Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3127 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.6093. Min Act: -0.0001

Data Index: 1032471 (C4 (Web Text))

Max Activating Token Index: 158

Text #1

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.1694. Min Act: -0.0001

Data Index: 74603 (C4 (Web Text))

Max Activating Token Index: 179

Text #2

Max Range: 1.6093. Min Range: -1.6093

Max Act: 1.0211. Min Act: -0.0001

Data Index: 726942 (C4 (Web Text))

Max Activating Token Index: 725

Text #3

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9397. Min Act: -0.0002

Data Index: 314183 (C4 (Web Text))

Max Activating Token Index: 930

Text #4

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9271. Min Act: -0.0001

Data Index: 814509 (C4 (Web Text))

Max Activating Token Index: 550

Text #5

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9394. Min Act: -0.0001

Data Index: 955802 (C4 (Web Text))

Max Activating Token Index: 667

Text #6

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.7795. Min Act: -0.0002

Data Index: 1484976 (Python Code)

Max Activating Token Index: 538

Text #7

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9152. Min Act: -0.0002

Data Index: 1233214 (C4 (Web Text))

Max Activating Token Index: 531

Text #8

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.9127. Min Act: -0.0001

Data Index: 861125 (C4 (Web Text))

Max Activating Token Index: 589

Text #9

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.8496. Min Act: -0.0001

Data Index: 1461808 (Python Code)

Max Activating Token Index: 73

Text #10

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.8524. Min Act: -0.0001

Data Index: 1318388 (C4 (Web Text))

Max Activating Token Index: 118

Text #11

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.7961. Min Act: -0.0001

Data Index: 115711 (C4 (Web Text))

Max Activating Token Index: 765

Text #12

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.7931. Min Act: -0.0002

Data Index: 1239631 (C4 (Web Text))

Max Activating Token Index: 767

Text #13

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.8129. Min Act: -0.0002

Data Index: 1136677 (C4 (Web Text))

Max Activating Token Index: 25

Text #14

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.8530. Min Act: -0.0002

Data Index: 1328922 (C4 (Web Text))

Max Activating Token Index: 137

Text #15

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.8767. Min Act: -0.0001

Data Index: 1104413 (C4 (Web Text))

Max Activating Token Index: 221

Text #16

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.7975. Min Act: -0.0001

Data Index: 1300094 (C4 (Web Text))

Max Activating Token Index: 968

Text #17

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.7888. Min Act: -0.0001

Data Index: 1405822 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.7567. Min Act: -0.0002

Data Index: 1239185 (C4 (Web Text))

Max Activating Token Index: 37

Text #19

Max Range: 1.6093. Min Range: -1.6093

Max Act: 0.7821. Min Act: -0.0001

Data Index: 306874 (C4 (Web Text))

Max Activating Token Index: 387