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

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

Neuron 3330 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: 0.8659. Min Range: -0.8659

Max Act: 0.8560. Min Act: -0.0000

Data Index: 86108 (C4 (Web Text))

Max Activating Token Index: 854

Text #1

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.8584. Min Act: -0.0000

Data Index: 1205272 (C4 (Web Text))

Max Activating Token Index: 766

Text #2

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.8036. Min Act: -0.0000

Data Index: 1155700 (C4 (Web Text))

Max Activating Token Index: 396

Text #3

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.8031. Min Act: -0.0000

Data Index: 1285056 (C4 (Web Text))

Max Activating Token Index: 885

Text #4

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.7363. Min Act: -0.0000

Data Index: 492194 (C4 (Web Text))

Max Activating Token Index: 264

Text #5

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.7961. Min Act: -0.0000

Data Index: 602564 (C4 (Web Text))

Max Activating Token Index: 833

Text #6

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.7219. Min Act: -0.0000

Data Index: 939197 (C4 (Web Text))

Max Activating Token Index: 292

Text #7

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.8659. Min Act: -0.0000

Data Index: 1311113 (C4 (Web Text))

Max Activating Token Index: 202

Text #8

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.8068. Min Act: -0.0000

Data Index: 64811 (C4 (Web Text))

Max Activating Token Index: 203

Text #9

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.7770. Min Act: -0.0000

Data Index: 332792 (C4 (Web Text))

Max Activating Token Index: 153

Text #10

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.7445. Min Act: -0.0000

Data Index: 158410 (C4 (Web Text))

Max Activating Token Index: 645

Text #11

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.7425. Min Act: 0.0001

Data Index: 1559874 (Python Code)

Max Activating Token Index: 37

Text #12

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.7033. Min Act: -0.0000

Data Index: 219701 (C4 (Web Text))

Max Activating Token Index: 640

Text #13

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.6603. Min Act: -0.0000

Data Index: 552928 (C4 (Web Text))

Max Activating Token Index: 97

Text #14

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.7219. Min Act: -0.0000

Data Index: 1245605 (C4 (Web Text))

Max Activating Token Index: 843

Text #15

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.7184. Min Act: -0.0000

Data Index: 1036775 (C4 (Web Text))

Max Activating Token Index: 542

Text #16

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.7113. Min Act: 0.0001

Data Index: 1530091 (Python Code)

Max Activating Token Index: 32

Text #17

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.6874. Min Act: -0.0000

Data Index: 264139 (C4 (Web Text))

Max Activating Token Index: 749

Text #18

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.6347. Min Act: 0.0002

Data Index: 1408587 (Python Code)

Max Activating Token Index: 41

Text #19

Max Range: 0.8659. Min Range: -0.8659

Max Act: 0.6707. Min Act: -0.0000

Data Index: 691182 (C4 (Web Text))

Max Activating Token Index: 732