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

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

Neuron 1875 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: 0.0395. Min Range: -0.0395

Max Act: 0.0395. Min Act: -0.0003

Data Index: 1339141 (C4 (Web Text))

Max Activating Token Index: 253

Text #1

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0363. Min Act: -0.0003

Data Index: 1523387 (Python Code)

Max Activating Token Index: 701

Text #2

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0367. Min Act: -0.0004

Data Index: 1079846 (C4 (Web Text))

Max Activating Token Index: 32

Text #3

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0381. Min Act: -0.0003

Data Index: 1639272 (Python Code)

Max Activating Token Index: 1020

Text #4

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0358. Min Act: -0.0003

Data Index: 1413615 (Python Code)

Max Activating Token Index: 54

Text #5

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0347. Min Act: -0.0003

Data Index: 1175397 (C4 (Web Text))

Max Activating Token Index: 679

Text #6

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0351. Min Act: -0.0003

Data Index: 334140 (C4 (Web Text))

Max Activating Token Index: 652

Text #7

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0350. Min Act: -0.0003

Data Index: 467884 (C4 (Web Text))

Max Activating Token Index: 251

Text #8

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0343. Min Act: -0.0003

Data Index: 647124 (C4 (Web Text))

Max Activating Token Index: 79

Text #9

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0324. Min Act: -0.0003

Data Index: 151632 (C4 (Web Text))

Max Activating Token Index: 294

Text #10

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0325. Min Act: -0.0003

Data Index: 1194121 (C4 (Web Text))

Max Activating Token Index: 75

Text #11

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0323. Min Act: -0.0003

Data Index: 1587898 (Python Code)

Max Activating Token Index: 136

Text #12

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0333. Min Act: -0.0003

Data Index: 1407660 (Python Code)

Max Activating Token Index: 948

Text #13

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0323. Min Act: -0.0003

Data Index: 999648 (C4 (Web Text))

Max Activating Token Index: 168

Text #14

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0320. Min Act: -0.0003

Data Index: 1279028 (C4 (Web Text))

Max Activating Token Index: 886

Text #15

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0334. Min Act: -0.0003

Data Index: 160712 (C4 (Web Text))

Max Activating Token Index: 943

Text #16

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0317. Min Act: -0.0003

Data Index: 623218 (C4 (Web Text))

Max Activating Token Index: 655

Text #17

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0316. Min Act: -0.0003

Data Index: 1215197 (C4 (Web Text))

Max Activating Token Index: 173

Text #18

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0315. Min Act: -0.0004

Data Index: 309494 (C4 (Web Text))

Max Activating Token Index: 49

Text #19

Max Range: 0.0395. Min Range: -0.0395

Max Act: 0.0316. Min Act: -0.0004

Data Index: 388162 (C4 (Web Text))

Max Activating Token Index: 43