MaxView

XPK Start: Tue Apr 21 15:24:30 UTC 2026
PyTorch was not found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.
`rope_parameters`'s factor field must be a float >= 1, got 40
`rope_parameters`'s beta_fast field must be a float, got 32
`rope_parameters`'s beta_slow field must be a float, got 1
DeepseekV32Config got `key=rope_scaling` in kwargs but hasn't set it as attribute. For RoPE standardization you need to set `self.rope_parameters` in model's config. 
2026-04-21 15:24:55.622367: 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)
I0421 15:24:55.831514 137312694798144 max_utils.py:273] Attempting to initialize the jax distributed system...
I0421 15:25:04.871715 137312694798144 distributed.py:149] Starting JAX distributed service on [::]:8482
I0421 15:25:04.874113 137312694798144 distributed.py:172] Connecting to JAX distributed service on mt-13-scan-layers-false-pt6j9-slice-job-0-0.mt-13-scan-layers-false-pt6j9:8482
I0421 15:25:06.070394 137312694798144 max_utils.py:284] Jax distributed system initialized!
I0421 15:25:12.278580 137312694798144 max_utils.py:800] System Information: Jax Version: 0.9.2
I0421 15:25:12.278705 137312694798144 max_utils.py:801] System Information: Jaxlib Version: 0.9.2
I0421 15:25:12.278750 137312694798144 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
I0421 15:25:12.278786 137312694798144 train_utils.py:361] WARNING: Sequence packing is essentially ignored for synthetic data. Please use a real dataset to use sequence packing.
I0421 15:25:12.970910 137312694798144 maxtext_utils.py:1565] Num_devices: 32, shape (1, 1, 1, 32, 1, 1, 1, 1, 1, 1, 1, 1, 1)
I0421 15:25:12.971188 137312694798144 checkpointing.py:677] Setting up checkpoint logger...
I0421 15:25:12.971242 137312694798144 checkpointing.py:233] Creating checkpoint manager with ocdbt=True and zarr3=True
I0421 15:25:12.971287 137312694798144 pytree_checkpoint_handler.py:592] save_device_host_concurrent_bytes=None
I0421 15:25:12.971654 137312694798144 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 0x7ce1e2c4b5c0>, enable_pinned_host_transfer=False, save_concurrent_bytes: 96000000000 (89.4 GiB), restore_concurrent_bytes: 96000000000 (89.4 GiB)
I0421 15:25:15.879271 137312694798144 checkpointing.py:265] Enabling policy for fixed interval checkpointing.
I0421 15:25:15.879501 137312694798144 checkpoint_manager.py:708] [process=4][thread=MainThread] CheckpointManager init: checkpointers=None, item_names=('items',), item_handlers={'items': <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ccd900bf350>}, handler_registry=None
I0421 15:25:15.879749 137312694798144 composite_checkpoint_handler.py:237] Deferred registration for item: "items". Adding handler `<orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ccd900bf350>` 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`.
I0421 15:25:15.879799 137312694798144 composite_checkpoint_handler.py:237] Deferred registration for item: "metrics". Adding handler `<orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ccd900c95b0>` 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`.
I0421 15:25:15.879835 137312694798144 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 0x7ccd900bf350>, ('items', <class 'orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeRestoreArgs'>): <orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler object at 0x7ccd900bf350>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonSaveArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ccd900c95b0>, ('metrics', <class 'orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonRestoreArgs'>): <orbax.checkpoint._src.handlers.json_checkpoint_handler.JsonCheckpointHandler object at 0x7ccd900c95b0>}).
I0421 15:25:15.880152 137312694798144 abstract_checkpointer.py:35] orbax-checkpoint version: 0.11.34
I0421 15:25:15.880223 137312694798144 async_checkpointer.py:192] [process=4][thread=MainThread] Using barrier_sync_fn: <function get_barrier_sync_fn.<locals>._fn at 0x7ccca85e1580> timeout: 1200 secs and primary_host=0 for async checkpoint writes
I0421 15:25:17.085316 137312694798144 checkpoint_manager.py:1812] Found 0 checkpoint steps in gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_13_scan_layers_false/checkpoints
I0421 15:25:17.087618 137312694798144 checkpoint_manager.py:929] [process=4][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_main_20260421_144122/linen_xpk_main_20260421_144122_13_scan_layers_false/checkpoints: <orbax.checkpoint.checkpoint_manager.CheckpointManager object at 0x7ccd900c03b0>
I0421 15:25:17.087740 137312694798144 checkpointing.py:301] Checkpoint manager created!
I0421 15:25:18.036080 137312694798144 nnx_wrappers.py:437] Unknown Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_norm_length', 'activation_embed').
I0421 15:25:18.036184 137312694798144 nnx_wrappers.py:437] Unknown Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 15:25:18.420008 137312694798144 attentions.py:1088] attentions/inputs_q Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 15:25:18.420099 137312694798144 attentions.py:1088] attentions/inputs_q Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 15:25:18.436298 137312694798144 attentions.py:1089] attentions/inputs_kv Logical: bfloat16[32,2048,2048]...................................... ('activation_batch', 'activation_attn_length', 'activation_attn_embed').
I0421 15:25:18.436355 137312694798144 attentions.py:1089] attentions/inputs_kv Physical: bfloat16[32,2048,2048]...................................... ('fsdp', None, None).
I0421 15:25:18.466481 137312694798144 attentions.py:1154] attentions/query Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 15:25:18.466557 137312694798144 attentions.py:1154] attentions/query Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 15:25:18.483054 137312694798144 attentions.py:1155] attentions/key Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 15:25:18.483116 137312694798144 attentions.py:1155] attentions/key Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 15:25:18.499375 137312694798144 attentions.py:1156] attentions/value Logical: bfloat16[32,2048,16,128].................................... ('activation_kv_batch', 'activation_attn_length', 'activation_kv_heads', 'activation_kv_head_dim').
I0421 15:25:18.499439 137312694798144 attentions.py:1156] attentions/value Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 15:25:18.528632 137312694798144 attentions.py:1198] attentions/out Logical: bfloat16[32,2048,16,128].................................... ('activation_batch', 'activation_attn_length', 'activation_heads', 'activation_kv').
I0421 15:25:18.528717 137312694798144 attentions.py:1198] attentions/out Physical: bfloat16[32,2048,16,128].................................... ('fsdp', None, None, None).
I0421 15:25:18.550309 137312694798144 linears.py:525] linears/x Logical: bfloat16[32,2048,7168]...................................... ('activation_batch', 'activation_length', 'activation_mlp').
I0421 15:25:18.550379 137312694798144 linears.py:525] linears/x Physical: bfloat16[32,2048,7168]...................................... ('fsdp', None, None).
I0421 15:25:21.423074 137312694798144 checkpointing.py:577] checkpoint manager exists so trying to load this run's existing checkpoint
I0421 15:25:21.423204 137312694798144 checkpointing.py:665] No existing checkpoints found, not restoring checkpoint.
fsdp: 32
I0421 15:25:28.671127 137312694798144 maxtext_utils.py:1668]  params/params/decoder/decoder_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.671260 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_0/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.671319 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_0/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.671379 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_0/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.671420 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_0/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.671455 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_0/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.671506 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.671557 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_0/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.671594 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_0/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.671628 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.671675 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_1/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.671709 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_1/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.671743 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_1/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.671774 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_1/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.671804 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_1/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.671836 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.671869 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_1/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.671900 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_1/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.671931 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.671961 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_10/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.671991 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_10/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.672021 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_10/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.672049 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_10/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.672077 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_10/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.672107 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.672137 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_10/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.672166 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_10/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.672195 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.672224 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_11/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.672253 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_11/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.672290 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_11/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.672319 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_11/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.672347 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_11/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.672377 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.672406 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_11/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.672436 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_11/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.672465 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.672494 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_12/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.672530 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_12/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.672581 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_12/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.672630 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_12/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.672676 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_12/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.672709 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.672741 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_12/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.672785 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_12/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.672817 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.672847 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_13/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.672876 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_13/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.672905 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_13/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.672933 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_13/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.672961 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_13/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.672991 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.673020 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_13/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.673049 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_13/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.673078 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.673107 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_14/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.673136 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_14/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.673165 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_14/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.673192 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_14/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.673220 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_14/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.673249 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.673283 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_14/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.673313 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_14/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.673341 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.673370 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_15/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.673398 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_15/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.673428 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_15/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.673455 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_15/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.673482 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_15/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.673512 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.673544 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_15/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.673575 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_15/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.673605 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.673645 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_2/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.673681 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_2/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.673713 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_2/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.673741 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_2/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.673769 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_2/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.673799 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.673831 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_2/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.673862 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_2/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.673892 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.673922 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_3/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.673952 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_3/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.673981 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_3/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.674008 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_3/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.674035 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_3/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.674064 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.674093 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_3/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.674124 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_3/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.674153 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.674182 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_4/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.674211 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_4/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.674240 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_4/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.674270 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_4/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.674298 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_4/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.674327 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.674356 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_4/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.674386 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_4/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.674415 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.674443 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_5/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.674472 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_5/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.674501 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_5/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.674527 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_5/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.674555 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_5/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.674583 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.674612 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_5/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.674651 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_5/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.674682 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.674712 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_6/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.674741 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_6/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.674769 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_6/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.674796 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_6/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.674823 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_6/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.674851 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.674880 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_6/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.674909 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_6/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.674937 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.674966 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_7/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.674994 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_7/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.675023 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_7/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.675050 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_7/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.675076 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_7/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.675105 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.675134 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_7/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.675163 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_7/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.675192 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.675220 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_8/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.675249 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_8/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.675282 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_8/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.675310 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_8/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.675337 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_8/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.675365 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.675395 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_8/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.675425 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_8/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.675453 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.675481 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_9/mlp/wi_0/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.675509 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_9/mlp/wi_1/kernel
    Shape:     float32[2048,7168]
    Logical:   P('embed', 'mlp')
    Physical:  ('fsdp', None)
I0421 15:25:28.675535 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_9/mlp/wo/kernel
    Shape:     float32[7168,2048]
    Logical:   P('mlp', 'embed')
    Physical:  (None, 'fsdp')
I0421 15:25:28.675562 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_9/post_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.675588 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_9/pre_self_attention_layer_norm/scale
    Shape:     float32[2048]
    Logical:   P('norm',)
    Physical:  (None,)
I0421 15:25:28.675617 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.675655 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_9/self_attention/out/kernel
    Shape:     float32[16,128,2048]
    Logical:   P('heads', 'kv', 'embed')
    Physical:  (None, None, 'fsdp')
I0421 15:25:28.675686 137312694798144 maxtext_utils.py:1668]  params/params/decoder/layers_9/self_attention/query/kernel
    Shape:     float32[2048,16,128]
    Logical:   P('embed', 'q_heads', 'kv')
    Physical:  ('fsdp', None, None)
I0421 15:25:28.675714 137312694798144 maxtext_utils.py:1668]  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)
I0421 15:25:28.675763 137312694798144 maxtext_utils.py:1668]  params/params/decoder/logits_dense/kernel
    Shape:     float32[2048,32000]
    Logical:   P('embed_vocab', 'vocab')
    Physical:  ('fsdp', None)
I0421 15:25:28.675808 137312694798144 maxtext_utils.py:1668]  params/params/token_embedder/embedding
    Shape:     float32[32000,2048]

    Logical:   P('vocab', 'embed_vocab')
    Physical:  (None, 'fsdp')
I0421 15:25:33.141902 137312694798144 train.py:155] train/xent Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 15:25:33.141993 137312694798144 train.py:155] train/xent Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 15:25:33.157498 137312694798144 train.py:162] train/z_loss Logical: float32[32,2048]............................................ ('activation_embed_and_logits_batch', 'activation_length').
I0421 15:25:33.157560 137312694798144 train.py:162] train/z_loss Physical: float32[32,2048]............................................ ('fsdp', None).
I0421 15:26:29.770126 137312694798144 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.
I0421 15:26:29.771283 137312694798144 metric_logger.py:301] number parameters: 1.104 billion
I0421 15:27:30.340874 137312694798144 checkpointing.py:772] Waiting for step 0 to finish before checkpoint...
I0421 15:27:30.575389 137312694798144 checkpointing.py:776] Waited 0.23449254035949707 seconds for step 0 to finish before starting checkpointing.
I0421 15:27:30.579014 137312694798144 checkpoint_manager.py:2009] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 15:27:30.580965 137312694798144 checkpoint_manager.py:1512] [process=4] Saving checkpoint at step 0
I0421 15:27:30.582568 137312694798144 event_tracking.py:70] [process=4] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_13_scan_layers_false/checkpoints/0.
I0421 15:27:30.898369 137312694798144 signaling_client.py:364] Using JaxDistributedSignalingClient
I0421 15:27:30.899411 137312694798144 jax_array_handlers.py:360] Scheduling D2H of 444 prioritized jax.Array.
I0421 15:27:30.899534 137312694798144 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
I0421 15:27:31.201298 137312694798144 base_pytree_checkpoint_handler.py:154] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.303432s
I0421 15:27:31.201462 137312694798144 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/blocking_gbytes_per_sec: 4.583 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.3365659713745117 s) (per-host)
I0421 15:27:31.201512 137312694798144 base_pytree_checkpoint_handler.py:768] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.336626s (batch_requests_ready=0.016863s, total_serialization_initiated=0.319696s, others=0.000067s)
I0421 15:27:31.201776 137312694798144 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.340892s (all_items=0.000017s, per_item={'items': '0.00001740'}, temp_paths=0.340875)
I0421 15:27:31.202611 137312694798144 event_tracking.py:125] [process=4] [async] Finished blocking save in 0.62 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_13_scan_layers_false/checkpoints/0.
I0421 15:27:31.202986 137184233367296 async_checkpointer.py:76] [process=4][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 15:47:31.202947
I0421 15:27:31.247580 137312694798144 checkpoint_manager.py:1560] [process=4][thread=MainThread][step=0] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 15:27:31.247952 137183730067200 async_checkpointer.py:280] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0421 15:27:31.248117 137312694798144 standard_logger.py:34] {'step': 0, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776785250.5789955, 'wait_for_prev_duration_secs': 6.103515625e-05, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776785250.5810044, 'checkpointer_blocking_duration_secs': 0.6221401691436768, 'get_old_steps_start_time': 1776785251.2031693, 'get_old_steps_duration_secs': 3.147125244140625e-05, 'checkpoint_manager_blocking_start_time': 1776785250.5770142, 'checkpoint_manager_blocking_duration_secs': 0.6710641384124756}
I0421 15:27:31.248304 137312694798144 checkpointing.py:408] Started an asynchronous checkpoint save for step 0
I0421 15:27:31.248356 137312694798144 max_utils.py:750] 
Memstats: After params initialized:
I0421 15:27:31.248410 137312694798144 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_16(process=4,(0,4,0,0))
I0421 15:27:31.248451 137312694798144 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_17(process=4,(1,4,0,0))
I0421 15:27:31.248475 137312694798144 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_20(process=4,(0,5,0,0))
I0421 15:27:31.248500 137312694798144 max_utils.py:756] 	Using (GB) 0.82 / 31.25 (2.624000%) on TPU_21(process=4,(1,5,0,0))
I0421 15:27:31.604334 137312694798144 metric_logger.py:196] completed step: 0, seconds: 60.569, TFLOP/s/device: 0.224, Tokens/s/device: 33.812, total_weights: 65536, loss: 10.877, lm_loss: 10.877, perplexity: 52938.617
I0421 15:27:31.842262 137312694798144 metric_logger.py:196] completed step: 1, seconds: 1.255, TFLOP/s/device: 10.828, Tokens/s/device: 1632.071, total_weights: 65536, loss: 10.877, lm_loss: 10.877, perplexity: 52938.617
I0421 15:27:32.264699 137312694798144 metric_logger.py:196] completed step: 2, seconds: 0.111, TFLOP/s/device: 122.816, Tokens/s/device: 18512.158, total_weights: 65536, loss: 10.268, lm_loss: 10.268, perplexity: 28794.738
I0421 15:27:32.400853 137312694798144 metric_logger.py:196] completed step: 3, seconds: 0.424, TFLOP/s/device: 32.076, Tokens/s/device: 4834.830, total_weights: 65536, loss: 9.741, lm_loss: 9.741, perplexity: 17000.666
I0421 15:27:32.681717 137312694798144 metric_logger.py:196] completed step: 4, seconds: 0.142, TFLOP/s/device: 95.900, Tokens/s/device: 14455.110, total_weights: 65536, loss: 9.285, lm_loss: 9.285, perplexity: 10779.159
I0421 15:27:32.693402 137312694798144 metric_logger.py:196] completed step: 5, seconds: 0.136, TFLOP/s/device: 99.993, Tokens/s/device: 15072.012, total_weights: 65536, loss: 8.901, lm_loss: 8.901, perplexity: 7335.745
I0421 15:27:34.288566    2891 google_auth_provider.cc:181] Running on GCE, using service account 562977990677-compute@developer.gserviceaccount.com
I0421 15:27:36.321767 137183738459904 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 444 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_13_scan_layers_false/checkpoints/0/items/array_metadatas/process_4
I0421 15:27:57.429850 137312694798144 metric_logger.py:196] completed step: 6, seconds: 0.282, TFLOP/s/device: 48.142, Tokens/s/device: 7256.442, total_weights: 65536, loss: 8.602, lm_loss: 8.602, perplexity: 5440.818
I0421 15:27:57.565966 137312694798144 metric_logger.py:196] completed step: 7, seconds: 24.616, TFLOP/s/device: 0.552, Tokens/s/device: 83.197, total_weights: 65536, loss: 8.393, lm_loss: 8.393, perplexity: 4418.067
I0421 15:27:57.702247 137312694798144 metric_logger.py:196] completed step: 8, seconds: 0.131, TFLOP/s/device: 103.728, Tokens/s/device: 15635.020, total_weights: 65536, loss: 8.264, lm_loss: 8.264, perplexity: 3882.134
I0421 15:27:57.837828 137312694798144 checkpointing.py:772] Waiting for step 9 to finish before checkpoint...
I0421 15:27:57.841170 137312694798144 checkpointing.py:776] Waited 0.0033617019653320312 seconds for step 9 to finish before starting checkpointing.
I0421 15:27:57.844074 137312694798144 checkpoint_manager.py:2020] [process=4][thread=MainThread][step=0][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 15:28:02.684596 137184233367296 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/gbytes_per_sec: 49.642 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 31.819656133651733 s) (per-host)
I0421 15:28:02.684752 137184233367296 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 31.481653s.
I0421 15:28:13.515404 137184233367296 async_checkpointer.py:160] [process=4][thread=async_save] Background save thread done. Time taken: 42.312289s.
I0421 15:28:13.515740 137183730067200 async_checkpointer.py:288] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 15:28:13.515892 137183730067200 async_checkpointer.py:298] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0421 15:28:13.515957 137183730067200 checkpoint_manager.py:2137] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is syncing with other hosts...
I0421 15:28:13.517633 137183730067200 checkpoint_manager.py:2146] [process=4][thread=save_finalize][step=0] CheckpointManager Save Finalize is done on all hosts.
I0421 15:28:13.517839 137312694798144 checkpoint_manager.py:2032] [process=4][thread=MainThread][step=0][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=0.
W0421 15:28:13.517987 137312694798144 checkpoint_manager.py:1452] Waiting for previous save to complete took 15.673912 seconds. If this number is high, consider checkpointing less frequently.
I0421 15:28:13.520310 137312694798144 checkpoint_manager.py:1512] [process=4] Saving checkpoint at step 9
I0421 15:28:13.522474 137312694798144 event_tracking.py:70] [process=4] [async] Started save checkpoint @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_13_scan_layers_false/checkpoints/9.
I0421 15:28:14.267524 137312694798144 jax_array_handlers.py:360] Scheduling D2H of 444 prioritized jax.Array.
I0421 15:28:14.267712 137312694798144 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
I0421 15:28:14.394895 137312694798144 base_pytree_checkpoint_handler.py:154] [process=4][thread=MainThread] Initiated "orbax.checkpoint._src.serialization.jax_array_handlers.ArrayHandler".serialize. Time taken: 0.128890s
I0421 15:28:14.395055 137312694798144 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/blocking_gbytes_per_sec: 9.596 GiB/s (total gbytes: 1.5 GiB) (time elapsed: 0.16074538230895996 s) (per-host)
I0421 15:28:14.395105 137312694798144 base_pytree_checkpoint_handler.py:768] [process=4][thread=MainThread] Initiated Pytree async_save. Time taken: 0.160803s (batch_requests_ready=0.016420s, total_serialization_initiated=0.144317s, others=0.000066s)
I0421 15:28:14.395381 137312694798144 composite_checkpoint_handler.py:715] [process=4][thread=MainThread] Initiated CompositeCheckpointHandler.async_save. Time taken: 0.165205s (all_items=0.000015s, per_item={'items': '0.00001550'}, temp_paths=0.165190)
I0421 15:28:14.396070 137312694798144 event_tracking.py:125] [process=4] [async] Finished blocking save in 0.88 seconds. Continuing save @ gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_13_scan_layers_false/checkpoints/9.
I0421 15:28:14.396462 137183696496384 async_checkpointer.py:76] [process=4][thread=async_save] Background save thread started. Deadline for this save operation is 2026-04-21 15:48:14.396421
I0421 15:28:14.399596 137312694798144 checkpoint_manager.py:1560] [process=4][thread=MainThread][step=9] Starting CheckpointManager Save Finalize thread=save_finalize
I0421 15:28:14.399899 137183730067200 async_checkpointer.py:280] [process=4][thread=save_finalize] Waiting for background save thread=async_save.
I0421 15:28:14.400063 137312694798144 standard_logger.py:34] {'step': 9, 'event_type': 'save', 'directory': 'gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_13_scan_layers_false/checkpoints', 'reached_preemption': False, 'preemption_received_at': None, 'synchronous': False, 'wait_for_prev_start_time': 1776785277.8440444, 'wait_for_prev_duration_secs': 15.673912286758423, 'time_between_consecutive_saves_sec': None, 'checkpointer_blocking_start_time': 1776785293.5203488, 'checkpointer_blocking_duration_secs': 0.8762662410736084, 'get_old_steps_start_time': 1776785294.3966508, 'get_old_steps_duration_secs': 3.0994415283203125e-05, 'checkpoint_manager_blocking_start_time': 1776785277.8420343, 'checkpoint_manager_blocking_duration_secs': 16.5579936504364}
I0421 15:28:14.400273 137312694798144 checkpointing.py:408] Started an asynchronous checkpoint save for step 9
I0421 15:28:14.400320 137312694798144 checkpoint_manager.py:2020] [process=4][thread=MainThread][step=9][wait_until_finished] Waiting for Save Finalize thread (save_finalize) to complete.
I0421 15:28:20.666081 137183738459904 array_metadata_store.py:203] [process=4][thread=array_type_handler] Wrote 444 array_metadata.ArrayMetadata to gs://lance-maxtext/linen_ckpt_xpk_main_20260421_144122/linen_xpk_main_20260421_144122_13_scan_layers_false/checkpoints/9/items/array_metadatas/process_4
I0421 15:28:57.894221 137183696496384 base_pytree_checkpoint_handler.py:130] [process=4] /jax/orbax/write/gbytes_per_sec: 36.180 MiB/s (total gbytes: 1.5 GiB) (time elapsed: 43.659871101379395 s) (per-host)
I0421 15:28:57.894346 137183696496384 async_checkpointer.py:90] [process=4][thread=async_save] 3 Handler Commit operations completed. Time taken: 43.497765s.
I0421 15:29:05.995011 137183696496384 async_checkpointer.py:160] [process=4][thread=async_save] Background save thread done. Time taken: 51.598414s.
I0421 15:29:05.995276 137183730067200 async_checkpointer.py:288] [process=4][thread=save_finalize] Done with waiting for background save thread=async_save.
I0421 15:29:05.995395 137183730067200 async_checkpointer.py:298] [process=4][thread=save_finalize] No errors found in background save thread=async_save.
I0421 15:29:05.995447 137183730067200 checkpoint_manager.py:2137] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is syncing with other hosts...
I0421 15:29:05.997152 137183730067200 checkpoint_manager.py:2146] [process=4][thread=save_finalize][step=9] CheckpointManager Save Finalize is done on all hosts.
I0421 15:29:05.997334 137312694798144 checkpoint_manager.py:2032] [process=4][thread=MainThread][step=9][wait_until_finished] Done waiting for Save Finalize thread (save_finalize) running at step=9.
I0421 15:29:05.997502 137312694798144 checkpoint_manager.py:2009] [process=4][thread=MainThread][wait_until_finished] No Save Finalize thread to wait for. Returning.
I0421 15:29:05.998449 137312694798144 metric_logger.py:196] completed step: 9, seconds: 0.136, TFLOP/s/device: 99.556, Tokens/s/device: 15006.192, total_weights: 65536, loss: 8.188, lm_loss: 8.188, perplexity: 3598.959
Per train step:
 Total TFLOPs: 13.59 
 split as 93.93% learnable weight flops and 6.07% attention flops
XPK End: Tue Apr 21 15:29:15 UTC 2026
EXIT_CODE=0