XPK Start: Sun Apr 19 00:27:22 UTC 2026 PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. 2026-04-19 00:27:46.799417: 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) I0419 00:27:46.980060 137633445746496 max_utils.py:273] Attempting to initialize the jax distributed system... I0419 00:27:56.020838 137633445746496 distributed.py:149] Starting JAX distributed service on [::]:8482 I0419 00:27:56.023205 137633445746496 distributed.py:172] Connecting to JAX distributed service on mt-11-optimizer-offload-i3h26-slice-job-0-0.mt-11-optimizer-offload-i3h26:8482 I0419 00:27:57.751448 137633445746496 max_utils.py:284] Jax distributed system initialized! I0419 00:28:04.088577 137633445746496 max_utils.py:800] System Information: Jax Version: 0.9.2 I0419 00:28:04.088678 137633445746496 max_utils.py:801] System Information: Jaxlib Version: 0.9.2 I0419 00:28:04.088719 137633445746496 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 I0419 00:28:04.088754 137633445746496 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0419 00:28:05.155566 137633445746496 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 00:28:05.156136 137633445746496 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 00:28:05.156332 137633445746496 checkpointing.py:688] Setting up checkpoint logger... I0419 00:28:05.156385 137633445746496 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0419 00:28:05.156430 137633445746496 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None I0419 00:28:05.156778 137633445746496 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 0x7d2cb9d18380>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0419 00:28:07.665228 137633445746496 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0419 00:28:07.665475 137633445746496 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 0x7d2b8506a150>}, handler_registry=None I0419 00:28:07.665719 137633445746496 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d2b8506a150>` 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`. I0419 00:28:07.665770 137633445746496 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d2b8527f680>` 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`. I0419 00:28:07.665806 137633445746496 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 0x7d2b8506a150>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7d2b8506a150>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d2b8527f680>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7d2b8527f680>}). I0419 00:28:07.666126 137633445746496 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34 I0419 00:28:07.666196 137633445746496 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7d17547dac00> timeout: 1200 secs and primary_host=0 for async checkpoint writes I0419 00:28:09.944633 137633445746496 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_11_optimizer_offload_true/checkpoints I0419 00:28:09.970190 137633445746496 checkpoint_manager.py:929] [process=6][thread=MainThread] CheckpointManager created, primary_host=0, CheckpointManagerOptions=CheckpointManagerOptions(save_interval_steps=1, max_to_keep=None, keep_time_interval=None, keep_period=None, should_keep_fn=None, best_fn=None, best_mode='max', keep_checkpoints_without_metrics=True, step_prefix=None, step_format_fixed_length=None, step_name_format=None, create=True, cleanup_tmp_directories=False, save_on_steps=frozenset(), single_host_load_and_broadcast=False, todelete_subdir=None, todelete_full_path=None, enable_background_delete=False, read_only=False, enable_async_checkpointing=True, async_options=None, multiprocessing_options=MultiprocessingOptions(primary_host=0, active_processes=None, barrier_sync_key_prefix=None), should_save_fn=None, file_options=FileOptions(path_permission_mode=None), save_root_metadata=True, temporary_path_class=None, save_decision_policy=FixedIntervalPolicy(interval=10), preservation_policy=LatestN(n=None), prevent_write_metrics=False, enable_should_save_is_saving_in_progress_check=True, enable_per_process_directory_creation=False, lightweight_initialize=False), root_directory=gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_11_optimizer_offload_true/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7d2b85069130> I0419 00:28:09.970338 137633445746496 checkpointing.py:302] Checkpoint manager created! I0419 00:28:10.912039 137633445746496 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0419 00:28:10.912143 137633445746496 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 00:28:11.292907 137633445746496 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 00:28:11.292998 137633445746496 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 00:28:11.309206 137633445746496 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 00:28:11.309274 137633445746496 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 00:28:11.332888 137633445746496 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 00:28:11.332954 137633445746496 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 00:28:11.349294 137633445746496 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 00:28:11.349348 137633445746496 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 00:28:11.365561 137633445746496 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 00:28:11.365621 137633445746496 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 00:28:11.390151 137633445746496 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0419 00:28:11.390218 137633445746496 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 00:28:11.411802 137633445746496 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0419 00:28:11.411872 137633445746496 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0419 00:28:11.623787 137633445746496 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0419 00:28:11.623895 137633445746496 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0419 00:28:13.190491 137633445746496 maxtext_utils.py:1827] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:28:13.190618 137633445746496 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) I0419 00:28:13.190671 137633445746496 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) I0419 00:28:13.190730 137633445746496 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') I0419 00:28:13.190784 137633445746496 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) I0419 00:28:13.190824 137633445746496 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) I0419 00:28:13.190876 137633445746496 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) I0419 00:28:13.190930 137633445746496 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') I0419 00:28:13.190970 137633445746496 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) I0419 00:28:13.191008 137633445746496 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) I0419 00:28:13.191058 137633445746496 maxtext_utils.py:1827] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: P('embed_vocab', 'vocab') Physical: ('fsdp', None) I0419 00:28:13.191105 137633445746496 maxtext_utils.py:1827] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: P('vocab', 'embed_vocab') Physical: (None, 'fsdp') I0419 00:28:13.689429 137633445746496 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 00:28:13.689523 137633445746496 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0419 00:28:13.704892 137633445746496 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 00:28:13.704955 137633445746496 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0419 00:28:24.732994 137633445746496 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. I0419 00:28:24.733873 137633445746496 metric_logger.py:301] number parameters: 1.104 billion I0419 00:28:36.129437 137633445746496 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0419 00:28:36.298491 137633445746496 checkpointing.py:798] Waited 0.16903424263000488 seconds for step 0 to finish before starting checkpointing. I0419 00:28:36.300909 137633445746496 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 00:28:36.302477 137633445746496 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0 I0419 00:28:36.304099 137633445746496 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_11_optimizer_offload_true/checkpoints/0. I0419 00:28:37.039390 137633445746496 signaling_client.py:364] Using JaxDistributedSignalingClient I0419 00:28:37.040430 137633445746496 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0419 00:28:37.040486 137633445746496 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 I0419 00:28:37.319772 137633445746496 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.280454s I0419 00:28:37.319941 137633445746496 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.377 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.2868976593017578 s) (per-host) I0419 00:28:37.319992 137633445746496 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.286959s (batch_requests_ready=0.002222s, total_serialization_initiated=0.284668s, others=0.000068s) I0419 00:28:37.320085 137633445746496 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.291049s (all_items=0.000019s, per_item={'items': '0.00001931'}, temp_paths=0.291029) I0419 00:28:37.320853 137633445746496 event_tracking.py:125] [process=6] [async] Finished blocking save in 1.02 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_11_optimizer_offload_true/checkpoints/0. I0419 00:28:37.321135 137498373187328 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 00:48:37.321110 I0419 00:28:37.323123 137633445746496 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0419 00:28:37.323411 137486466545408 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0419 00:28:37.323565 137633445746496 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_11_optimizer_offload_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776558516.3008904, 'wait_for_prev_duration_secs': 6.0558319091796875e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776558516.302515, 'checkpointer_blocking_duration_secs': 1.018751859664917, 'get_old_steps_start_time': 1776558517.3213003, 'get_old_steps_duration_secs': 2.9802322387695312e-05, 'checkpoint_manager_blocking_start_time': 1776558516.299018, 'checkpoint_manager_blocking_duration_secs': 1.0245084762573242} I0419 00:28:37.323667 137633445746496 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0419 00:28:37.323715 137633445746496 max_utils.py:750] Memstats: After params initialized: I0419 00:28:37.323760 137633445746496 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_24(process=6,(0,6,0,0)) I0419 00:28:37.323791 137633445746496 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_25(process=6,(1,6,0,0)) I0419 00:28:37.323816 137633445746496 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_28(process=6,(0,7,0,0)) I0419 00:28:37.323839 137633445746496 max_utils.py:756] Using (GB) 0.17 / 31.25 (0.544000%) on TPU_29(process=6,(1,7,0,0)) I0419 00:28:37.634044 137633445746496 metric_logger.py:196] completed step: 0, seconds: 11.395, TFLOP/s/device: 1.192, Tokens/s/device: 179.721, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0419 00:28:37.835204 137633445746496 metric_logger.py:196] completed step: 1, seconds: 1.503, TFLOP/s/device: 9.040, Tokens/s/device: 1362.548, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0419 00:28:38.217100 137633445746496 metric_logger.py:196] completed step: 2, seconds: 0.038, TFLOP/s/device: 353.509, Tokens/s/device: 53284.766, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129 I0419 00:28:38.379925 137633445746496 metric_logger.py:196] completed step: 3, seconds: 0.382, TFLOP/s/device: 35.553, Tokens/s/device: 5358.886, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.047 I0419 00:28:38.705953 137633445746496 metric_logger.py:196] completed step: 4, seconds: 0.168, TFLOP/s/device: 80.642, Tokens/s/device: 12155.313, total_weights: 65536, loss: 9.278, lm_loss: 9.278, perplexity: 10694.968 I0419 00:28:38.712047 137633445746496 metric_logger.py:196] completed step: 5, seconds: 0.163, TFLOP/s/device: 83.504, Tokens/s/device: 12586.656, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.520 I0419 00:28:41.499805 2773 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0419 00:28:44.912901 137498356401920 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_11_optimizer_offload_true/checkpoints/0/items/array_metadatas/process_6 I0419 00:29:00.214021 137633445746496 metric_logger.py:196] completed step: 6, seconds: 0.327, TFLOP/s/device: 41.583, Tokens/s/device: 6267.885, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.688 I0419 00:29:00.376923 137633445746496 metric_logger.py:196] completed step: 7, seconds: 21.340, TFLOP/s/device: 0.637, Tokens/s/device: 95.970, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4377.010 I0419 00:29:00.539587 137633445746496 metric_logger.py:196] completed step: 8, seconds: 0.168, TFLOP/s/device: 81.072, Tokens/s/device: 12220.081, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3845.943 I0419 00:29:00.701809 137633445746496 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0419 00:29:00.702491 137633445746496 checkpointing.py:798] Waited 0.0007033348083496094 seconds for step 9 to finish before starting checkpointing. I0419 00:29:00.704512 137633445746496 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 00:29:15.821626 137498373187328 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 40.722 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 38.788562536239624 s) (per-host) I0419 00:29:15.821731 137498373187328 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 38.500510s. I0419 00:29:25.494388 137498373187328 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 48.173152s. I0419 00:29:25.494673 137486466545408 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 00:29:25.494794 137486466545408 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0419 00:29:25.494845 137486466545408 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0419 00:29:25.496470 137486466545408 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0419 00:29:25.496599 137633445746496 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. W0419 00:29:25.496666 137633445746496 checkpoint_manager.py:1452] Waiting for previous save to complete took 24.792163 seconds. If this number is high, consider checkpointing less frequently. I0419 00:29:25.498591 137633445746496 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9 I0419 00:29:25.500663 137633445746496 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_11_optimizer_offload_true/checkpoints/9. I0419 00:29:26.203714 137633445746496 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0419 00:29:26.203806 137633445746496 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 I0419 00:29:26.240698 137633445746496 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.038008s I0419 00:29:26.240872 137633445746496 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 37.321 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.041330575942993164 s) (per-host) I0419 00:29:26.240922 137633445746496 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.041391s (batch_requests_ready=0.001781s, total_serialization_initiated=0.039541s, others=0.000069s) I0419 00:29:26.241013 137633445746496 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.045611s (all_items=0.000016s, per_item={'items': '0.00001621'}, temp_paths=0.045595) I0419 00:29:26.241714 137633445746496 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.74 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_11_optimizer_offload_true/checkpoints/9. I0419 00:29:26.242055 137478933567232 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 00:49:26.242019 I0419 00:29:26.244343 137633445746496 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0419 00:29:26.244628 137486466545408 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0419 00:29:26.244765 137633445746496 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_11_optimizer_offload_true/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776558540.70448, 'wait_for_prev_duration_secs': 24.792162656784058, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776558565.4986293, 'checkpointer_blocking_duration_secs': 0.7435343265533447, 'get_old_steps_start_time': 1776558566.2421865, 'get_old_steps_duration_secs': 3.123283386230469e-05, 'checkpoint_manager_blocking_start_time': 1776558540.7027147, 'checkpoint_manager_blocking_duration_secs': 25.54201078414917} I0419 00:29:26.244874 137633445746496 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0419 00:29:26.244925 137633445746496 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 00:29:33.446857 137498356401920 array_metadata_store.py:203] [process=6][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260418_235007/linen_xpk_feat_nnx_post_train_fixes_20260418_235007_11_optimizer_offload_true/checkpoints/9/items/array_metadatas/process_6 I0419 00:30:09.862959 137478933567232 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 36.175 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.66337871551514 s) (per-host) I0419 00:30:09.863087 137478933567232 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.620300s. I0419 00:30:20.067528 137478933567232 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 53.824724s. I0419 00:30:20.067823 137486466545408 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 00:30:20.067965 137486466545408 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0419 00:30:20.068023 137486466545408 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0419 00:30:20.069716 137486466545408 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0419 00:30:20.069900 137633445746496 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. I0419 00:30:20.070056 137633445746496 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 00:30:20.070979 137633445746496 metric_logger.py:196] completed step: 9, seconds: 0.163, TFLOP/s/device: 83.555, Tokens/s/device: 12594.319, 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: Sun Apr 19 00:30:32 UTC 2026 EXIT_CODE=0