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

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

Neuron 4444 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2626. Min Act: -0.0001

Data Index: 1241719 (C4 (Web Text))

Max Activating Token Index: 967

Text #1

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2620. Min Act: -0.0001

Data Index: 423355 (C4 (Web Text))

Max Activating Token Index: 877

Text #2

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2589. Min Act: -0.0001

Data Index: 316830 (C4 (Web Text))

Max Activating Token Index: 311

Text #3

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2403. Min Act: -0.0001

Data Index: 625551 (C4 (Web Text))

Max Activating Token Index: 617

Text #4

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2528. Min Act: -0.0001

Data Index: 432993 (C4 (Web Text))

Max Activating Token Index: 389

Text #5

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2613. Min Act: -0.0001

Data Index: 343596 (C4 (Web Text))

Max Activating Token Index: 4

Text #6

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2448. Min Act: -0.0001

Data Index: 202710 (C4 (Web Text))

Max Activating Token Index: 411

Text #7

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2372. Min Act: -0.0001

Data Index: 1447127 (Python Code)

Max Activating Token Index: 152

Text #8

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2503. Min Act: -0.0001

Data Index: 446431 (C4 (Web Text))

Max Activating Token Index: 51

Text #9

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2342. Min Act: -0.0001

Data Index: 1192466 (C4 (Web Text))

Max Activating Token Index: 699

Text #10

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2432. Min Act: -0.0001

Data Index: 948979 (C4 (Web Text))

Max Activating Token Index: 610

Text #11

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2322. Min Act: -0.0001

Data Index: 146363 (C4 (Web Text))

Max Activating Token Index: 506

Text #12

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2373. Min Act: -0.0001

Data Index: 857353 (C4 (Web Text))

Max Activating Token Index: 136

Text #13

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2303. Min Act: -0.0001

Data Index: 1248144 (C4 (Web Text))

Max Activating Token Index: 288

Text #14

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2290. Min Act: -0.0001

Data Index: 1378501 (Python Code)

Max Activating Token Index: 117

Text #15

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2267. Min Act: -0.0001

Data Index: 25997 (C4 (Web Text))

Max Activating Token Index: 662

Text #16

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2435. Min Act: -0.0001

Data Index: 1020942 (C4 (Web Text))

Max Activating Token Index: 100

Text #17

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2379. Min Act: -0.0001

Data Index: 12727 (C4 (Web Text))

Max Activating Token Index: 549

Text #18

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2248. Min Act: -0.0001

Data Index: 824372 (C4 (Web Text))

Max Activating Token Index: 467

Text #19

Max Range: 0.2626. Min Range: -0.2626

Max Act: 0.2433. Min Act: -0.0001

Data Index: 85651 (C4 (Web Text))

Max Activating Token Index: 598