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

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

Neuron 1061 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.5596. Min Act: -0.0000

Data Index: 496651 (C4 (Web Text))

Max Activating Token Index: 715

Text #1

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.2766. Min Act: -0.0000

Data Index: 612990 (C4 (Web Text))

Max Activating Token Index: 493

Text #2

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.2640. Min Act: -0.0000

Data Index: 1261302 (C4 (Web Text))

Max Activating Token Index: 624

Text #3

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.1459. Min Act: -0.0000

Data Index: 1452455 (Python Code)

Max Activating Token Index: 74

Text #4

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.1698. Min Act: -0.0000

Data Index: 1049540 (C4 (Web Text))

Max Activating Token Index: 24

Text #5

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.1960. Min Act: -0.0000

Data Index: 233269 (C4 (Web Text))

Max Activating Token Index: 941

Text #6

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.1367. Min Act: -0.0000

Data Index: 1419856 (Python Code)

Max Activating Token Index: 221

Text #7

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.1812. Min Act: -0.0000

Data Index: 984379 (C4 (Web Text))

Max Activating Token Index: 204

Text #8

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0231. Min Act: -0.0000

Data Index: 300859 (C4 (Web Text))

Max Activating Token Index: 400

Text #9

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0953. Min Act: -0.0000

Data Index: 1139204 (C4 (Web Text))

Max Activating Token Index: 24

Text #10

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0734. Min Act: -0.0000

Data Index: 26267 (C4 (Web Text))

Max Activating Token Index: 20

Text #11

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0667. Min Act: -0.0000

Data Index: 721876 (C4 (Web Text))

Max Activating Token Index: 95

Text #12

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0613. Min Act: -0.0000

Data Index: 684702 (C4 (Web Text))

Max Activating Token Index: 176

Text #13

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0208. Min Act: -0.0000

Data Index: 352789 (C4 (Web Text))

Max Activating Token Index: 187

Text #14

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0089. Min Act: -0.0000

Data Index: 1437876 (Python Code)

Max Activating Token Index: 704

Text #15

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0773. Min Act: -0.0000

Data Index: 1121018 (C4 (Web Text))

Max Activating Token Index: 22

Text #16

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0027. Min Act: -0.0000

Data Index: 1195825 (C4 (Web Text))

Max Activating Token Index: 736

Text #17

Max Range: 1.5596. Min Range: -1.5596

Max Act: 0.9834. Min Act: -0.0000

Data Index: 642255 (C4 (Web Text))

Max Activating Token Index: 292

Text #18

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0530. Min Act: -0.0000

Data Index: 831352 (C4 (Web Text))

Max Activating Token Index: 97

Text #19

Max Range: 1.5596. Min Range: -1.5596

Max Act: 1.0101. Min Act: -0.0000

Data Index: 374941 (C4 (Web Text))

Max Activating Token Index: 13