Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 153 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2324. Min Act: -0.0003

Data Index: 1632728 (Python Code)

Max Activating Token Index: 780

Text #1

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2398. Min Act: -0.0003

Data Index: 1175437 (C4 (Web Text))

Max Activating Token Index: 948

Text #2

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2349. Min Act: -0.0003

Data Index: 382073 (C4 (Web Text))

Max Activating Token Index: 602

Text #3

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2174. Min Act: -0.0003

Data Index: 442845 (C4 (Web Text))

Max Activating Token Index: 724

Text #4

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2233. Min Act: -0.0003

Data Index: 1525445 (Python Code)

Max Activating Token Index: 525

Text #5

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2271. Min Act: -0.0003

Data Index: 1035978 (C4 (Web Text))

Max Activating Token Index: 32

Text #6

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2265. Min Act: -0.0004

Data Index: 1219441 (C4 (Web Text))

Max Activating Token Index: 862

Text #7

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2221. Min Act: -0.0003

Data Index: 204547 (C4 (Web Text))

Max Activating Token Index: 555

Text #8

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2133. Min Act: -0.0003

Data Index: 638173 (C4 (Web Text))

Max Activating Token Index: 871

Text #9

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2095. Min Act: -0.0003

Data Index: 1544937 (Python Code)

Max Activating Token Index: 949

Text #10

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2221. Min Act: -0.0003

Data Index: 507104 (C4 (Web Text))

Max Activating Token Index: 1001

Text #11

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2216. Min Act: -0.0003

Data Index: 105051 (C4 (Web Text))

Max Activating Token Index: 835

Text #12

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2146. Min Act: -0.0003

Data Index: 889594 (C4 (Web Text))

Max Activating Token Index: 684

Text #13

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2130. Min Act: -0.0003

Data Index: 1177603 (C4 (Web Text))

Max Activating Token Index: 139

Text #14

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2118. Min Act: -0.0003

Data Index: 679411 (C4 (Web Text))

Max Activating Token Index: 690

Text #15

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2119. Min Act: -0.0003

Data Index: 89470 (C4 (Web Text))

Max Activating Token Index: 367

Text #16

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2088. Min Act: -0.0003

Data Index: 583284 (C4 (Web Text))

Max Activating Token Index: 79

Text #17

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2032. Min Act: -0.0003

Data Index: 1198932 (C4 (Web Text))

Max Activating Token Index: 707

Text #18

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2000. Min Act: -0.0003

Data Index: 324632 (C4 (Web Text))

Max Activating Token Index: 854

Text #19

Max Range: 0.2398. Min Range: -0.2398

Max Act: 0.2038. Min Act: -0.0003

Data Index: 119429 (C4 (Web Text))

Max Activating Token Index: 799