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

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

Neuron 0 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: 0.0174. Min Range: -0.0174

Max Act: 0.0174. Min Act: -0.0002

Data Index: 828447 (C4 (Web Text))

Max Activating Token Index: 16

Text #1

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0156. Min Act: -0.0002

Data Index: 1529005 (Python Code)

Max Activating Token Index: 19

Text #2

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0145. Min Act: -0.0003

Data Index: 285741 (C4 (Web Text))

Max Activating Token Index: 759

Text #3

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0143. Min Act: -0.0003

Data Index: 1600404 (Python Code)

Max Activating Token Index: 540

Text #4

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0139. Min Act: -0.0003

Data Index: 1076107 (C4 (Web Text))

Max Activating Token Index: 12

Text #5

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0132. Min Act: -0.0003

Data Index: 231905 (C4 (Web Text))

Max Activating Token Index: 20

Text #6

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0127. Min Act: -0.0002

Data Index: 573162 (C4 (Web Text))

Max Activating Token Index: 14

Text #7

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0129. Min Act: -0.0003

Data Index: 813230 (C4 (Web Text))

Max Activating Token Index: 961

Text #8

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0126. Min Act: -0.0002

Data Index: 1532868 (Python Code)

Max Activating Token Index: 7

Text #9

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0119. Min Act: -0.0003

Data Index: 1287929 (C4 (Web Text))

Max Activating Token Index: 151

Text #10

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0116. Min Act: -0.0003

Data Index: 847205 (C4 (Web Text))

Max Activating Token Index: 904

Text #11

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0120. Min Act: -0.0003

Data Index: 1121634 (C4 (Web Text))

Max Activating Token Index: 545

Text #12

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0118. Min Act: -0.0003

Data Index: 767089 (C4 (Web Text))

Max Activating Token Index: 914

Text #13

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0118. Min Act: -0.0003

Data Index: 1400850 (Python Code)

Max Activating Token Index: 16

Text #14

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0110. Min Act: -0.0003

Data Index: 790942 (C4 (Web Text))

Max Activating Token Index: 231

Text #15

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0114. Min Act: -0.0003

Data Index: 761775 (C4 (Web Text))

Max Activating Token Index: 150

Text #16

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0114. Min Act: -0.0002

Data Index: 1175712 (C4 (Web Text))

Max Activating Token Index: 14

Text #17

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0116. Min Act: -0.0003

Data Index: 1204976 (C4 (Web Text))

Max Activating Token Index: 873

Text #18

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0109. Min Act: -0.0003

Data Index: 406030 (C4 (Web Text))

Max Activating Token Index: 826

Text #19

Max Range: 0.0174. Min Range: -0.0174

Max Act: 0.0112. Min Act: -0.0002

Data Index: 216433 (C4 (Web Text))

Max Activating Token Index: 668