_save. I0421 12:24:20.755399 138650933319488 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260421_114057/linen_xpk_feat_nnx_post_train_fixes_20260421_114057_11_optimizer_offload_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776774237.1352024, 'wait_for_prev_duration_secs': 23.280061960220337, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776774260.4171338, 'checkpointer_blocking_duration_secs': 0.33559107780456543, 'get_old_steps_start_time': 1776774260.7527597, 'get_old_steps_duration_secs': 3.0517578125e-05, 'checkpoint_manager_blocking_start_time': 1776774237.1334546, 'checkpoint_manager_blocking_duration_secs': 23.62191104888916} I0421 12:24:20.755503 138650933319488 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0421 12:24:20.755548 138650933319488 checkpoint_manager.py:2020] [process=7][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0421 12:24:26.335646 138496287160064 array_metadata_store.py:203] [process=7][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260421_114057/linen_xpk_feat_nnx_post_train_fixes_20260421_114057_11_optimizer_offload_true/checkpoints/9/items/array_metadatas/process_7 I0421 12:25:02.187876 138519759423232 base_pytree_checkpoint_handler.py:130] [process=7] /jax/orbax/write/gbytes_per_sec: 38.084 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.474695920944214 s) (per-host) I0421 12:25:02.188009 138519759423232 async_checkpointer.py:90] [process=7][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.435319s. I0421 12:25:10.965957 138519759423232 async_checkpointer.py:160] [process=7][thread=async_save] Background save thread done. Time taken: 50.213251s. I0421 12:25:10.966259 138519767815936 async_checkpointer.py:288] [process=7][thread=save_finalize] Done with waiting for background save thread=async_save. I0421 12:25:10.966372 138519767815936 async_checkpointer.py:298] [process=7][thread=save_finalize] No errors found in background save thread=async_save. I0421 12:25:10.966419 138519767815936 checkpoint_manager.py:2137] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0421 12:25:10.967857 138519767815936 checkpoint_manager.py:2146] [process=7][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0421 12:25:10.968029 138650933319488 checkpoint_manager.py:2032] [process=7][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0421 12:25:10.968178 138650933319488 checkpoint_manager.py:2009] [process=7][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0421 12:25:10.969140 138650933319488 metric_logger.py:196] completed step: 9, seconds: 0.162, TFLOP/s/device: 83.622, Tokens/s/device: 12604.396, total_weights: 65536, loss: 8.179, lm_loss: 8.179, perplexity: 3564.739 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Tue Apr 21 12:25:20 UTC 2026 EXIT_CODE=0