Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 173 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1893. Min Act: -0.0001

Data Index: 628830 (C4 (Web Text))

Max Activating Token Index: 52

Text #1

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1341. Min Act: -0.0001

Data Index: 1338468 (C4 (Web Text))

Max Activating Token Index: 789

Text #2

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1315. Min Act: -0.0001

Data Index: 17997 (C4 (Web Text))

Max Activating Token Index: 268

Text #3

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1246. Min Act: -0.0001

Data Index: 162711 (C4 (Web Text))

Max Activating Token Index: 755

Text #4

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1208. Min Act: -0.0001

Data Index: 1592474 (Python Code)

Max Activating Token Index: 394

Text #5

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1199. Min Act: -0.0001

Data Index: 1378721 (Python Code)

Max Activating Token Index: 338

Text #6

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1188. Min Act: -0.0001

Data Index: 1207843 (C4 (Web Text))

Max Activating Token Index: 622

Text #7

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1275. Min Act: -0.0001

Data Index: 640203 (C4 (Web Text))

Max Activating Token Index: 263

Text #8

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1174. Min Act: -0.0001

Data Index: 584942 (C4 (Web Text))

Max Activating Token Index: 447

Text #9

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1184. Min Act: -0.0001

Data Index: 436733 (C4 (Web Text))

Max Activating Token Index: 398

Text #10

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1148. Min Act: -0.0001

Data Index: 1621537 (Python Code)

Max Activating Token Index: 65

Text #11

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1193. Min Act: -0.0001

Data Index: 1260282 (C4 (Web Text))

Max Activating Token Index: 745

Text #12

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1192. Min Act: -0.0001

Data Index: 760617 (C4 (Web Text))

Max Activating Token Index: 939

Text #13

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1123. Min Act: -0.0001

Data Index: 871164 (C4 (Web Text))

Max Activating Token Index: 312

Text #14

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1231. Min Act: -0.0001

Data Index: 1630756 (Python Code)

Max Activating Token Index: 830

Text #15

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1155. Min Act: -0.0001

Data Index: 1209722 (C4 (Web Text))

Max Activating Token Index: 922

Text #16

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1136. Min Act: -0.0001

Data Index: 538516 (C4 (Web Text))

Max Activating Token Index: 953

Text #17

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1094. Min Act: -0.0001

Data Index: 229858 (C4 (Web Text))

Max Activating Token Index: 847

Text #18

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1140. Min Act: -0.0001

Data Index: 524098 (C4 (Web Text))

Max Activating Token Index: 799

Text #19

Max Range: 0.1893. Min Range: -0.1893

Max Act: 0.1058. Min Act: -0.0001

Data Index: 145592 (C4 (Web Text))

Max Activating Token Index: 355