Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1844 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 10.4372. Min Range: -10.4372

Max Act: 10.4372. Min Act: -0.0003

Data Index: 1057103 (C4 (Web Text))

Max Activating Token Index: 195

Text #1

Max Range: 10.4372. Min Range: -10.4372

Max Act: 10.1973. Min Act: -0.0002

Data Index: 1243438 (C4 (Web Text))

Max Activating Token Index: 478

Text #2

Max Range: 10.4372. Min Range: -10.4372

Max Act: 10.1780. Min Act: -0.0003

Data Index: 412875 (C4 (Web Text))

Max Activating Token Index: 398

Text #3

Max Range: 10.4372. Min Range: -10.4372

Max Act: 10.1334. Min Act: -0.0003

Data Index: 1347820 (C4 (Web Text))

Max Activating Token Index: 139

Text #4

Max Range: 10.4372. Min Range: -10.4372

Max Act: 10.0177. Min Act: -0.0003

Data Index: 1087152 (C4 (Web Text))

Max Activating Token Index: 211

Text #5

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.9998. Min Act: -0.0003

Data Index: 1037021 (C4 (Web Text))

Max Activating Token Index: 361

Text #6

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.9945. Min Act: -0.0003

Data Index: 62241 (C4 (Web Text))

Max Activating Token Index: 126

Text #7

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.9854. Min Act: -0.0003

Data Index: 1299992 (C4 (Web Text))

Max Activating Token Index: 318

Text #8

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.9286. Min Act: -0.0003

Data Index: 1245403 (C4 (Web Text))

Max Activating Token Index: 381

Text #9

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.8235. Min Act: -0.0003

Data Index: 894711 (C4 (Web Text))

Max Activating Token Index: 557

Text #10

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.7175. Min Act: -0.0003

Data Index: 171757 (C4 (Web Text))

Max Activating Token Index: 305

Text #11

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.8178. Min Act: -0.0003

Data Index: 1297628 (C4 (Web Text))

Max Activating Token Index: 804

Text #12

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.8124. Min Act: -0.0003

Data Index: 216898 (C4 (Web Text))

Max Activating Token Index: 147

Text #13

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.8073. Min Act: -0.0003

Data Index: 238884 (C4 (Web Text))

Max Activating Token Index: 315

Text #14

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.7971. Min Act: -0.0003

Data Index: 804818 (C4 (Web Text))

Max Activating Token Index: 5

Text #15

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.7851. Min Act: -0.0003

Data Index: 740637 (C4 (Web Text))

Max Activating Token Index: 310

Text #16

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.7741. Min Act: -0.0002

Data Index: 262488 (C4 (Web Text))

Max Activating Token Index: 218

Text #17

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.7560. Min Act: -0.0003

Data Index: 542231 (C4 (Web Text))

Max Activating Token Index: 772

Text #18

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.7415. Min Act: -0.0003

Data Index: 1097859 (C4 (Web Text))

Max Activating Token Index: 351

Text #19

Max Range: 10.4372. Min Range: -10.4372

Max Act: 9.7315. Min Act: -0.0003

Data Index: 183313 (C4 (Web Text))

Max Activating Token Index: 586