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

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

Neuron 4677 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.3924. Min Act: -0.0000

Data Index: 1073887 (C4 (Web Text))

Max Activating Token Index: 962

Text #1

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.1612. Min Act: -0.0000

Data Index: 1556523 (Python Code)

Max Activating Token Index: 102

Text #2

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.2534. Min Act: -0.0000

Data Index: 315867 (C4 (Web Text))

Max Activating Token Index: 165

Text #3

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.1318. Min Act: -0.0000

Data Index: 1648206 (Python Code)

Max Activating Token Index: 154

Text #4

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.2414. Min Act: -0.0000

Data Index: 566208 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.0934. Min Act: -0.0000

Data Index: 686277 (C4 (Web Text))

Max Activating Token Index: 414

Text #6

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.1281. Min Act: -0.0000

Data Index: 240454 (C4 (Web Text))

Max Activating Token Index: 43

Text #7

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.1309. Min Act: -0.0000

Data Index: 83783 (C4 (Web Text))

Max Activating Token Index: 964

Text #8

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.1971. Min Act: -0.0000

Data Index: 536133 (C4 (Web Text))

Max Activating Token Index: 24

Text #9

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.1514. Min Act: -0.0000

Data Index: 1410613 (Python Code)

Max Activating Token Index: 263

Text #10

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.0663. Min Act: -0.0000

Data Index: 836549 (C4 (Web Text))

Max Activating Token Index: 637

Text #11

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.0248. Min Act: -0.0000

Data Index: 381001 (C4 (Web Text))

Max Activating Token Index: 933

Text #12

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.0717. Min Act: -0.0000

Data Index: 1414496 (Python Code)

Max Activating Token Index: 783

Text #13

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.0809. Min Act: -0.0000

Data Index: 1169454 (C4 (Web Text))

Max Activating Token Index: 355

Text #14

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.0566. Min Act: -0.0000

Data Index: 1238417 (C4 (Web Text))

Max Activating Token Index: 273

Text #15

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.1171. Min Act: -0.0000

Data Index: 767016 (C4 (Web Text))

Max Activating Token Index: 784

Text #16

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.0611. Min Act: -0.0000

Data Index: 1092474 (C4 (Web Text))

Max Activating Token Index: 439

Text #17

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.0725. Min Act: -0.0000

Data Index: 914279 (C4 (Web Text))

Max Activating Token Index: 404

Text #18

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.0363. Min Act: -0.0000

Data Index: 699891 (C4 (Web Text))

Max Activating Token Index: 23

Text #19

Max Range: 1.3924. Min Range: -1.3924

Max Act: 1.0730. Min Act: -0.0000

Data Index: 21726 (C4 (Web Text))

Max Activating Token Index: 318