XPK Start: Mon Apr 20 12:32:22 UTC 2026 PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. 2026-04-20 12:32:47.006309: 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) I0420 12:32:47.183878 140452174579520 max_utils.py:273] Attempting to initialize the jax distributed system... I0420 12:32:56.224682 140452174579520 distributed.py:149] Starting JAX distributed service on [::]:8482 I0420 12:32:56.226977 140452174579520 distributed.py:172] Connecting to JAX distributed service on mt-07-eval-efj4x-slice-job-0-0.mt-07-eval-efj4x:8482 I0420 12:32:57.615227 140452174579520 max_utils.py:284] Jax distributed system initialized! I0420 12:33:04.162870 140452174579520 max_utils.py:800] System Information: Jax Version: 0.9.2 I0420 12:33:04.162972 140452174579520 max_utils.py:801] System Information: Jaxlib Version: 0.9.2 I0420 12:33:04.163010 140452174579520 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 I0420 12:33:04.163050 140452174579520 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0420 12:33:04.858377 140452174579520 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0420 12:33:04.859015 140452174579520 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0420 12:33:04.859217 140452174579520 checkpointing.py:688] Setting up checkpoint logger... I0420 12:33:04.859279 140452174579520 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0420 12:33:04.859339 140452174579520 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None I0420 12:33:04.859700 140452174579520 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 0x7fbcda6380e0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0420 12:33:08.254151 140452174579520 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0420 12:33:08.254339 140452174579520 checkpoint_manager.py:708] [process=2][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fa8783ae1e0>}, handler_registry=None I0420 12:33:08.254567 140452174579520 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fa8783ae1e0>` 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`. I0420 12:33:08.254614 140452174579520 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fa8783af500>` 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`. I0420 12:33:08.254649 140452174579520 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 0x7fa8783ae1e0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7fa8783ae1e0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fa8783af500>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7fa8783af500>}). I0420 12:33:08.254977 140452174579520 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34 I0420 12:33:08.255046 140452174579520 async_checkpointer.py:192] [process=2][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7fa87812b880> timeout: 1200 secs and primary_host=0 for async checkpoint writes I0420 12:33:09.380800 140452174579520 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158_07_eval/checkpoints I0420 12:33:09.405031 140452174579520 checkpoint_manager.py:929] [process=2][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_linen_converter_and_sharding_tools_20260420_112158/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7fa87850c4d0> I0420 12:33:09.405159 140452174579520 checkpointing.py:302] Checkpoint manager created! I0420 12:33:10.344145 140452174579520 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0420 12:33:10.344247 140452174579520 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 12:33:10.724076 140452174579520 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0420 12:33:10.724174 140452174579520 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 12:33:10.740518 140452174579520 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0420 12:33:10.740577 140452174579520 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0420 12:33:10.764339 140452174579520 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 12:33:10.764405 140452174579520 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 12:33:10.780795 140452174579520 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 12:33:10.780853 140452174579520 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 12:33:10.797327 140452174579520 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0420 12:33:10.797388 140452174579520 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 12:33:10.821988 140452174579520 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0420 12:33:10.822056 140452174579520 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0420 12:33:10.842810 140452174579520 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0420 12:33:10.842871 140452174579520 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0420 12:33:11.062083 140452174579520 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0420 12:33:11.062192 140452174579520 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0420 12:33:12.500848 140452174579520 maxtext_utils.py:1827] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0420 12:33:12.500972 140452174579520 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) I0420 12:33:12.501027 140452174579520 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) I0420 12:33:12.501103 140452174579520 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') I0420 12:33:12.501157 140452174579520 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) I0420 12:33:12.501198 140452174579520 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) I0420 12:33:12.501252 140452174579520 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) I0420 12:33:12.501303 140452174579520 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') I0420 12:33:12.501343 140452174579520 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) I0420 12:33:12.501380 140452174579520 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) I0420 12:33:12.501428 140452174579520 maxtext_utils.py:1827] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: P('embed_vocab', 'vocab') Physical: ('fsdp', None) I0420 12:33:12.501478 140452174579520 maxtext_utils.py:1827] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: P('vocab', 'embed_vocab') Physical: (None, 'fsdp') I0420 12:33:12.999836 140452174579520 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0420 12:33:12.999938 140452174579520 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0420 12:33:13.015608 140452174579520 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0420 12:33:13.015668 140452174579520 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0420 12:33:24.114912 140452174579520 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. I0420 12:33:24.115778 140452174579520 metric_logger.py:301] number parameters: 1.104 billion I0420 12:33:35.371386 140452174579520 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0420 12:33:35.525123 140452174579520 checkpointing.py:798] Waited 0.15370535850524902 seconds for step 0 to finish before starting checkpointing. I0420 12:33:35.527474 140452174579520 checkpoint_manager.py:2009] [process=2][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0420 12:33:35.529168 140452174579520 checkpoint_manager.py:1512] [process=2] Saving checkpoint at step 0 I0420 12:33:35.530865 140452174579520 event_tracking.py:70] [process=2] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158_07_eval/checkpoints/0. I0420 12:33:35.928788 140452174579520 signaling_client.py:364] Using JaxDistributedSignalingClient I0420 12:33:35.929638 140452174579520 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0420 12:33:35.929696 140452174579520 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 I0420 12:33:36.205618 140452174579520 base_pytree_checkpoint_handler.py:154] [process=2][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.276920s I0420 12:33:36.205813 140452174579520 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/blocking_gbytes_per_sec: 5.460 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.28254270553588867 s) (per-host) I0420 12:33:36.205871 140452174579520 base_pytree_checkpoint_handler.py:768] [process=2][thread=MainThread] Initiated Pytree async_save. Time taken: 0.282613s (batch_requests_ready=0.002371s, total_serialization_initiated=0.280163s, others=0.000079s) I0420 12:33:36.205965 140452174579520 composite_checkpoint_handler.py:715] [process=2][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.286715s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.286697) I0420 12:33:36.206836 140452174579520 event_tracking.py:125] [process=2] [async] Finished blocking save in 0.68 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158_07_eval/checkpoints/0. I0420 12:33:36.207190 140323006449408 async_checkpointer.py:76] [process=2][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 12:53:36.207153 I0420 12:33:36.211896 140452174579520 checkpoint_manager.py:1560] [process=2][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0420 12:33:36.212207 140320339654400 async_checkpointer.py:280] [process=2][thread=save_finalize] Waiting for background save thread=async_save. I0420 12:33:36.212361 140452174579520 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776688415.5274553, 'wait_for_prev_duration_secs': 6.222724914550781e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776688415.5292056, 'checkpointer_blocking_duration_secs': 0.6781480312347412, 'get_old_steps_start_time': 1776688416.207379, 'get_old_steps_duration_secs': 3.62396240234375e-05, 'checkpoint_manager_blocking_start_time': 1776688415.5256765, 'checkpoint_manager_blocking_duration_secs': 0.6866483688354492} I0420 12:33:36.212483 140452174579520 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0420 12:33:36.212530 140452174579520 max_utils.py:750] Memstats: After params initialized: I0420 12:33:36.212578 140452174579520 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_8(process=2,(0,2,0,0)) I0420 12:33:36.212614 140452174579520 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_9(process=2,(1,2,0,0)) I0420 12:33:36.212640 140452174579520 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_12(process=2,(0,3,0,0)) I0420 12:33:36.212663 140452174579520 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_13(process=2,(1,3,0,0)) I0420 12:33:36.525482 140452174579520 metric_logger.py:196] completed step: 0, seconds: 11.256, TFLOP/s/device: 1.207, Tokens/s/device: 181.955, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0420 12:33:36.703853 140452174579520 metric_logger.py:196] completed step: 1, seconds: 1.153, TFLOP/s/device: 11.789, Tokens/s/device: 1776.942, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875 I0420 12:33:37.115934 140452174579520 metric_logger.py:196] completed step: 2, seconds: 0.031, TFLOP/s/device: 441.227, Tokens/s/device: 66506.462, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129 I0420 12:33:37.263298 140452174579520 metric_logger.py:196] completed step: 3, seconds: 0.413, TFLOP/s/device: 32.929, Tokens/s/device: 4963.465, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203 I0420 12:33:37.558298 140452174579520 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.667, Tokens/s/device: 13364.831, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614 I0420 12:33:37.564440 140452174579520 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.311, Tokens/s/device: 13914.178, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717 I0420 12:33:39.646289 2824 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0420 12:33:41.680221 140321402173184 array_metadata_store.py:203] [process=2][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158_07_eval/checkpoints/0/items/array_metadatas/process_2 I0420 12:33:59.197421 140452174579520 metric_logger.py:196] completed step: 6, seconds: 0.296, TFLOP/s/device: 45.969, Tokens/s/device: 6928.914, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759 I0420 12:33:59.344707 140452174579520 metric_logger.py:196] completed step: 7, seconds: 21.486, TFLOP/s/device: 0.632, Tokens/s/device: 95.317, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905 I0420 12:33:59.492096 140452174579520 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.130, Tokens/s/device: 13434.706, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406 I0420 12:33:59.638702 140452174579520 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0420 12:33:59.639372 140452174579520 checkpointing.py:798] Waited 0.0006899833679199219 seconds for step 9 to finish before starting checkpointing. I0420 12:33:59.641424 140452174579520 checkpoint_manager.py:2020] [process=2][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0420 12:34:14.919250 140323006449408 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/gbytes_per_sec: 40.507 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 38.99594211578369 s) (per-host) I0420 12:34:14.919379 140323006449408 async_checkpointer.py:90] [process=2][thread=async_save] 3 Handler Commit operations completed. Time taken: 38.712073s. I0420 12:34:22.371732 140323006449408 async_checkpointer.py:160] [process=2][thread=async_save] Background save thread done. Time taken: 46.164390s. I0420 12:34:22.371987 140320339654400 async_checkpointer.py:288] [process=2][thread=save_finalize] Done with waiting for background save thread=async_save. I0420 12:34:22.372062 140320339654400 async_checkpointer.py:298] [process=2][thread=save_finalize] No errors found in background save thread=async_save. I0420 12:34:22.372110 140320339654400 checkpoint_manager.py:2137] [process=2][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0420 12:34:22.373814 140320339654400 checkpoint_manager.py:2146] [process=2][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0420 12:34:22.374010 140452174579520 checkpoint_manager.py:2032] [process=2][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0. W0420 12:34:22.374165 140452174579520 checkpoint_manager.py:1452] Waiting for previous save to complete took 22.732736 seconds. If this number is high, consider checkpointing less frequently. I0420 12:34:22.375850 140452174579520 checkpoint_manager.py:1512] [process=2] Saving checkpoint at step 9 I0420 12:34:22.377875 140452174579520 event_tracking.py:70] [process=2] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158_07_eval/checkpoints/9. I0420 12:34:22.668082 140452174579520 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array. I0420 12:34:22.668176 140452174579520 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 I0420 12:34:22.710128 140452174579520 base_pytree_checkpoint_handler.py:154] [process=2][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.043292s I0420 12:34:22.710297 140452174579520 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/blocking_gbytes_per_sec: 32.912 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04686903953552246 s) (per-host) I0420 12:34:22.710348 140452174579520 base_pytree_checkpoint_handler.py:768] [process=2][thread=MainThread] Initiated Pytree async_save. Time taken: 0.046930s (batch_requests_ready=0.001925s, total_serialization_initiated=0.044936s, others=0.000068s) I0420 12:34:22.710442 140452174579520 composite_checkpoint_handler.py:715] [process=2][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.050946s (all_items=0.000022s, per_item={'items': '0.00002170'}, temp_paths=0.050925) I0420 12:34:22.711129 140452174579520 event_tracking.py:125] [process=2] [async] Finished blocking save in 0.34 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158_07_eval/checkpoints/9. I0420 12:34:22.711479 140320339654400 async_checkpointer.py:76] [process=2][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-20 12:54:22.711443 I0420 12:34:22.713445 140452174579520 checkpoint_manager.py:1560] [process=2][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0420 12:34:22.713682 140316391819008 async_checkpointer.py:280] [process=2][thread=save_finalize] Waiting for background save thread=async_save. I0420 12:34:22.713856 140452174579520 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776688439.6413882, 'wait_for_prev_duration_secs': 22.732736110687256, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776688462.37589, 'checkpointer_blocking_duration_secs': 0.3357408046722412, 'get_old_steps_start_time': 1776688462.7116563, 'get_old_steps_duration_secs': 2.9802322387695312e-05, 'checkpoint_manager_blocking_start_time': 1776688439.6396122, 'checkpoint_manager_blocking_duration_secs': 23.074209928512573} I0420 12:34:22.713977 140452174579520 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0420 12:34:22.714021 140452174579520 checkpoint_manager.py:2020] [process=2][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0420 12:34:28.384886 140321402173184 array_metadata_store.py:203] [process=2][thread=array_type_handler] Wrote 39 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158/linen_xpk_feat_nnx_linen_converter_and_sharding_tools_20260420_112158_07_eval/checkpoints/9/items/array_metadatas/process_2 I0420 12:35:04.707186 140320339654400 base_pytree_checkpoint_handler.py:130] [process=2] /jax/orbax/write/gbytes_per_sec: 37.570 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 42.043707609176636 s) (per-host) I0420 12:35:04.707311 140320339654400 async_checkpointer.py:90] [process=2][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.995716s. I0420 12:35:13.877005 140320339654400 async_checkpointer.py:160] [process=2][thread=async_save] Background save thread done. Time taken: 51.165395s. I0420 12:35:13.877327 140316391819008 async_checkpointer.py:288] [process=2][thread=save_finalize] Done with waiting for background save thread=async_save. I0420 12:35:13.877443 140316391819008 async_checkpointer.py:298] [process=2][thread=save_finalize] No errors found in background save thread=async_save. I0420 12:35:13.877490 140316391819008 checkpoint_manager.py:2137] [process=2][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0420 12:35:13.881699 140316391819008 checkpoint_manager.py:2146] [process=2][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0420 12:35:13.881916 140452174579520 checkpoint_manager.py:2032] [process=2][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9. I0420 12:35:13.882068 140452174579520 checkpoint_manager.py:2009] [process=2][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0420 12:35:13.883026 140452174579520 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.253, Tokens/s/device: 13905.392, total_weights: 65536, loss: 8.179, lm_loss: 8.179, perplexity: 3564.635 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Mon Apr 20 12:35:22 UTC 2026 EXIT_CODE=0