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

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

Neuron 3512 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: 0.1960. Min Range: -0.1960

Max Act: 0.1960. Min Act: -0.0001

Data Index: 708489 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1407. Min Act: -0.0001

Data Index: 861242 (C4 (Web Text))

Max Activating Token Index: 58

Text #2

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1345. Min Act: -0.0002

Data Index: 1184871 (C4 (Web Text))

Max Activating Token Index: 35

Text #3

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1265. Min Act: -0.0001

Data Index: 936608 (C4 (Web Text))

Max Activating Token Index: 333

Text #4

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1190. Min Act: -0.0002

Data Index: 57314 (C4 (Web Text))

Max Activating Token Index: 244

Text #5

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1194. Min Act: -0.0002

Data Index: 533957 (C4 (Web Text))

Max Activating Token Index: 351

Text #6

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1185. Min Act: -0.0002

Data Index: 728814 (C4 (Web Text))

Max Activating Token Index: 54

Text #7

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1182. Min Act: -0.0001

Data Index: 288281 (C4 (Web Text))

Max Activating Token Index: 9

Text #8

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1179. Min Act: -0.0001

Data Index: 1158276 (C4 (Web Text))

Max Activating Token Index: 14

Text #9

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1232. Min Act: -0.0001

Data Index: 1484274 (Python Code)

Max Activating Token Index: 894

Text #10

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1163. Min Act: -0.0002

Data Index: 1092472 (C4 (Web Text))

Max Activating Token Index: 85

Text #11

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1162. Min Act: -0.0002

Data Index: 915083 (C4 (Web Text))

Max Activating Token Index: 46

Text #12

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1164. Min Act: -0.0001

Data Index: 498489 (C4 (Web Text))

Max Activating Token Index: 36

Text #13

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1155. Min Act: -0.0002

Data Index: 1167022 (C4 (Web Text))

Max Activating Token Index: 82

Text #14

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1140. Min Act: -0.0002

Data Index: 1162191 (C4 (Web Text))

Max Activating Token Index: 319

Text #15

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1160. Min Act: -0.0002

Data Index: 510926 (C4 (Web Text))

Max Activating Token Index: 6

Text #16

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1066. Min Act: -0.0001

Data Index: 237758 (C4 (Web Text))

Max Activating Token Index: 204

Text #17

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1137. Min Act: -0.0002

Data Index: 637062 (C4 (Web Text))

Max Activating Token Index: 344

Text #18

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1098. Min Act: -0.0002

Data Index: 1071915 (C4 (Web Text))

Max Activating Token Index: 808

Text #19

Max Range: 0.1960. Min Range: -0.1960

Max Act: 0.1144. Min Act: -0.0002

Data Index: 993413 (C4 (Web Text))

Max Activating Token Index: 170