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

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

Neuron 2108 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1255. Min Act: -0.0001

Data Index: 1618150 (Python Code)

Max Activating Token Index: 772

Text #1

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1247. Min Act: -0.0001

Data Index: 1021585 (C4 (Web Text))

Max Activating Token Index: 97

Text #2

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1162. Min Act: -0.0002

Data Index: 683270 (C4 (Web Text))

Max Activating Token Index: 24

Text #3

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1231. Min Act: -0.0002

Data Index: 371392 (C4 (Web Text))

Max Activating Token Index: 732

Text #4

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1175. Min Act: -0.0001

Data Index: 714997 (C4 (Web Text))

Max Activating Token Index: 992

Text #5

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1175. Min Act: -0.0002

Data Index: 751946 (C4 (Web Text))

Max Activating Token Index: 4

Text #6

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1156. Min Act: -0.0002

Data Index: 379959 (C4 (Web Text))

Max Activating Token Index: 598

Text #7

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1152. Min Act: -0.0001

Data Index: 789046 (C4 (Web Text))

Max Activating Token Index: 287

Text #8

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1036. Min Act: -0.0001

Data Index: 930618 (C4 (Web Text))

Max Activating Token Index: 159

Text #9

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1118. Min Act: -0.0001

Data Index: 266277 (C4 (Web Text))

Max Activating Token Index: 385

Text #10

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1116. Min Act: -0.0001

Data Index: 1219079 (C4 (Web Text))

Max Activating Token Index: 878

Text #11

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1156. Min Act: -0.0001

Data Index: 807246 (C4 (Web Text))

Max Activating Token Index: 21

Text #12

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1114. Min Act: -0.0002

Data Index: 30039 (C4 (Web Text))

Max Activating Token Index: 520

Text #13

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1114. Min Act: -0.0001

Data Index: 712361 (C4 (Web Text))

Max Activating Token Index: 888

Text #14

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1200. Min Act: -0.0002

Data Index: 24260 (C4 (Web Text))

Max Activating Token Index: 241

Text #15

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1118. Min Act: -0.0002

Data Index: 1134108 (C4 (Web Text))

Max Activating Token Index: 812

Text #16

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1009. Min Act: -0.0002

Data Index: 211747 (C4 (Web Text))

Max Activating Token Index: 776

Text #17

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1068. Min Act: -0.0002

Data Index: 1336448 (C4 (Web Text))

Max Activating Token Index: 7

Text #18

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.1067. Min Act: -0.0002

Data Index: 1210777 (C4 (Web Text))

Max Activating Token Index: 705

Text #19

Max Range: 0.1255. Min Range: -0.1255

Max Act: 0.0997. Min Act: -0.0001

Data Index: 1517903 (Python Code)

Max Activating Token Index: 21