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

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

Neuron 5040 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1586. Min Act: -0.0000

Data Index: 1509870 (Python Code)

Max Activating Token Index: 445

Text #1

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1459. Min Act: -0.0000

Data Index: 51363 (C4 (Web Text))

Max Activating Token Index: 677

Text #2

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1099. Min Act: -0.0000

Data Index: 154620 (C4 (Web Text))

Max Activating Token Index: 710

Text #3

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1223. Min Act: -0.0000

Data Index: 701230 (C4 (Web Text))

Max Activating Token Index: 551

Text #4

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1323. Min Act: -0.0000

Data Index: 506607 (C4 (Web Text))

Max Activating Token Index: 532

Text #5

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1221. Min Act: -0.0000

Data Index: 996060 (C4 (Web Text))

Max Activating Token Index: 599

Text #6

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1149. Min Act: -0.0000

Data Index: 322953 (C4 (Web Text))

Max Activating Token Index: 867

Text #7

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1142. Min Act: -0.0000

Data Index: 1428339 (Python Code)

Max Activating Token Index: 455

Text #8

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1094. Min Act: -0.0000

Data Index: 369439 (C4 (Web Text))

Max Activating Token Index: 554

Text #9

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1208. Min Act: -0.0000

Data Index: 790925 (C4 (Web Text))

Max Activating Token Index: 893

Text #10

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1251. Min Act: -0.0000

Data Index: 1047160 (C4 (Web Text))

Max Activating Token Index: 226

Text #11

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1114. Min Act: -0.0000

Data Index: 1058171 (C4 (Web Text))

Max Activating Token Index: 360

Text #12

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1138. Min Act: -0.0000

Data Index: 1275138 (C4 (Web Text))

Max Activating Token Index: 591

Text #13

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1120. Min Act: -0.0000

Data Index: 273856 (C4 (Web Text))

Max Activating Token Index: 655

Text #14

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1084. Min Act: -0.0000

Data Index: 336515 (C4 (Web Text))

Max Activating Token Index: 419

Text #15

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1112. Min Act: -0.0000

Data Index: 202842 (C4 (Web Text))

Max Activating Token Index: 733

Text #16

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1102. Min Act: -0.0000

Data Index: 1231952 (C4 (Web Text))

Max Activating Token Index: 630

Text #17

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1090. Min Act: -0.0000

Data Index: 1053579 (C4 (Web Text))

Max Activating Token Index: 787

Text #18

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1071. Min Act: -0.0000

Data Index: 1622831 (Python Code)

Max Activating Token Index: 626

Text #19

Max Range: 0.1586. Min Range: -0.1586

Max Act: 0.1059. Min Act: -0.0000

Data Index: 181051 (C4 (Web Text))

Max Activating Token Index: 667