XPK Start: Sun Apr 19 00:28:37 UTC 2026 PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. 2026-04-19 00:29:01.332071: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303) I0419 00:29:01.508568 133761152444224 max_utils.py:273] Attempting to initialize the jax distributed system... I0419 00:29:10.549320 133761152444224 distributed.py:149] Starting JAX distributed service on [::]:8482 I0419 00:29:10.551628 133761152444224 distributed.py:172] Connecting to JAX distributed service on mt-13-scan-layers-false-mqmer-slice-job-0-0.mt-13-scan-layers-false-mqmer:8482 I0419 00:29:12.534826 133761152444224 max_utils.py:284] Jax distributed system initialized! I0419 00:29:18.580118 133761152444224 max_utils.py:800] System Information: Jax Version: 0.9.2 I0419 00:29:18.580227 133761152444224 max_utils.py:801] System Information: Jaxlib Version: 0.9.2 I0419 00:29:18.580293 133761152444224 max_utils.py:802] System Information: Jax Backend: PJRT C API TFRT TPU v6 lite Built on Mar 4 2026 11:32:08 (1772652728) cl/878335365 I0419 00:29:18.580331 133761152444224 train_utils.py:378] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing. I0419 00:29:19.641421 133761152444224 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 00:29:19.642005 133761152444224 maxtext_utils.py:1718] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1) I0419 00:29:19.642181 133761152444224 checkpointing.py:688] Setting up checkpoint logger... I0419 00:29:19.642242 133761152444224 checkpointing.py:234] Creating checkpoint manager with ocdbt=True and zarr3=True I0419 00:29:19.642287 133761152444224 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None I0419 00:29:19.642611 133761152444224 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 0x79a6fa3481d0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB) I0419 00:29:23.401954 133761152444224 checkpointing.py:266] Enabling policy for fixed interval checkpointing. I0419 00:29:23.402260 133761152444224 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 0x79a5ee7eb140>}, handler_registry=None I0419 00:29:23.402501 133761152444224 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79a5ee7eb140>` for item "items" and save args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>` to `_handler_registry`. I0419 00:29:23.402550 133761152444224 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79a5ee7ebd40>` for item "metrics" and save args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>` and restore args `<class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>` to `_handler_registry`. I0419 00:29:23.402585 133761152444224 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 0x79a5ee7eb140>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x79a5ee7eb140>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79a5ee7ebd40>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x79a5ee7ebd40>}). I0419 00:29:23.402900 133761152444224 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34 I0419 00:29:23.402970 133761152444224 async_checkpointer.py:192] [process=3][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7992d8656c00> timeout: 1200 secs and primary_host=0 for async checkpoint writes I0419 00:29:24.521595 133761152444224 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_13_scan_layers_false/checkpoints I0419 00:29:24.933810 133761152444224 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_13_scan_layers_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x79a5ee7e9c10> I0419 00:29:24.933992 133761152444224 checkpointing.py:302] Checkpoint manager created! I0419 00:29:25.879271 133761152444224 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed'). I0419 00:29:25.879379 133761152444224 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 00:29:26.262356 133761152444224 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 00:29:26.262451 133761152444224 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 00:29:26.278535 133761152444224 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed'). I0419 00:29:26.278594 133761152444224 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None). I0419 00:29:26.307978 133761152444224 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 00:29:26.308047 133761152444224 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 00:29:26.324496 133761152444224 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 00:29:26.324558 133761152444224 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 00:29:26.340496 133761152444224 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim'). I0419 00:29:26.340555 133761152444224 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 00:29:26.369577 133761152444224 attentions.py:1197] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv'). I0419 00:29:26.369647 133761152444224 attentions.py:1197] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None). I0419 00:29:26.391579 133761152444224 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp'). I0419 00:29:26.391643 133761152444224 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None). I0419 00:29:29.262651 133761152444224 checkpointing.py:578] checkpoint manager exists so trying to load this run's existing checkpoint I0419 00:29:29.262777 133761152444224 checkpointing.py:676] No existing checkpoints found, not restoring checkpoint. fsdp: 32 I0419 00:29:36.510007 133761152444224 maxtext_utils.py:1827] params/params/decoder/decoder_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.510136 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_0/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.510189 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_0/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.510262 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_0/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.510306 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_0/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.510344 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_0/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.510396 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_0/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.510448 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_0/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.510488 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_0/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.510522 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_0/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.510556 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_1/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.510589 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_1/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.510621 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_1/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.510652 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_1/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.510682 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_1/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.510715 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_1/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.510747 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_1/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.510780 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_1/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.510810 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_1/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.510841 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_10/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.510870 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_10/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.510900 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_10/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.510934 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_10/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.510963 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_10/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.510993 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_10/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.511023 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_10/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.511052 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_10/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.511083 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_10/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.511113 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_11/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.511142 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_11/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.511172 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_11/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.511200 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_11/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.511228 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_11/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.511271 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_11/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.511302 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_11/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.511332 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_11/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.511362 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_11/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.511392 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_12/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.511421 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_12/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.511451 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_12/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.511479 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_12/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.511506 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_12/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.511535 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_12/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.511565 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_12/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.511595 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_12/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.511623 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_12/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.511652 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_13/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.511681 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_13/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.511709 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_13/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.511736 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_13/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.511764 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_13/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.511793 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_13/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.511822 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_13/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.511851 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_13/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.511880 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_13/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.511909 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_14/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.511944 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_14/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.511974 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_14/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.512003 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_14/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.512030 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_14/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.512059 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_14/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.512088 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_14/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.512117 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_14/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.512146 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_14/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.512175 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_15/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.512204 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_15/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.512243 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_15/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.512274 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_15/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.512302 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_15/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.512331 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_15/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.512360 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_15/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.512390 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_15/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.512419 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_15/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.512448 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_2/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.512478 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_2/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.512507 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_2/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.512534 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_2/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.512562 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_2/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.512593 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_2/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.512623 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_2/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.512653 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_2/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.512682 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_2/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.512711 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_3/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.512740 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_3/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.512769 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_3/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.512797 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_3/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.512824 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_3/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.512853 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_3/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.512882 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_3/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.512927 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_3/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.512979 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_3/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.513027 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_4/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.513059 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_4/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.513088 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_4/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.513116 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_4/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.513143 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_4/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.513176 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_4/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.513208 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_4/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.513260 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_4/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.513295 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_4/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.513326 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_5/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.513355 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_5/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.513384 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_5/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.513411 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_5/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.513439 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_5/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.513468 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_5/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.513497 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_5/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.513526 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_5/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.513555 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_5/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.513584 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_6/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.513613 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_6/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.513642 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_6/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.513669 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_6/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.513696 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_6/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.513726 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_6/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.513755 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_6/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.513784 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_6/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.513813 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_6/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.513843 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_7/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.513872 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_7/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.513901 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_7/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.513933 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_7/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.513962 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_7/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.513992 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_7/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.514022 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_7/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.514051 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_7/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.514080 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_7/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.514109 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_8/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.514137 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_8/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.514166 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_8/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.514193 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_8/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.514220 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_8/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.514263 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_8/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.514294 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_8/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.514326 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_8/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.514356 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_8/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.514386 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_9/mlp/wi_0/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.514415 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_9/mlp/wi_1/kernel Shape: float32[2048,7168] Logical: P('embed', 'mlp') Physical: ('fsdp', None) I0419 00:29:36.514444 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_9/mlp/wo/kernel Shape: float32[7168,2048] Logical: P('mlp', 'embed') Physical: (None, 'fsdp') I0419 00:29:36.514472 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_9/post_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.514499 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_9/pre_self_attention_layer_norm/scale Shape: float32[2048] Logical: P('norm',) Physical: (None,) I0419 00:29:36.514528 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_9/self_attention/key/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.514559 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_9/self_attention/out/kernel Shape: float32[16,128,2048] Logical: P('heads', 'kv', 'embed') Physical: (None, None, 'fsdp') I0419 00:29:36.514588 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_9/self_attention/query/kernel Shape: float32[2048,16,128] Logical: P('embed', 'q_heads', 'kv') Physical: ('fsdp', None, None) I0419 00:29:36.514617 133761152444224 maxtext_utils.py:1827] params/params/decoder/layers_9/self_attention/value/kernel Shape: float32[2048,16,128] Logical: P('embed', 'kv_heads', 'kv_head_dim') Physical: ('fsdp', None, None) I0419 00:29:36.514666 133761152444224 maxtext_utils.py:1827] params/params/decoder/logits_dense/kernel Shape: float32[2048,32000] Logical: P('embed_vocab', 'vocab') Physical: ('fsdp', None) I0419 00:29:36.514709 133761152444224 maxtext_utils.py:1827] params/params/token_embedder/embedding Shape: float32[32000,2048] Logical: P('vocab', 'embed_vocab') Physical: (None, 'fsdp') I0419 00:29:41.091063 133761152444224 train.py:157] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 00:29:41.091166 133761152444224 train.py:157] train/xent Physical: float32[32,2048]............................................ ('fsdp', None). I0419 00:29:41.106627 133761152444224 train.py:164] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length'). I0419 00:29:41.106687 133761152444224 train.py:164] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None). I0419 00:30:37.852154 133761152444224 max_utils.py:791] Total memory size: 1.8 GB, Output size: 0.4 GB, Temp size: 1.5 GB, Argument size: 0.4 GB, Host temp size: 0.0 GB. I0419 00:30:37.853326 133761152444224 metric_logger.py:301] number parameters: 1.104 billion I0419 00:31:38.596080 133761152444224 checkpointing.py:794] Waiting for step 0 to finish before checkpoint... I0419 00:31:38.830101 133761152444224 checkpointing.py:798] Waited 0.23400139808654785 seconds for step 0 to finish before starting checkpointing. I0419 00:31:38.833878 133761152444224 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 00:31:38.835961 133761152444224 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 0 I0419 00:31:38.837394 133761152444224 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_13_scan_layers_false/checkpoints/0. I0419 00:31:39.640294 133761152444224 signaling_client.py:364] Using JaxDistributedSignalingClient I0419 00:31:39.641312 133761152444224 jax_array_handlers.py:360] Scheduling D2H of 444 prioritized jax.Array. I0419 00:31:39.641438 133761152444224 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0419 00:31:39.940024 133761152444224 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.300264s I0419 00:31:39.940200 133761152444224 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 4.601 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.33524036407470703 s) (per-host) I0419 00:31:39.940262 133761152444224 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.335311s (batch_requests_ready=0.018381s, total_serialization_initiated=0.316851s, others=0.000078s) I0419 00:31:39.940507 133761152444224 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.339669s (all_items=0.000019s, per_item={'items': '0.00001860'}, temp_paths=0.339651) I0419 00:31:39.941381 133761152444224 event_tracking.py:125] [process=3] [async] Finished blocking save in 1.11 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_13_scan_layers_false/checkpoints/0. I0419 00:31:39.943090 133636541564672 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 00:51:39.943048 I0419 00:31:39.992040 133761152444224 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize I0419 00:31:39.992394 133633328731904 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save. I0419 00:31:39.992561 133761152444224 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_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776558698.8338513, 'wait_for_prev_duration_secs': 7.128715515136719e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776558698.8359993, 'checkpointer_blocking_duration_secs': 1.107196569442749, 'get_old_steps_start_time': 1776558699.9432225, 'get_old_steps_duration_secs': 4.100799560546875e-05, 'checkpoint_manager_blocking_start_time': 1776558698.831677, 'checkpoint_manager_blocking_duration_secs': 1.1608424186706543} I0419 00:31:39.992770 133761152444224 checkpointing.py:409] Started an asynchronous checkpoint save for step 0 I0419 00:31:39.992830 133761152444224 max_utils.py:750] Memstats: After params initialized: I0419 00:31:39.992883 133761152444224 max_utils.py:756] Using (GB) 0.82 / 31.25 (2.624000%) on TPU_10(process=3,(2,2,0,0)) I0419 00:31:39.992916 133761152444224 max_utils.py:756] Using (GB) 0.82 / 31.25 (2.624000%) on TPU_11(process=3,(3,2,0,0)) I0419 00:31:39.992942 133761152444224 max_utils.py:756] Using (GB) 0.82 / 31.25 (2.624000%) on TPU_14(process=3,(2,3,0,0)) I0419 00:31:39.992967 133761152444224 max_utils.py:756] Using (GB) 0.82 / 31.25 (2.624000%) on TPU_15(process=3,(3,3,0,0)) I0419 00:31:40.345950 133761152444224 metric_logger.py:196] completed step: 0, seconds: 60.743, TFLOP/s/device: 0.224, Tokens/s/device: 33.716, total_weights: 65536, loss: 10.877, lm_loss: 10.877, perplexity: 52938.617 I0419 00:31:40.526575 133761152444224 metric_logger.py:196] completed step: 1, seconds: 1.740, TFLOP/s/device: 7.807, Tokens/s/device: 1176.729, total_weights: 65536, loss: 10.877, lm_loss: 10.877, perplexity: 52938.617 I0419 00:31:40.964830 133761152444224 metric_logger.py:196] completed step: 2, seconds: 0.054, TFLOP/s/device: 251.679, Tokens/s/device: 37935.761, total_weights: 65536, loss: 10.268, lm_loss: 10.268, perplexity: 28795.404 I0419 00:31:41.101251 133761152444224 metric_logger.py:196] completed step: 3, seconds: 0.439, TFLOP/s/device: 30.983, Tokens/s/device: 4670.169, total_weights: 65536, loss: 9.741, lm_loss: 9.741, perplexity: 17000.498 I0419 00:31:41.379742 133761152444224 metric_logger.py:196] completed step: 4, seconds: 0.143, TFLOP/s/device: 95.016, Tokens/s/device: 14321.879, total_weights: 65536, loss: 9.285, lm_loss: 9.285, perplexity: 10779.109 I0419 00:31:41.392031 133761152444224 metric_logger.py:196] completed step: 5, seconds: 0.136, TFLOP/s/device: 99.826, Tokens/s/device: 15046.875, total_weights: 65536, loss: 8.900, lm_loss: 8.900, perplexity: 7335.493 I0419 00:31:43.506481 2859 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com I0419 00:31:46.055957 133636549957376 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 444 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_13_scan_layers_false/checkpoints/0/items/array_metadatas/process_3 I0419 00:32:08.251446 133761152444224 metric_logger.py:196] completed step: 6, seconds: 0.280, TFLOP/s/device: 48.543, Tokens/s/device: 7316.925, total_weights: 65536, loss: 8.602, lm_loss: 8.602, perplexity: 5441.066 I0419 00:32:08.390080 133761152444224 metric_logger.py:196] completed step: 7, seconds: 26.728, TFLOP/s/device: 0.508, Tokens/s/device: 76.623, total_weights: 65536, loss: 8.394, lm_loss: 8.394, perplexity: 4418.581 I0419 00:32:08.526650 133761152444224 metric_logger.py:196] completed step: 8, seconds: 0.142, TFLOP/s/device: 95.482, Tokens/s/device: 14392.028, total_weights: 65536, loss: 8.264, lm_loss: 8.264, perplexity: 3882.439 I0419 00:32:08.663548 133761152444224 checkpointing.py:794] Waiting for step 9 to finish before checkpoint... I0419 00:32:08.667018 133761152444224 checkpointing.py:798] Waited 0.00348663330078125 seconds for step 9 to finish before starting checkpointing. I0419 00:32:08.670266 133761152444224 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 00:32:13.253341 133636541564672 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 46.944 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 33.64833617210388 s) (per-host) I0419 00:32:13.253465 133636541564672 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 33.309592s. I0419 00:32:22.675459 133636541564672 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 42.731571s. I0419 00:32:22.675746 133633328731904 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 00:32:22.675867 133633328731904 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save. I0419 00:32:22.675917 133633328731904 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts... I0419 00:32:22.679860 133633328731904 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts. I0419 00:32:22.680028 133761152444224 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. W0419 00:32:22.680182 133761152444224 checkpoint_manager.py:1452] Waiting for previous save to complete took 14.009925 seconds. If this number is high, consider checkpointing less frequently. I0419 00:32:22.682749 133761152444224 checkpoint_manager.py:1512] [process=3] Saving checkpoint at step 9 I0419 00:32:22.684759 133761152444224 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_13_scan_layers_false/checkpoints/9. I0419 00:32:23.428703 133761152444224 jax_array_handlers.py:360] Scheduling D2H of 444 prioritized jax.Array. I0419 00:32:23.428874 133761152444224 replica_slices.py:424] Transferring arrays to host memory with options: use_replica_parallel=True, min_slice_bytes_for_replica_parallel=None, max_replicas_for_replica_parallel=None, enable_pinned_host_transfer=False I0419 00:32:23.548752 133761152444224 base_pytree_checkpoint_handler.py:154] [process=3][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.121464s I0419 00:32:23.548913 133761152444224 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/blocking_gbytes_per_sec: 9.972 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.15469622611999512 s) (per-host) I0419 00:32:23.548962 133761152444224 base_pytree_checkpoint_handler.py:768] [process=3][thread=MainThread] Initiated Pytree async_save. Time taken: 0.154754s (batch_requests_ready=0.017576s, total_serialization_initiated=0.137113s, others=0.000065s) I0419 00:32:23.549231 133761152444224 composite_checkpoint_handler.py:715] [process=3][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.159112s (all_items=0.000016s, per_item={'items': '0.00001645'}, temp_paths=0.159095) I0419 00:32:23.550002 133761152444224 event_tracking.py:125] [process=3] [async] Finished blocking save in 0.87 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_13_scan_layers_false/checkpoints/9. I0419 00:32:23.550402 133633328731904 async_checkpointer.py:76] [process=3][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-19 00:52:23.550363 I0419 00:32:23.562127 133761152444224 checkpoint_manager.py:1560] [process=3][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize I0419 00:32:23.562375 133635451217664 async_checkpointer.py:280] [process=3][thread=save_finalize] Waiting for background save thread=async_save. I0419 00:32:23.562503 133761152444224 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_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776558728.6702178, 'wait_for_prev_duration_secs': 14.009925127029419, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776558742.6827862, 'checkpointer_blocking_duration_secs': 0.867755651473999, 'get_old_steps_start_time': 1776558743.5505667, 'get_old_steps_duration_secs': 3.0279159545898438e-05, 'checkpoint_manager_blocking_start_time': 1776558728.667884, 'checkpoint_manager_blocking_duration_secs': 14.894585371017456} I0419 00:32:23.562707 133761152444224 checkpointing.py:409] Started an asynchronous checkpoint save for step 9 I0419 00:32:23.562752 133761152444224 checkpoint_manager.py:2020] [process=3][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete. I0419 00:32:30.675582 133636549957376 array_metadata_store.py:203] [process=3][thread=array_type_handler] Wrote 444 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_13_scan_layers_false/checkpoints/9/items/array_metadatas/process_3 I0419 00:33:07.509313 133633328731904 base_pytree_checkpoint_handler.py:130] [process=3] /jax/orbax/write/gbytes_per_sec: 35.806 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 44.11504888534546 s) (per-host) I0419 00:33:07.509441 133633328731904 async_checkpointer.py:90] [process=3][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.958928s. I0419 00:33:16.636157 133633328731904 async_checkpointer.py:160] [process=3][thread=async_save] Background save thread done. Time taken: 53.085619s. I0419 00:33:16.636503 133635451217664 async_checkpointer.py:288] [process=3][thread=save_finalize] Done with waiting for background save thread=async_save. I0419 00:33:16.636635 133635451217664 async_checkpointer.py:298] [process=3][thread=save_finalize] No errors found in background save thread=async_save. I0419 00:33:16.636684 133635451217664 checkpoint_manager.py:2137] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts... I0419 00:33:16.639323 133635451217664 checkpoint_manager.py:2146] [process=3][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts. I0419 00:33:16.639504 133761152444224 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 00:33:16.639649 133761152444224 checkpoint_manager.py:2009] [process=3][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning. I0419 00:33:16.640587 133761152444224 metric_logger.py:196] completed step: 9, seconds: 0.139, TFLOP/s/device: 98.052, Tokens/s/device: 14779.534, total_weights: 65536, loss: 8.188, lm_loss: 8.188, perplexity: 3598.943 Per train step: Total TFLOPs: 13.59 split as 93.93% learnable weight flops and 6.07% attention flops XPK End: Sun Apr 19 00:33:28 UTC 2026 EXIT_CODE=0