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

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

Neuron 1459 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: 0.0485. Min Range: -0.0485

Max Act: 0.0485. Min Act: -0.0003

Data Index: 419892 (C4 (Web Text))

Max Activating Token Index: 241

Text #1

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0464. Min Act: -0.0003

Data Index: 473271 (C4 (Web Text))

Max Activating Token Index: 639

Text #2

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0445. Min Act: -0.0003

Data Index: 1262231 (C4 (Web Text))

Max Activating Token Index: 985

Text #3

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0444. Min Act: -0.0003

Data Index: 802747 (C4 (Web Text))

Max Activating Token Index: 589

Text #4

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0444. Min Act: -0.0003

Data Index: 446205 (C4 (Web Text))

Max Activating Token Index: 1013

Text #5

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0418. Min Act: -0.0003

Data Index: 622276 (C4 (Web Text))

Max Activating Token Index: 367

Text #6

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0403. Min Act: -0.0001

Data Index: 12406 (C4 (Web Text))

Max Activating Token Index: 862

Text #7

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0414. Min Act: -0.0002

Data Index: 729794 (C4 (Web Text))

Max Activating Token Index: 964

Text #8

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0412. Min Act: -0.0003

Data Index: 218655 (C4 (Web Text))

Max Activating Token Index: 452

Text #9

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0418. Min Act: -0.0003

Data Index: 1338808 (C4 (Web Text))

Max Activating Token Index: 556

Text #10

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0396. Min Act: -0.0003

Data Index: 1322192 (C4 (Web Text))

Max Activating Token Index: 859

Text #11

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0421. Min Act: -0.0003

Data Index: 749715 (C4 (Web Text))

Max Activating Token Index: 872

Text #12

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0420. Min Act: -0.0003

Data Index: 1470947 (Python Code)

Max Activating Token Index: 524

Text #13

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0401. Min Act: -0.0002

Data Index: 1079178 (C4 (Web Text))

Max Activating Token Index: 426

Text #14

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0397. Min Act: -0.0003

Data Index: 441302 (C4 (Web Text))

Max Activating Token Index: 184

Text #15

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0395. Min Act: -0.0003

Data Index: 1471042 (Python Code)

Max Activating Token Index: 961

Text #16

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0381. Min Act: -0.0003

Data Index: 1594033 (Python Code)

Max Activating Token Index: 918

Text #17

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0383. Min Act: -0.0003

Data Index: 905387 (C4 (Web Text))

Max Activating Token Index: 152

Text #18

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0373. Min Act: -0.0002

Data Index: 1398873 (Python Code)

Max Activating Token Index: 650

Text #19

Max Range: 0.0485. Min Range: -0.0485

Max Act: 0.0380. Min Act: -0.0003

Data Index: 835855 (C4 (Web Text))

Max Activating Token Index: 490