Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1581 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.5440. Min Act: -0.0003

Data Index: 89326 (C4 (Web Text))

Max Activating Token Index: 268

Text #1

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.5053. Min Act: -0.0003

Data Index: 257149 (C4 (Web Text))

Max Activating Token Index: 663

Text #2

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4934. Min Act: -0.0003

Data Index: 799906 (C4 (Web Text))

Max Activating Token Index: 205

Text #3

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4873. Min Act: -0.0003

Data Index: 294113 (C4 (Web Text))

Max Activating Token Index: 41

Text #4

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4315. Min Act: -0.0003

Data Index: 890971 (C4 (Web Text))

Max Activating Token Index: 82

Text #5

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4614. Min Act: -0.0003

Data Index: 278093 (C4 (Web Text))

Max Activating Token Index: 483

Text #6

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4607. Min Act: -0.0003

Data Index: 286697 (C4 (Web Text))

Max Activating Token Index: 134

Text #7

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4263. Min Act: -0.0003

Data Index: 1218333 (C4 (Web Text))

Max Activating Token Index: 18

Text #8

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4401. Min Act: -0.0003

Data Index: 444561 (C4 (Web Text))

Max Activating Token Index: 726

Text #9

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4503. Min Act: -0.0003

Data Index: 712431 (C4 (Web Text))

Max Activating Token Index: 931

Text #10

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4137. Min Act: -0.0003

Data Index: 454896 (C4 (Web Text))

Max Activating Token Index: 418

Text #11

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.3872. Min Act: -0.0003

Data Index: 779437 (C4 (Web Text))

Max Activating Token Index: 419

Text #12

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4199. Min Act: -0.0003

Data Index: 788131 (C4 (Web Text))

Max Activating Token Index: 434

Text #13

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4198. Min Act: -0.0003

Data Index: 303291 (C4 (Web Text))

Max Activating Token Index: 963

Text #14

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4131. Min Act: -0.0003

Data Index: 1002940 (C4 (Web Text))

Max Activating Token Index: 877

Text #15

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4155. Min Act: -0.0003

Data Index: 1071773 (C4 (Web Text))

Max Activating Token Index: 68

Text #16

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4404. Min Act: -0.0003

Data Index: 940091 (C4 (Web Text))

Max Activating Token Index: 968

Text #17

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4015. Min Act: -0.0003

Data Index: 1280207 (C4 (Web Text))

Max Activating Token Index: 648

Text #18

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.3899. Min Act: -0.0003

Data Index: 697808 (C4 (Web Text))

Max Activating Token Index: 854

Text #19

Max Range: 0.5440. Min Range: -0.5440

Max Act: 0.4083. Min Act: -0.0003

Data Index: 171564 (C4 (Web Text))

Max Activating Token Index: 134