Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5103 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.1229. Min Act: -0.0000

Data Index: 827878 (C4 (Web Text))

Max Activating Token Index: 306

Text #1

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.1136. Min Act: -0.0000

Data Index: 960199 (C4 (Web Text))

Max Activating Token Index: 305

Text #2

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.1227. Min Act: -0.0000

Data Index: 725759 (C4 (Web Text))

Max Activating Token Index: 167

Text #3

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.1082. Min Act: -0.0000

Data Index: 772205 (C4 (Web Text))

Max Activating Token Index: 129

Text #4

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0983. Min Act: -0.0000

Data Index: 940221 (C4 (Web Text))

Max Activating Token Index: 832

Text #5

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0917. Min Act: -0.0000

Data Index: 1262628 (C4 (Web Text))

Max Activating Token Index: 807

Text #6

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0795. Min Act: -0.0000

Data Index: 779936 (C4 (Web Text))

Max Activating Token Index: 851

Text #7

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0858. Min Act: -0.0000

Data Index: 812939 (C4 (Web Text))

Max Activating Token Index: 94

Text #8

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0750. Min Act: -0.0000

Data Index: 1460469 (Python Code)

Max Activating Token Index: 911

Text #9

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0713. Min Act: -0.0000

Data Index: 1270760 (C4 (Web Text))

Max Activating Token Index: 747

Text #10

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0608. Min Act: -0.0000

Data Index: 1279028 (C4 (Web Text))

Max Activating Token Index: 1012

Text #11

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0567. Min Act: -0.0000

Data Index: 1471799 (Python Code)

Max Activating Token Index: 305

Text #12

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0609. Min Act: -0.0000

Data Index: 495065 (C4 (Web Text))

Max Activating Token Index: 777

Text #13

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0621. Min Act: -0.0000

Data Index: 1653810 (Python Code)

Max Activating Token Index: 419

Text #14

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0527. Min Act: -0.0000

Data Index: 29459 (C4 (Web Text))

Max Activating Token Index: 962

Text #15

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0559. Min Act: -0.0000

Data Index: 669691 (C4 (Web Text))

Max Activating Token Index: 892

Text #16

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0583. Min Act: -0.0000

Data Index: 135078 (C4 (Web Text))

Max Activating Token Index: 221

Text #17

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0602. Min Act: -0.0000

Data Index: 1519487 (Python Code)

Max Activating Token Index: 495

Text #18

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0542. Min Act: -0.0000

Data Index: 1403690 (Python Code)

Max Activating Token Index: 529

Text #19

Max Range: 0.1229. Min Range: -0.1229

Max Act: 0.0518. Min Act: -0.0000

Data Index: 1055086 (C4 (Web Text))

Max Activating Token Index: 507