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

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

Neuron 813 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.2273. Min Act: -0.0000

Data Index: 80457 (C4 (Web Text))

Max Activating Token Index: 95

Text #1

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.2397. Min Act: -0.0000

Data Index: 494252 (C4 (Web Text))

Max Activating Token Index: 44

Text #2

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.2189. Min Act: -0.0000

Data Index: 176444 (C4 (Web Text))

Max Activating Token Index: 971

Text #3

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.2026. Min Act: -0.0000

Data Index: 1306271 (C4 (Web Text))

Max Activating Token Index: 447

Text #4

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.2077. Min Act: -0.0000

Data Index: 148380 (C4 (Web Text))

Max Activating Token Index: 908

Text #5

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1978. Min Act: -0.0001

Data Index: 548218 (C4 (Web Text))

Max Activating Token Index: 76

Text #6

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1973. Min Act: -0.0000

Data Index: 426670 (C4 (Web Text))

Max Activating Token Index: 747

Text #7

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1843. Min Act: -0.0000

Data Index: 1114259 (C4 (Web Text))

Max Activating Token Index: 728

Text #8

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1967. Min Act: -0.0000

Data Index: 1218284 (C4 (Web Text))

Max Activating Token Index: 136

Text #9

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1958. Min Act: -0.0000

Data Index: 926158 (C4 (Web Text))

Max Activating Token Index: 21

Text #10

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1968. Min Act: -0.0000

Data Index: 1058520 (C4 (Web Text))

Max Activating Token Index: 863

Text #11

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1763. Min Act: -0.0000

Data Index: 268949 (C4 (Web Text))

Max Activating Token Index: 271

Text #12

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1799. Min Act: -0.0000

Data Index: 1103 (C4 (Web Text))

Max Activating Token Index: 220

Text #13

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1862. Min Act: -0.0000

Data Index: 884651 (C4 (Web Text))

Max Activating Token Index: 277

Text #14

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1847. Min Act: -0.0000

Data Index: 1121236 (C4 (Web Text))

Max Activating Token Index: 701

Text #15

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1787. Min Act: -0.0000

Data Index: 394925 (C4 (Web Text))

Max Activating Token Index: 938

Text #16

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1816. Min Act: -0.0000

Data Index: 628524 (C4 (Web Text))

Max Activating Token Index: 250

Text #17

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1920. Min Act: -0.0000

Data Index: 1230940 (C4 (Web Text))

Max Activating Token Index: 1007

Text #18

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1908. Min Act: -0.0001

Data Index: 439023 (C4 (Web Text))

Max Activating Token Index: 15

Text #19

Max Range: 0.2397. Min Range: -0.2397

Max Act: 0.1717. Min Act: -0.0000

Data Index: 1274516 (C4 (Web Text))

Max Activating Token Index: 55