XPK Start: Wed Apr 22 12:58:48 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-22 12:59:12.660495: 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)
I0422 12:59:12.840179 135198618732352 max_utils.py:273] Attempting to initialize the jax distributed system...
I0422 12:59:21.882208 135198618732352 distributed.py:149] Starting JAX distributed service on [::]:8482
I0422 12:59:21.884442 135198618732352 distributed.py:172] Connecting to JAX distributed service on mt-07-eval-w7rkg-slice-job-0-0.mt-07-eval-w7rkg:8482
I0422 12:59:24.496226 135198618732352 max_utils.py:284] Jax distributed system initialized!
I0422 12:59:30.683948 135198618732352 max_utils.py:800] System Information: Jax Version: 0.9.2
I0422 12:59:30.684050 135198618732352 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0422 12:59:30.684089 135198618732352 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
I0422 12:59:30.684144 135198618732352 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0422 12:59:31.747356 135198618732352 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0422 12:59:31.747938 135198618732352 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0422 12:59:31.748130 135198618732352 checkpointing.py:688] Setting up checkpoint logger...
I0422 12:59:31.748180 135198618732352 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0422 12:59:31.748222 135198618732352 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0422 12:59:31.748561 135198618732352 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 0x7af5cdb14470>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0422 12:59:34.256309 135198618732352 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0422 12:59:34.256553 135198618732352 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 0x7af49da7b7d0>}, handler_registry=None
I0422 12:59:34.256789 135198618732352 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7af49da7b7d0>` 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`.
I0422 12:59:34.256837 135198618732352 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7af49da7a210>` 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`.
I0422 12:59:34.256873 135198618732352 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 0x7af49da7b7d0>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7af49da7b7d0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7af49da7a210>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7af49da7a210>}).
I0422 12:59:34.257201 135198618732352 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0422 12:59:34.257274 135198618732352 async_checkpointer.py:192] [process=6][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7ae1446639c0> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0422 12:59:35.344036 135198618732352 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_07_eval/checkpoints
I0422 12:59:35.787924 135198618732352 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_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_07_eval/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7af49db98b90>
I0422 12:59:35.788112 135198618732352 checkpointing.py:302] Checkpoint manager created!
I0422 12:59:36.732263 135198618732352 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0422 12:59:36.732367 135198618732352 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 12:59:37.111887 135198618732352 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 12:59:37.111983 135198618732352 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 12:59:37.128312 135198618732352 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0422 12:59:37.128371 135198618732352 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0422 12:59:37.151902 135198618732352 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0422 12:59:37.151969 135198618732352 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 12:59:37.168346 135198618732352 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0422 12:59:37.168408 135198618732352 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 12:59:37.184714 135198618732352 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0422 12:59:37.184771 135198618732352 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 12:59:37.209456 135198618732352 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0422 12:59:37.209527 135198618732352 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0422 12:59:37.231283 135198618732352 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0422 12:59:37.231346 135198618732352 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0422 12:59:37.445834 135198618732352 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0422 12:59:37.445938 135198618732352 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0422 12:59:38.868836 135198618732352 maxtext_utils.py:1827] params/params/decoder/decoder_norm/scale
Shape: float32[2048]
Logical: P('norm',)
Physical: (None,)
I0422 12:59:38.868959 135198618732352 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)
I0422 12:59:38.869011 135198618732352 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)
I0422 12:59:38.869143 135198618732352 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')
I0422 12:59:38.869212 135198618732352 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)
I0422 12:59:38.869254 135198618732352 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)
I0422 12:59:38.869312 135198618732352 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)
I0422 12:59:38.869367 135198618732352 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')
I0422 12:59:38.869405 135198618732352 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)
I0422 12:59:38.869441 135198618732352 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)
I0422 12:59:38.869487 135198618732352 maxtext_utils.py:1827] params/params/decoder/logits_dense/kernel
Shape: float32[2048,32000]
Logical: P('embed_vocab', 'vocab')
Physical: ('fsdp', None)
I0422 12:59:38.869535 135198618732352 maxtext_utils.py:1827] params/params/token_embedder/embedding
Shape: float32[32000,2048]
Logical: P('vocab', 'embed_vocab')
Physical: (None, 'fsdp')
I0422 12:59:39.361211 135198618732352 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0422 12:59:39.361303 135198618732352 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0422 12:59:39.376817 135198618732352 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0422 12:59:39.376880 135198618732352 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0422 12:59:50.387206 135198618732352 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.
I0422 12:59:50.388036 135198618732352 metric_logger.py:301] number parameters: 1.104 billion
I0422 13:00:01.659282 135198618732352 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0422 13:00:01.813980 135198618732352 checkpointing.py:798] Waited 0.15467429161071777 seconds for step 0 to finish before starting checkpointing.
I0422 13:00:01.816454 135198618732352 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 13:00:01.818190 135198618732352 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 0
I0422 13:00:01.819577 135198618732352 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_07_eval/checkpoints/0.
I0422 13:00:02.560819 135198618732352 signaling_client.py:364] Using JaxDistributedSignalingClient
I0422 13:00:02.561748 135198618732352 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0422 13:00:02.561806 135198618732352 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
I0422 13:00:02.844143 135198618732352 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.283398s
I0422 13:00:02.844313 135198618732352 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 5.340 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.288877010345459 s) (per-host)
I0422 13:00:02.844370 135198618732352 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.288945s (batch_requests_ready=0.002313s, total_serialization_initiated=0.286556s, others=0.000076s)
I0422 13:00:02.844468 135198618732352 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.293180s (all_items=0.000024s, per_item={'items': '0.00002384'}, temp_paths=0.293157)
I0422 13:00:02.845273 135198618732352 event_tracking.py:125] [process=6] [async] Finished blocking save in 1.03 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_07_eval/checkpoints/0.
I0422 13:00:02.845621 135069430961920 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 13:20:02.845583
I0422 13:00:02.847657 135198618732352 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 13:00:02.847901 135067860031232 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0422 13:00:02.848025 135198618732352 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776862801.816437, 'wait_for_prev_duration_secs': 5.9604644775390625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776862801.8182266, 'checkpointer_blocking_duration_secs': 1.027500867843628, 'get_old_steps_start_time': 1776862802.8457532, 'get_old_steps_duration_secs': 2.9802322387695312e-05, 'checkpoint_manager_blocking_start_time': 1776862801.8144975, 'checkpoint_manager_blocking_duration_secs': 1.0334892272949219}
I0422 13:00:02.848143 135198618732352 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0422 13:00:02.848193 135198618732352 max_utils.py:750]
Memstats: After params initialized:
I0422 13:00:02.848241 135198618732352 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_24(process=6,(0,6,0,0))
I0422 13:00:02.848272 135198618732352 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_25(process=6,(1,6,0,0))
I0422 13:00:02.848298 135198618732352 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_28(process=6,(0,7,0,0))
I0422 13:00:02.848322 135198618732352 max_utils.py:756] Using (GB) 0.43 / 31.25 (1.376000%) on TPU_29(process=6,(1,7,0,0))
I0422 13:00:03.159906 135198618732352 metric_logger.py:196] completed step: 0, seconds: 11.271, TFLOP/s/device: 1.205, Tokens/s/device: 181.703, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0422 13:00:03.449561 135198618732352 metric_logger.py:196] completed step: 1, seconds: 1.499, TFLOP/s/device: 9.063, Tokens/s/device: 1366.126, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52779.875
I0422 13:00:03.871697 135198618732352 metric_logger.py:196] completed step: 2, seconds: 0.142, TFLOP/s/device: 95.372, Tokens/s/device: 14375.562, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28637.129
I0422 13:00:04.019140 135198618732352 metric_logger.py:196] completed step: 3, seconds: 0.422, TFLOP/s/device: 32.165, Tokens/s/device: 4848.290, total_weights: 65536, loss: 9.734, lm_loss: 9.734, perplexity: 16889.203
I0422 13:00:04.313928 135198618732352 metric_logger.py:196] completed step: 4, seconds: 0.153, TFLOP/s/device: 88.956, Tokens/s/device: 13408.406, total_weights: 65536, loss: 9.277, lm_loss: 9.277, perplexity: 10694.614
I0422 13:00:04.319755 135198618732352 metric_logger.py:196] completed step: 5, seconds: 0.147, TFLOP/s/device: 92.152, Tokens/s/device: 13890.208, total_weights: 65536, loss: 8.892, lm_loss: 8.892, perplexity: 7270.717
I0422 13:00:06.041530 2866 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0422 13:00:08.088158 135068880971520 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_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_07_eval/checkpoints/0/items/array_metadatas/process_6
I0422 13:00:26.400522 135198618732352 metric_logger.py:196] completed step: 6, seconds: 0.295, TFLOP/s/device: 46.020, Tokens/s/device: 6936.682, total_weights: 65536, loss: 8.592, lm_loss: 8.592, perplexity: 5390.759
I0422 13:00:26.547872 135198618732352 metric_logger.py:196] completed step: 7, seconds: 21.934, TFLOP/s/device: 0.619, Tokens/s/device: 93.370, total_weights: 65536, loss: 8.384, lm_loss: 8.384, perplexity: 4376.905
I0422 13:00:26.695151 135198618732352 metric_logger.py:196] completed step: 8, seconds: 0.152, TFLOP/s/device: 89.384, Tokens/s/device: 13472.886, total_weights: 65536, loss: 8.255, lm_loss: 8.255, perplexity: 3846.406
I0422 13:00:26.841724 135198618732352 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0422 13:00:26.842381 135198618732352 checkpointing.py:798] Waited 0.0006794929504394531 seconds for step 9 to finish before starting checkpointing.
I0422 13:00:26.844611 135198618732352 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 13:00:40.397176 135069430961920 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 41.742 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 37.84167718887329 s) (per-host)
I0422 13:00:40.397303 135069430961920 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 37.551034s.
I0422 13:00:47.198160 135069430961920 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 44.351874s.
I0422 13:00:47.198439 135067860031232 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 13:00:47.198555 135067860031232 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0422 13:00:47.198618 135067860031232 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0422 13:00:47.201357 135067860031232 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0422 13:00:47.201559 135198618732352 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.
W0422 13:00:47.201699 135198618732352 checkpoint_manager.py:1452] Waiting for previous save to complete took 20.357088 seconds. If this number is high, consider checkpointing less frequently.
I0422 13:00:47.203592 135198618732352 checkpoint_manager.py:1512] [process=6] Saving checkpoint at step 9
I0422 13:00:47.205549 135198618732352 event_tracking.py:70] [process=6] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_07_eval/checkpoints/9.
I0422 13:00:47.495914 135198618732352 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0422 13:00:47.496006 135198618732352 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
I0422 13:00:47.539860 135198618732352 base_pytree_checkpoint_handler.py:154] [process=6][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.045009s
I0422 13:00:47.540006 135198618732352 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/blocking_gbytes_per_sec: 31.791 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04852294921875 s) (per-host)
I0422 13:00:47.540055 135198618732352 base_pytree_checkpoint_handler.py:768] [process=6][thread=MainThread] Initiated Pytree async_save. Time taken: 0.048580s (batch_requests_ready=0.001892s, total_serialization_initiated=0.046625s, others=0.000064s)
I0422 13:00:47.540153 135198618732352 composite_checkpoint_handler.py:715] [process=6][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.052609s (all_items=0.000015s, per_item={'items': '0.00001502'}, temp_paths=0.052594)
I0422 13:00:47.540821 135198618732352 event_tracking.py:125] [process=6] [async] Finished blocking save in 0.34 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_07_eval/checkpoints/9.
I0422 13:00:47.541180 135067860031232 async_checkpointer.py:76] [process=6][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-22 13:20:47.541144
I0422 13:00:47.543058 135198618732352 checkpoint_manager.py:1560] [process=6][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0422 13:00:47.543375 135045197043456 async_checkpointer.py:280] [process=6][thread=save_finalize] Waiting for background save thread=async_save.
I0422 13:00:47.543508 135198618732352 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_feat_nnx_post_train_fixes_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_07_eval/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776862826.8445807, 'wait_for_prev_duration_secs': 20.357088327407837, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776862847.2036304, 'checkpointer_blocking_duration_secs': 0.33765506744384766, 'get_old_steps_start_time': 1776862847.5413103, 'get_old_steps_duration_secs': 2.9087066650390625e-05, 'checkpoint_manager_blocking_start_time': 1776862826.8426142, 'checkpoint_manager_blocking_duration_secs': 20.700862169265747}
I0422 13:00:47.543667 135198618732352 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0422 13:00:47.543712 135198618732352 checkpoint_manager.py:2020] [process=6][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0422 13:00:52.667717 135068880971520 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_20260422_123906/linen_xpk_feat_nnx_post_train_fixes_20260422_123906_07_eval/checkpoints/9/items/array_metadatas/process_6
I0422 13:01:28.954051 135067860031232 base_pytree_checkpoint_handler.py:130] [process=6] /jax/orbax/write/gbytes_per_sec: 38.097 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 41.4625244140625 s) (per-host)
I0422 13:01:28.954201 135067860031232 async_checkpointer.py:90] [process=6][thread=async_save] 3 Handler Commit operations completed. Time taken: 41.412402s.
I0422 13:01:35.657381 135067860031232 async_checkpointer.py:160] [process=6][thread=async_save] Background save thread done. Time taken: 48.115566s.
I0422 13:01:35.657654 135045197043456 async_checkpointer.py:288] [process=6][thread=save_finalize] Done with waiting for background save thread=async_save.
I0422 13:01:35.657781 135045197043456 async_checkpointer.py:298] [process=6][thread=save_finalize] No errors found in background save thread=async_save.
I0422 13:01:35.657827 135045197043456 checkpoint_manager.py:2137] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0422 13:01:35.660577 135045197043456 checkpoint_manager.py:2146] [process=6][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0422 13:01:35.660692 135198618732352 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.
I0422 13:01:35.660765 135198618732352 checkpoint_manager.py:2009] [process=6][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0422 13:01:35.661708 135198618732352 metric_logger.py:196] completed step: 9, seconds: 0.147, TFLOP/s/device: 92.252, Tokens/s/device: 13905.203, 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: Wed Apr 22 13:01:48 UTC 2026
EXIT_CODE=0