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

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

Neuron 4626 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: 0.0290. Min Range: -0.0290

Max Act: 0.0290. Min Act: -0.0002

Data Index: 933761 (C4 (Web Text))

Max Activating Token Index: 672

Text #1

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0282. Min Act: -0.0002

Data Index: 1433287 (Python Code)

Max Activating Token Index: 723

Text #2

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0253. Min Act: -0.0002

Data Index: 675980 (C4 (Web Text))

Max Activating Token Index: 457

Text #3

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0256. Min Act: -0.0002

Data Index: 658398 (C4 (Web Text))

Max Activating Token Index: 588

Text #4

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0245. Min Act: -0.0002

Data Index: 545229 (C4 (Web Text))

Max Activating Token Index: 386

Text #5

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0230. Min Act: -0.0002

Data Index: 745467 (C4 (Web Text))

Max Activating Token Index: 780

Text #6

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0244. Min Act: -0.0002

Data Index: 1553079 (Python Code)

Max Activating Token Index: 486

Text #7

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0236. Min Act: -0.0002

Data Index: 1201874 (C4 (Web Text))

Max Activating Token Index: 222

Text #8

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0216. Min Act: -0.0002

Data Index: 1535508 (Python Code)

Max Activating Token Index: 239

Text #9

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0230. Min Act: -0.0002

Data Index: 1642459 (Python Code)

Max Activating Token Index: 971

Text #10

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0211. Min Act: -0.0002

Data Index: 636110 (C4 (Web Text))

Max Activating Token Index: 839

Text #11

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0227. Min Act: -0.0002

Data Index: 1110942 (C4 (Web Text))

Max Activating Token Index: 143

Text #12

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0214. Min Act: -0.0002

Data Index: 1451521 (Python Code)

Max Activating Token Index: 225

Text #13

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0227. Min Act: -0.0002

Data Index: 629961 (C4 (Web Text))

Max Activating Token Index: 198

Text #14

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0223. Min Act: -0.0002

Data Index: 1598952 (Python Code)

Max Activating Token Index: 351

Text #15

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0226. Min Act: -0.0002

Data Index: 1428262 (Python Code)

Max Activating Token Index: 735

Text #16

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0219. Min Act: -0.0002

Data Index: 1656485 (Python Code)

Max Activating Token Index: 386

Text #17

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0167. Min Act: -0.0002

Data Index: 1637164 (Python Code)

Max Activating Token Index: 203

Text #18

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0209. Min Act: -0.0002

Data Index: 1378959 (Python Code)

Max Activating Token Index: 725

Text #19

Max Range: 0.0290. Min Range: -0.0290

Max Act: 0.0222. Min Act: -0.0002

Data Index: 551961 (C4 (Web Text))

Max Activating Token Index: 291