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

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

Neuron 3701 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0211. Min Act: -0.0002

Data Index: 1249859 (C4 (Web Text))

Max Activating Token Index: 292

Text #1

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0240. Min Act: -0.0002

Data Index: 654409 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0211. Min Act: -0.0002

Data Index: 164134 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0180. Min Act: -0.0002

Data Index: 667149 (C4 (Web Text))

Max Activating Token Index: 108

Text #4

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0194. Min Act: -0.0002

Data Index: 666716 (C4 (Web Text))

Max Activating Token Index: 411

Text #5

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0179. Min Act: -0.0002

Data Index: 580584 (C4 (Web Text))

Max Activating Token Index: 567

Text #6

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0172. Min Act: -0.0002

Data Index: 276228 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0122. Min Act: -0.0002

Data Index: 1426954 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0156. Min Act: -0.0002

Data Index: 422000 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0175. Min Act: -0.0002

Data Index: 175400 (C4 (Web Text))

Max Activating Token Index: 62

Text #10

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0164. Min Act: -0.0002

Data Index: 840557 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0157. Min Act: -0.0002

Data Index: 633774 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0148. Min Act: -0.0002

Data Index: 1547703 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0171. Min Act: -0.0002

Data Index: 595894 (C4 (Web Text))

Max Activating Token Index: 78

Text #14

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0174. Min Act: -0.0002

Data Index: 1176021 (C4 (Web Text))

Max Activating Token Index: 531

Text #15

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0163. Min Act: -0.0002

Data Index: 804141 (C4 (Web Text))

Max Activating Token Index: 152

Text #16

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0178. Min Act: -0.0002

Data Index: 475319 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0177. Min Act: -0.0002

Data Index: 476583 (C4 (Web Text))

Max Activating Token Index: 81

Text #18

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0168. Min Act: -0.0002

Data Index: 979511 (C4 (Web Text))

Max Activating Token Index: 251

Text #19

Max Range: 0.0240. Min Range: -0.0240

Max Act: 0.0157. Min Act: -0.0002

Data Index: 935838 (C4 (Web Text))

Max Activating Token Index: 719