Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 563 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.3145. Min Act: -0.0000

Data Index: 1134988 (C4 (Web Text))

Max Activating Token Index: 855

Text #1

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.3076. Min Act: -0.0000

Data Index: 797781 (C4 (Web Text))

Max Activating Token Index: 100

Text #2

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2865. Min Act: -0.0000

Data Index: 738359 (C4 (Web Text))

Max Activating Token Index: 685

Text #3

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2873. Min Act: -0.0000

Data Index: 404865 (C4 (Web Text))

Max Activating Token Index: 368

Text #4

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2844. Min Act: -0.0000

Data Index: 71268 (C4 (Web Text))

Max Activating Token Index: 757

Text #5

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2749. Min Act: -0.0000

Data Index: 682352 (C4 (Web Text))

Max Activating Token Index: 850

Text #6

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2734. Min Act: -0.0000

Data Index: 1234129 (C4 (Web Text))

Max Activating Token Index: 215

Text #7

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2551. Min Act: -0.0000

Data Index: 484584 (C4 (Web Text))

Max Activating Token Index: 814

Text #8

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2548. Min Act: -0.0000

Data Index: 828733 (C4 (Web Text))

Max Activating Token Index: 50

Text #9

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2631. Min Act: -0.0000

Data Index: 979163 (C4 (Web Text))

Max Activating Token Index: 480

Text #10

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2544. Min Act: -0.0000

Data Index: 1357340 (C4 (Web Text))

Max Activating Token Index: 53

Text #11

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2432. Min Act: -0.0000

Data Index: 1269727 (C4 (Web Text))

Max Activating Token Index: 519

Text #12

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2591. Min Act: -0.0000

Data Index: 379790 (C4 (Web Text))

Max Activating Token Index: 70

Text #13

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2577. Min Act: -0.0000

Data Index: 1602671 (Python Code)

Max Activating Token Index: 335

Text #14

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2557. Min Act: -0.0000

Data Index: 690080 (C4 (Web Text))

Max Activating Token Index: 351

Text #15

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2528. Min Act: 0.0000

Data Index: 1620675 (Python Code)

Max Activating Token Index: 470

Text #16

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2346. Min Act: -0.0000

Data Index: 405827 (C4 (Web Text))

Max Activating Token Index: 1008

Text #17

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2408. Min Act: -0.0000

Data Index: 396158 (C4 (Web Text))

Max Activating Token Index: 358

Text #18

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2399. Min Act: -0.0000

Data Index: 498843 (C4 (Web Text))

Max Activating Token Index: 725

Text #19

Max Range: 0.3145. Min Range: -0.3145

Max Act: 0.2367. Min Act: -0.0000

Data Index: 536636 (C4 (Web Text))

Max Activating Token Index: 453