Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1670 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1326. Min Act: -0.0005

Data Index: 1560489 (Python Code)

Max Activating Token Index: 15

Text #1

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1279. Min Act: -0.0005

Data Index: 44709 (C4 (Web Text))

Max Activating Token Index: 344

Text #2

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1277. Min Act: -0.0005

Data Index: 475621 (C4 (Web Text))

Max Activating Token Index: 16

Text #3

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1237. Min Act: -0.0005

Data Index: 1004694 (C4 (Web Text))

Max Activating Token Index: 720

Text #4

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1238. Min Act: -0.0005

Data Index: 782681 (C4 (Web Text))

Max Activating Token Index: 19

Text #5

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1161. Min Act: -0.0005

Data Index: 1251367 (C4 (Web Text))

Max Activating Token Index: 33

Text #6

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1247. Min Act: -0.0005

Data Index: 717153 (C4 (Web Text))

Max Activating Token Index: 440

Text #7

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1173. Min Act: -0.0005

Data Index: 1288108 (C4 (Web Text))

Max Activating Token Index: 650

Text #8

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1081. Min Act: -0.0005

Data Index: 219382 (C4 (Web Text))

Max Activating Token Index: 822

Text #9

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1162. Min Act: -0.0005

Data Index: 1096137 (C4 (Web Text))

Max Activating Token Index: 557

Text #10

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1219. Min Act: -0.0005

Data Index: 564097 (C4 (Web Text))

Max Activating Token Index: 600

Text #11

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1154. Min Act: -0.0005

Data Index: 228096 (C4 (Web Text))

Max Activating Token Index: 966

Text #12

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1198. Min Act: -0.0004

Data Index: 238993 (C4 (Web Text))

Max Activating Token Index: 343

Text #13

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1138. Min Act: -0.0005

Data Index: 1060293 (C4 (Web Text))

Max Activating Token Index: 775

Text #14

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1169. Min Act: -0.0005

Data Index: 686619 (C4 (Web Text))

Max Activating Token Index: 461

Text #15

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1153. Min Act: -0.0005

Data Index: 1554497 (Python Code)

Max Activating Token Index: 530

Text #16

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1147. Min Act: -0.0005

Data Index: 894240 (C4 (Web Text))

Max Activating Token Index: 728

Text #17

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1180. Min Act: -0.0005

Data Index: 1608857 (Python Code)

Max Activating Token Index: 879

Text #18

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1110. Min Act: -0.0005

Data Index: 1515760 (Python Code)

Max Activating Token Index: 390

Text #19

Max Range: 0.1326. Min Range: -0.1326

Max Act: 0.1188. Min Act: -0.0005

Data Index: 1618237 (Python Code)

Max Activating Token Index: 792