Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 16 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0452. Min Act: -0.0004

Data Index: 1036360 (C4 (Web Text))

Max Activating Token Index: 557

Text #1

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0416. Min Act: -0.0004

Data Index: 1347264 (C4 (Web Text))

Max Activating Token Index: 664

Text #2

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0388. Min Act: -0.0004

Data Index: 969099 (C4 (Web Text))

Max Activating Token Index: 788

Text #3

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0390. Min Act: -0.0004

Data Index: 524976 (C4 (Web Text))

Max Activating Token Index: 382

Text #4

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0381. Min Act: -0.0004

Data Index: 1202074 (C4 (Web Text))

Max Activating Token Index: 468

Text #5

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0345. Min Act: -0.0004

Data Index: 79643 (C4 (Web Text))

Max Activating Token Index: 47

Text #6

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0371. Min Act: -0.0004

Data Index: 826669 (C4 (Web Text))

Max Activating Token Index: 857

Text #7

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0361. Min Act: -0.0004

Data Index: 1298599 (C4 (Web Text))

Max Activating Token Index: 790

Text #8

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0355. Min Act: -0.0004

Data Index: 65600 (C4 (Web Text))

Max Activating Token Index: 439

Text #9

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0358. Min Act: -0.0004

Data Index: 1653296 (Python Code)

Max Activating Token Index: 802

Text #10

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0359. Min Act: -0.0004

Data Index: 991875 (C4 (Web Text))

Max Activating Token Index: 55

Text #11

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0346. Min Act: -0.0004

Data Index: 946061 (C4 (Web Text))

Max Activating Token Index: 611

Text #12

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0332. Min Act: -0.0004

Data Index: 777970 (C4 (Web Text))

Max Activating Token Index: 596

Text #13

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0368. Min Act: -0.0004

Data Index: 20436 (C4 (Web Text))

Max Activating Token Index: 182

Text #14

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0345. Min Act: -0.0004

Data Index: 806811 (C4 (Web Text))

Max Activating Token Index: 35

Text #15

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0351. Min Act: -0.0004

Data Index: 1541122 (Python Code)

Max Activating Token Index: 339

Text #16

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0358. Min Act: -0.0004

Data Index: 1058392 (C4 (Web Text))

Max Activating Token Index: 139

Text #17

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0349. Min Act: -0.0004

Data Index: 631592 (C4 (Web Text))

Max Activating Token Index: 831

Text #18

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0335. Min Act: -0.0004

Data Index: 387099 (C4 (Web Text))

Max Activating Token Index: 168

Text #19

Max Range: 0.0452. Min Range: -0.0452

Max Act: 0.0363. Min Act: -0.0004

Data Index: 617898 (C4 (Web Text))

Max Activating Token Index: 561