MaxView

XPK Start: Sat Apr 18 23:57:24 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
2026-04-18 23:57:48.993006: 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)
I0418 23:57:49.172495 138550587356992 max_utils.py:273] Attempting to initialize the jax distributed system...
I0418 23:57:58.212410 138550587356992 distributed.py:149] Starting JAX distributed service on [::]:8482
I0418 23:57:58.214644 138550587356992 distributed.py:172] Connecting to JAX distributed service on mt-04-int8-kzxxs-slice-job-0-0.mt-04-int8-kzxxs:8482
I0418 23:58:00.369896 138550587356992 max_utils.py:284] Jax distributed system initialized!
I0418 23:58:06.474664 138550587356992 max_utils.py:800] System Information: Jax Version: 0.9.2
I0418 23:58:06.474766 138550587356992 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0418 23:58:06.474806 138550587356992 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
I0418 23:58:06.474843 138550587356992 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0418 23:58:07.176111 138550587356992 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0418 23:58:07.176707 138550587356992 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0418 23:58:07.177005 138550587356992 checkpointing.py:688] Setting up checkpoint logger...
I0418 23:58:07.177059 138550587356992 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True
I0418 23:58:07.177101 138550587356992 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0418 23:58:07.177443 138550587356992 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 0x7e024091c5c0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0418 23:58:11.101686 138550587356992 checkpointing.py:266] Enabling policy for fixed interval checkpointing.
I0418 23:58:11.101919 138550587356992 checkpoint_manager.py:708] [process=3][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7dee445b2390>}, handler_registry=None
I0418 23:58:11.102159 138550587356992 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7dee445b2390>` 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`.
I0418 23:58:11.102205 138550587356992 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7dee445b55b0>` 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`.
I0418 23:58:11.102252 138550587356992 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 0x7dee445b2390>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7dee445b2390>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7dee445b55b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7dee445b55b0>}).
I0418 23:58:11.102568 138550587356992 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0418 23:58:11.102636 138550587356992 async_checkpointer.py:192] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7dee2c693740> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0418 23:58:12.502130 138550587356992 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_04_int8/checkpoints
I0418 23:58:13.784795 138550587356992 checkpoint_manager.py:929] [process=3][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_04_int8/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7dee445b13d0>
I0418 23:58:13.784963 138550587356992 checkpointing.py:302] Checkpoint manager created!
I0418 23:58:14.945681 138550587356992 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0418 23:58:14.945786 138550587356992 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 23:58:15.327989 138550587356992 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 23:58:15.328081 138550587356992 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 23:58:15.344400 138550587356992 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0418 23:58:15.344459 138550587356992 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0418 23:58:15.409807 138550587356992 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 23:58:15.409896 138550587356992 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 23:58:15.426346 138550587356992 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 23:58:15.426405 138550587356992 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 23:58:15.442729 138550587356992 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0418 23:58:15.442791 138550587356992 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 23:58:15.467770 138550587356992 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0418 23:58:15.467842 138550587356992 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0418 23:58:15.533768 138550587356992 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0418 23:58:15.533854 138550587356992 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0418 23:58:16.046668 138550587356992 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint
I0418 23:58:16.046782 138550587356992 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0418 23:58:18.504601 138550587356992 maxtext_utils.py:1827]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0418 23:58:18.504728 138550587356992 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)
I0418 23:58:18.504781 138550587356992 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)
I0418 23:58:18.504838 138550587356992 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')
I0418 23:58:18.504899 138550587356992 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)
I0418 23:58:18.504939 138550587356992 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)
I0418 23:58:18.504996 138550587356992 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)
I0418 23:58:18.505050 138550587356992 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')
I0418 23:58:18.505090 138550587356992 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)
I0418 23:58:18.505127 138550587356992 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)
I0418 23:58:18.505175 138550587356992 maxtext_utils.py:1827]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)
I0418 23:58:18.505221 138550587356992 maxtext_utils.py:1827]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]
    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')

I0418 23:58:19.893128 138550587356992 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0418 23:58:19.893221 138550587356992 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0418 23:58:19.908958 138550587356992 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0418 23:58:19.909022 138550587356992 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0418 23:58:33.344213 138550587356992 max_utils.py:791] Total memory size: 1.8 GB, Output size: 0.4 GB, Temp size: 1.4 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB.
I0418 23:58:33.345000 138550587356992 metric_logger.py:301] number parameters: 1.104 billion
I0418 23:58:48.256581 138550587356992 checkpointing.py:794] Waiting for step 0 to finish before checkpoint...
I0418 23:58:48.499054 138550587356992 checkpointing.py:798] Waited 0.24245786666870117 seconds for step 0 to finish before starting checkpointing.
I0418 23:58:48.501357 138550587356992 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0418 23:58:48.503335 138550587356992 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 0
I0418 23:58:48.504637 138550587356992 event_tracking.py:70] [process=3] [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_04_int8/checkpoints/0.
I0418 23:58:49.254961 138550587356992 signaling_client.py:364] Using JaxDistributedSignalingClient
I0418 23:58:49.255952 138550587356992 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 23:58:49.256011 138550587356992 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
I0418 23:58:49.542834 138550587356992 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.287944s
I0418 23:58:49.543032 138550587356992 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 5.253 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.29363012313842773 s) (per-host)
I0418 23:58:49.543098 138550587356992 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.293703s (batch_requests_ready=0.002176s, total_serialization_initiated=0.291445s, others=0.000082s)
I0418 23:58:49.543215 138550587356992 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.298205s (all_items=0.000018s, per_item={'items': '0.00001812'}, temp_paths=0.298187)
I0418 23:58:49.544060 138550587356992 event_tracking.py:125] [process=3] [async] Finished blocking save in 1.04 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_04_int8/checkpoints/0.
I0418 23:58:49.544404 138417733195520 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 00:18:49.544365
I0418 23:58:49.546513 138550587356992 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 23:58:49.546809 138417170016000 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0418 23:58:49.546945 138550587356992 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_04_int8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776556728.5013385, 'wait_for_prev_duration_secs': 6.270408630371094e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776556728.5033743, 'checkpointer_blocking_duration_secs': 1.0411837100982666, 'get_old_steps_start_time': 1776556729.5445845, 'get_old_steps_duration_secs': 3.075599670410156e-05, 'checkpoint_manager_blocking_start_time': 1776556728.499584, 'checkpoint_manager_blocking_duration_secs': 1.0473191738128662}
I0418 23:58:49.547060 138550587356992 checkpointing.py:409] Started an asynchronous checkpoint save for step 0
I0418 23:58:49.547112 138550587356992 max_utils.py:750] 
Memstats: After params initialized:
I0418 23:58:49.547165 138550587356992 max_utils.py:756] 	Using (GB) 0.44 / 31.25 (1.408000%) on TPU_10(process=3,(2,2,0,0))
I0418 23:58:49.547198 138550587356992 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_11(process=3,(3,2,0,0))
I0418 23:58:49.547225 138550587356992 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_14(process=3,(2,3,0,0))
I0418 23:58:49.547269 138550587356992 max_utils.py:756] 	Using (GB) 0.43 / 31.25 (1.376000%) on TPU_15(process=3,(3,3,0,0))
I0418 23:58:49.876166 138550587356992 metric_logger.py:196] completed step: 0, seconds: 14.911, TFLOP/s/device: 0.911, Tokens/s/device: 137.344, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52796.277
I0418 23:58:50.128955 138550587356992 metric_logger.py:196] completed step: 1, seconds: 1.618, TFLOP/s/device: 8.397, Tokens/s/device: 1265.726, total_weights: 65536, loss: 10.874, lm_loss: 10.874, perplexity: 52796.277
I0418 23:58:50.534330 138550587356992 metric_logger.py:196] completed step: 2, seconds: 0.024, TFLOP/s/device: 575.628, Tokens/s/device: 86764.955, total_weights: 65536, loss: 10.262, lm_loss: 10.262, perplexity: 28634.820
I0418 23:58:50.763818 138550587356992 metric_logger.py:196] completed step: 3, seconds: 0.406, TFLOP/s/device: 33.480, Tokens/s/device: 5046.448, total_weights: 65536, loss: 9.731, lm_loss: 9.731, perplexity: 16827.199
I0418 23:58:51.224440 138550587356992 metric_logger.py:196] completed step: 4, seconds: 0.235, TFLOP/s/device: 57.841, Tokens/s/device: 8718.418, total_weights: 65536, loss: 9.272, lm_loss: 9.272, perplexity: 10638.531
I0418 23:58:51.230470 138550587356992 metric_logger.py:196] completed step: 5, seconds: 0.229, TFLOP/s/device: 59.223, Tokens/s/device: 8926.703, total_weights: 65536, loss: 8.887, lm_loss: 8.887, perplexity: 7233.941
I0418 23:58:54.986165    2937 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0418 23:58:58.284332 138417716410112 array_metadata_store.py:203] [process=3][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_04_int8/checkpoints/0/items/array_metadatas/process_3
I0418 23:59:22.188863 138550587356992 metric_logger.py:196] completed step: 6, seconds: 0.461, TFLOP/s/device: 29.467, Tokens/s/device: 4441.582, total_weights: 65536, loss: 8.588, lm_loss: 8.588, perplexity: 5365.468
I0418 23:59:22.418505 138550587356992 metric_logger.py:196] completed step: 7, seconds: 30.730, TFLOP/s/device: 0.442, Tokens/s/device: 66.645, total_weights: 65536, loss: 8.380, lm_loss: 8.380, perplexity: 4358.779
I0418 23:59:22.648374 138550587356992 metric_logger.py:196] completed step: 8, seconds: 0.234, TFLOP/s/device: 58.045, Tokens/s/device: 8749.220, total_weights: 65536, loss: 8.251, lm_loss: 8.251, perplexity: 3831.453
I0418 23:59:22.877547 138550587356992 checkpointing.py:794] Waiting for step 9 to finish before checkpoint...
I0418 23:59:22.878253 138550587356992 checkpointing.py:798] Waited 0.0007078647613525391 seconds for step 9 to finish before starting checkpointing.
I0418 23:59:22.880409 138550587356992 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 23:59:25.527564 138417733195520 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 43.541 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 36.27812600135803 s) (per-host)
I0418 23:59:25.527685 138417733195520 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 35.983168s.
I0418 23:59:36.246391 138417733195520 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 46.701852s.
I0418 23:59:36.246681 138417170016000 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0418 23:59:36.246806 138417170016000 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0418 23:59:36.246857 138417170016000 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0418 23:59:36.248442 138417170016000 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0418 23:59:36.248607 138550587356992 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0418 23:59:36.248740 138550587356992 checkpoint_manager.py:1452] Waiting for previous save to complete took 13.368331 seconds. If this number is high, consider checkpointing less frequently.
I0418 23:59:36.250490 138550587356992 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 9
I0418 23:59:36.252575 138550587356992 event_tracking.py:70] [process=3] [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_04_int8/checkpoints/9.
I0418 23:59:36.972715 138550587356992 jax_array_handlers.py:360] Scheduling D2H of 39 prioritized jax.Array.
I0418 23:59:36.972810 138550587356992 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
I0418 23:59:37.010722 138550587356992 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.039151s
I0418 23:59:37.010891 138550587356992 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 36.226 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.04258251190185547 s) (per-host)
I0418 23:59:37.010941 138550587356992 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.042642s (batch_requests_ready=0.001767s, total_serialization_initiated=0.040809s, others=0.000066s)
I0418 23:59:37.011033 138550587356992 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.047029s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=0.047014)
I0418 23:59:37.011695 138550587356992 event_tracking.py:125] [process=3] [async] Finished blocking save in 0.76 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_04_int8/checkpoints/9.
I0418 23:59:37.012033 138417170016000 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 00:19:37.011997
I0418 23:59:37.017825 138550587356992 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0418 23:59:37.018112 138417716410112 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save.
I0418 23:59:37.018288 138550587356992 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_04_int8/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776556762.880379, 'wait_for_prev_duration_secs': 13.36833143234253, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776556776.2505283, 'checkpointer_blocking_duration_secs': 0.7616519927978516, 'get_old_steps_start_time': 1776556777.0122018, 'get_old_steps_duration_secs': 4.553794860839844e-05, 'checkpoint_manager_blocking_start_time': 1776556762.8784773, 'checkpoint_manager_blocking_duration_secs': 14.13977575302124}
I0418 23:59:37.018403 138550587356992 checkpointing.py:409] Started an asynchronous checkpoint save for step 9
I0418 23:59:37.018448 138550587356992 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0418 23:59:45.713421 138394650674944 array_metadata_store.py:203] [process=3][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_04_int8/checkpoints/9/items/array_metadatas/process_3
I0419 00:00:21.367738 138417170016000 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 35.577 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 44.3993980884552 s) (per-host)
I0419 00:00:21.367848 138417170016000 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 44.355701s.
I0419 00:00:28.548726 138417170016000 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 51.536562s.
I0419 00:00:28.549015 138417716410112 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save.
I0419 00:00:28.549151 138417716410112 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save.
I0419 00:00:28.549201 138417716410112 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0419 00:00:28.550826 138417716410112 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0419 00:00:28.551008 138550587356992 checkpoint_manager.py:2032] [process=3][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0419 00:00:28.551154 138550587356992 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0419 00:00:28.551949 138550587356992 metric_logger.py:196] completed step: 9, seconds: 0.230, TFLOP/s/device: 59.195, Tokens/s/device: 8922.464, total_weights: 65536, loss: 8.175, lm_loss: 8.175, perplexity: 3552.333
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Sun Apr 19 00:00:39 UTC 2026
EXIT_CODE=0