Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 4459 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.5709. Min Act: -0.0000

Data Index: 107742 (C4 (Web Text))

Max Activating Token Index: 539

Text #1

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.5895. Min Act: -0.0000

Data Index: 512719 (C4 (Web Text))

Max Activating Token Index: 967

Text #2

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.5877. Min Act: -0.0000

Data Index: 991649 (C4 (Web Text))

Max Activating Token Index: 530

Text #3

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.5462. Min Act: -0.0000

Data Index: 1120213 (C4 (Web Text))

Max Activating Token Index: 494

Text #4

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.5212. Min Act: -0.0000

Data Index: 723268 (C4 (Web Text))

Max Activating Token Index: 66

Text #5

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.5097. Min Act: -0.0000

Data Index: 796765 (C4 (Web Text))

Max Activating Token Index: 20

Text #6

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4891. Min Act: -0.0000

Data Index: 56862 (C4 (Web Text))

Max Activating Token Index: 882

Text #7

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.5376. Min Act: -0.0000

Data Index: 419143 (C4 (Web Text))

Max Activating Token Index: 196

Text #8

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4703. Min Act: -0.0000

Data Index: 1201544 (C4 (Web Text))

Max Activating Token Index: 957

Text #9

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4850. Min Act: -0.0000

Data Index: 774752 (C4 (Web Text))

Max Activating Token Index: 222

Text #10

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4824. Min Act: -0.0000

Data Index: 1031332 (C4 (Web Text))

Max Activating Token Index: 859

Text #11

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4839. Min Act: -0.0000

Data Index: 1276310 (C4 (Web Text))

Max Activating Token Index: 248

Text #12

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4888. Min Act: -0.0000

Data Index: 252670 (C4 (Web Text))

Max Activating Token Index: 938

Text #13

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4888. Min Act: -0.0000

Data Index: 203600 (C4 (Web Text))

Max Activating Token Index: 671

Text #14

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4860. Min Act: -0.0000

Data Index: 415905 (C4 (Web Text))

Max Activating Token Index: 923

Text #15

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4950. Min Act: -0.0000

Data Index: 115388 (C4 (Web Text))

Max Activating Token Index: 710

Text #16

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.5090. Min Act: -0.0000

Data Index: 874373 (C4 (Web Text))

Max Activating Token Index: 735

Text #17

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4885. Min Act: -0.0000

Data Index: 117073 (C4 (Web Text))

Max Activating Token Index: 83

Text #18

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4343. Min Act: -0.0000

Data Index: 82082 (C4 (Web Text))

Max Activating Token Index: 262

Text #19

Max Range: 0.5895. Min Range: -0.5895

Max Act: 0.4347. Min Act: -0.0000

Data Index: 954133 (C4 (Web Text))

Max Activating Token Index: 635