XPK Start: Fri Apr 17 03:30:42 UTC 2026 PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. 2026-04-17 03:31:07.275147: 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) I0417 03:31:07.452695 136554047944512 max_utils.py:273] Attempting to initialize the jax distributed system... I0417 03:31:16.494760 136554047944512 distributed.py:149] Starting JAX distributed service on [::]:8482 I0417 03:31:16.497505 136554047944512 distributed.py:172] Connecting to JAX distributed service on mt-11-optimizer-offload-z3i7a-slice-job-0-0.mt-11-optimizer-offload-z3i7a:8482 I0417 03:31:17.318645 136554047944512 max_utils.py:284] Jax distributed system initialized! I0417 03:31:23.447244 136554047944512 max_utils.py:800] System Information: Jax Version: 0.9.2 I0417 03:31:23.447345 136554047944512 max_utils.py:801] System Information: Jaxlib Version: 0.9.2 I0417 03:31:23.447385 136554047944512 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 I0417 03:31:23.447419 136554047944512 train_utils.py:377] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0417 03:31:24.127968 136554047944512 maxtext_utils.py:1687] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0417 03:31:24.128539 136554047944512 maxtext_utils.py:1687] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0417 03:31:24.128720 136554047944512 checkpointing.py:688] Setting up checkpoint logger... I0417 03:31:24.128769 136554047944512 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0417 03:31:24.128811 136554047944512 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None I0417 03:31:24.129145 136554047944512 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 0x7c316611c500>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0417 03:31:27.483890 136554047944512 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0417 03:31:27.484124 136554047944512 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 0x7c1d4c3f87a0>}, handler_registry=None I0417 03:31:27.484354 136554047944512 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c1d4c3f87a0>` 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`. I0417 03:31:27.484400 136554047944512 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c1d4c3f8f50>` 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`. I0417 03:31:27.484434 136554047944512 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 0x7c1d4c3f87a0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7c1d4c3f87a0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c1d4c3f8f50>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7c1d4c3f8f50>}). I0417 03:31:27.484751 136554047944512 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34 I0417 03:31:27.484833 136554047944512 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7c1d4c1887c0> timeout: 1200 secs and primary_host=0 for async checkpoint writes I0417 03:31:28.259970 136554047944512 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/linen_xpk_feat_nnx_post_train_fixes_20260417_023710_11_optimizer_offload_false/checkpoints I0417 03:31:28.632111 136554047944512 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_20260417_023710/linen_xpk_feat_nnx_post_train_fixes_20260417_023710_11_optimizer_offload_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7c1d4c5f2fc0> I0417 03:31:28.632278 136554047944512 checkpointing.py:302] Checkpoint manager created! I0417 03:31:29.596263 136554047944512 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0417 03:31:29.596362 136554047944512 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0417 03:31:29.992421 136554047944512 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0417 03:31:29.992519 136554047944512 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0417 03:31:30.008891 136554047944512 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0417 03:31:30.008947 136554047944512 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0417 03:31:30.032509 136554047944512 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0417 03:31:30.032576 136554047944512 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0417 03:31:30.049018 136554047944512 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0417 03:31:30.049078 136554047944512 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0417 03:31:30.065720 136554047944512 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0417 03:31:30.065782 136554047944512 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0417 03:31:30.095469 136554047944512 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0417 03:31:30.095569 136554047944512 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0417 03:31:30.117326 136554047944512 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0417 03:31:30.117393 136554047944512 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0417 03:31:30.332037 136554047944512 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0417 03:31:30.332144 136554047944512 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0417 03:31:31.750077 136554047944512 maxtext_utils.py:1796] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0417 03:31:31.750205 136554047944512 maxtext_utils.py:1796] params/params/decoder/layers/mlp/wi_0/kernel Shape: float32[2048,16,7168] Logical: P('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0417 03:31:31.750260 136554047944512 maxtext_utils.py:1796] params/params/decoder/layers/mlp/wi_1/kernel Shape: float32[2048,16,7168] Logical: P('embed', 'layers', 'mlp') Physical: ('fsdp', None, None) I0417 03:31:31.750318 136554047944512 maxtext_utils.py:1796] params/params/decoder/layers/mlp/wo/kernel Shape: float32[7168,16,2048] Logical: P('mlp', 'layers', 'embed') Physical: (None, None, 'fsdp') I0417 03:31:31.750370 136554047944512 maxtext_utils.py:1796] params/params/decoder/layers/post_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: P('norm', 'layers') Physical: (None, None) I0417 03:31:31.750409 136554047944512 maxtext_utils.py:1796] params/params/decoder/layers/pre_self_attention_layer_norm/scale Shape: float32[2048,16] Logical: P('norm', 'layers') Physical: (None, None) I0417 03:31:31.750461 136554047944512 maxtext_utils.py:1796] 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) I0417 03:31:31.750514 136554047944512 maxtext_utils.py:1796] 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') I0417 03:31:31.750555 136554047944512 maxtext_utils.py:1796] 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) I0417 03:31:31.750591 136554047944512 maxtext_utils.py:1796] 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) I0417 03:31:31.750639 136554047944512 maxtext_utils.py:1796] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: P('embed_vocab', 'vocab') Physical: ('fsdp', None) I0417 03:31:31.750688 136554047944512 maxtext_utils.py:1796] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: P('vocab', 'embed_vocab') Physical: (None, 'fsdp') I0417 03:31:32.241601 136554047944512 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0417 03:31:32.241698 136554047944512 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0417 03:31:32.257414 136554047944512 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0417 03:31:32.257475 136554047944512 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0417 03:31:43.054279 136554047944512 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. I0417 03:31:43.055033 136554047944512 metric_logger.py:289] number parameters: 1.104 billion I0417 03:31:54.427754 136554047944512 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0417 03:31:54.581616 136554047944512 checkpointing.py:798] Waited 0.15384626388549805 seconds for step 0 to finish before starting checkpointing. I0417 03:31:54.584042 136554047944512 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0417 03:31:54.585907 136554047944512 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0 I0417 03:31:54.587268 136554047944512 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/linen_xpk_feat_nnx_post_train_fixes_20260417_023710_11_optimizer_offload_false/checkpoints/0. I0417 03:31:55.330271 136554047944512 signaling_client.py:364] Using JaxDistributedSignalingClient I0417 03:31:55.331318 136554047944512 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0417 03:31:55.331373 136554047944512 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 I0417 03:31:55.606011 136554047944512 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.275812s I0417 03:31:55.606182 136554047944512 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.486 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28119325637817383 s) (per-host) I0417 03:31:55.606236 136554047944512 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.281255s (batch_requests_ready=0.002174s, total_serialization_initiated=0.279011s, others=0.000070s) I0417 03:31:55.606324 136554047944512 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.285299s (all_items=0.000016s, per_item={'items': '0.00001621'}, temp_paths=0.285283) I0417 03:31:55.607200 136554047944512 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_20260417_023710/linen_xpk_feat_nnx_post_train_fixes_20260417_023710_11_optimizer_offload_false/checkpoints/0. I0417 03:31:55.607500 136427000575744 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-17 03:51:55.607466 I0417 03:31:55.615710 136554047944512 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0417 03:31:55.615982 136425619449600 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0417 03:31:55.616115 136554047944512 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/linen_xpk_feat_nnx_post_train_fixes_20260417_023710_11_optimizer_offload_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776396714.584023, 'wait_for_prev_duration_secs': 6.246566772460938e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776396714.5859456, 'checkpointer_blocking_duration_secs': 1.0217175483703613, 'get_old_steps_start_time': 1776396715.6076894, 'get_old_steps_duration_secs': 3.0517578125e-05, 'checkpoint_manager_blocking_start_time': 1776396714.5821383, 'checkpoint_manager_blocking_duration_secs': 1.0339372158050537} I0417 03:31:55.616222 136554047944512 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0417 03:31:55.616272 136554047944512 max_utils.py:750] Memstats: After params initialized: I0417 03:31:55.616321 136554047944512 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0)) I0417 03:31:55.616352 136554047944512 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0)) I0417 03:31:55.616377 136554047944512 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0)) I0417 03:31:55.616402 136554047944512 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0)) I0417 03:31:55.928805 136554047944512 metric_logger.py:185] completed step: 0, seconds: 11.373, TFLOP/s/device: 1.195, Tokens/s/device: 180.082, total_weights: 65536, loss: 10.874 I0417 03:31:56.100806 136554047944512 metric_logger.py:185] completed step: 1, seconds: 1.500, TFLOP/s/device: 9.059, Tokens/s/device: 1365.532, total_weights: 65536, loss: 10.874 I0417 03:31:56.531566 136554047944512 metric_logger.py:185] completed step: 2, seconds: 0.025, TFLOP/s/device: 550.844, Tokens/s/device: 83029.271, total_weights: 65536, loss: 10.262 I0417 03:31:56.678998 136554047944512 metric_logger.py:185] completed step: 3, seconds: 0.431, TFLOP/s/device: 31.552, Tokens/s/device: 4755.834, total_weights: 65536, loss: 9.734 I0417 03:31:56.973981 136554047944512 metric_logger.py:185] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.623, Tokens/s/device: 13358.206, total_weights: 65536, loss: 9.277 I0417 03:31:56.979436 136554047944512 metric_logger.py:185] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.246, Tokens/s/device: 13904.259, total_weights: 65536, loss: 8.892 I0417 03:32:00.361414 2821 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0417 03:32:02.375979 136425627842304 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_20260417_023710/linen_xpk_feat_nnx_post_train_fixes_20260417_023710_11_optimizer_offload_false/checkpoints/0/items/array_metadatas/process_6 I0417 03:32:19.338488 136554047944512 metric_logger.py:185] completed step: 6, seconds: 0.295, TFLOP/s/device: 46.024, Tokens/s/device: 6937.270, total_weights: 65536, loss: 8.592 I0417 03:32:19.485715 136554047944512 metric_logger.py:185] completed step: 7, seconds: 22.212, TFLOP/s/device: 0.612, Tokens/s/device: 92.202, total_weights: 65536, loss: 8.384 I0417 03:32:19.633016 136554047944512 metric_logger.py:185] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.182, Tokens/s/device: 13442.466, total_weights: 65536, loss: 8.255 I0417 03:32:19.780144 136554047944512 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0417 03:32:19.780721 136554047944512 checkpointing.py:798] Waited 0.0005972385406494141 seconds for step 9 to finish before starting checkpointing. I0417 03:32:19.782862 136554047944512 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0417 03:32:32.195328 136427000575744 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 42.842 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.87029790878296 s) (per-host) I0417 03:32:32.195456 136427000575744 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 36.587836s. I0417 03:32:42.383612 136427000575744 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 46.775970s. I0417 03:32:42.383907 136425619449600 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0417 03:32:42.383972 136425619449600 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0417 03:32:42.384020 136425619449600 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0417 03:32:42.385929 136425619449600 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0417 03:32:42.386097 136554047944512 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. W0417 03:32:42.386256 136554047944512 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.603394 seconds. If this number is high, consider checkpointing less frequently. I0417 03:32:42.387979 136554047944512 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9 I0417 03:32:42.389902 136554047944512 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/linen_xpk_feat_nnx_post_train_fixes_20260417_023710_11_optimizer_offload_false/checkpoints/9. I0417 03:32:43.097691 136554047944512 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0417 03:32:43.097786 136554047944512 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 I0417 03:32:43.139374 136554047944512 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.042801s I0417 03:32:43.139542 136554047944512 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 33.383 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04620790481567383 s) (per-host) I0417 03:32:43.139598 136554047944512 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.046273s (batch_requests_ready=0.001771s, total_serialization_initiated=0.044428s, others=0.000073s) I0417 03:32:43.139686 136554047944512 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.050244s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=0.050228) I0417 03:32:43.140395 136554047944512 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.75 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/linen_xpk_feat_nnx_post_train_fixes_20260417_023710_11_optimizer_offload_false/checkpoints/9. I0417 03:32:43.140686 136425619449600 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-17 03:52:43.140654 I0417 03:32:43.142716 136554047944512 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0417 03:32:43.143046 136403681777408 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save. I0417 03:32:43.143212 136554047944512 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260417_023710/linen_xpk_feat_nnx_post_train_fixes_20260417_023710_11_optimizer_offload_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776396739.7828176, 'wait_for_prev_duration_secs': 22.603394031524658, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776396762.3880177, 'checkpointer_blocking_duration_secs': 0.7528266906738281, 'get_old_steps_start_time': 1776396763.1408675, 'get_old_steps_duration_secs': 3.123283386230469e-05, 'checkpoint_manager_blocking_start_time': 1776396739.7809672, 'checkpoint_manager_blocking_duration_secs': 23.36220932006836} I0417 03:32:43.143329 136554047944512 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0417 03:32:43.143375 136554047944512 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0417 03:32:50.322446 136425627842304 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_20260417_023710/linen_xpk_feat_nnx_post_train_fixes_20260417_023710_11_optimizer_offload_false/checkpoints/9/items/array_metadatas/process_6 I0417 03:33:26.760407 136425619449600 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 36.174 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.66703248023987 s) (per-host) I0417 03:33:26.760535 136425619449600 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.619735s. I0417 03:33:34.752276 136425619449600 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 51.611462s. I0417 03:33:34.752590 136403681777408 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save. I0417 03:33:34.752715 136403681777408 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save. I0417 03:33:34.752763 136403681777408 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0417 03:33:34.754355 136403681777408 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0417 03:33:34.754515 136554047944512 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. I0417 03:33:34.754674 136554047944512 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0417 03:33:34.755347 136554047944512 metric_logger.py:185] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.383, Tokens/s/device: 13924.963, total_weights: 65536, loss: 8.179 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Fri Apr 17 03:33:43 UTC 2026 EXIT_CODE=0