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

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

Neuron 100 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.6514. Min Act: -0.0000

Data Index: 1212195 (C4 (Web Text))

Max Activating Token Index: 604

Text #1

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.6342. Min Act: -0.0000

Data Index: 318420 (C4 (Web Text))

Max Activating Token Index: 295

Text #2

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.8047. Min Act: -0.0000

Data Index: 420066 (C4 (Web Text))

Max Activating Token Index: 709

Text #3

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.8072. Min Act: -0.0000

Data Index: 1036571 (C4 (Web Text))

Max Activating Token Index: 689

Text #4

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.8009. Min Act: -0.0000

Data Index: 470198 (C4 (Web Text))

Max Activating Token Index: 297

Text #5

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.7945. Min Act: -0.0000

Data Index: 111998 (C4 (Web Text))

Max Activating Token Index: 819

Text #6

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.7585. Min Act: -0.0000

Data Index: 544290 (C4 (Web Text))

Max Activating Token Index: 907

Text #7

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.7548. Min Act: -0.0000

Data Index: 903848 (C4 (Web Text))

Max Activating Token Index: 745

Text #8

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.5914. Min Act: -0.0000

Data Index: 635808 (C4 (Web Text))

Max Activating Token Index: 663

Text #9

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.6491. Min Act: -0.0000

Data Index: 268000 (C4 (Web Text))

Max Activating Token Index: 593

Text #10

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.7929. Min Act: -0.0000

Data Index: 940720 (C4 (Web Text))

Max Activating Token Index: 823

Text #11

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.5953. Min Act: -0.0000

Data Index: 301116 (C4 (Web Text))

Max Activating Token Index: 642

Text #12

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.2669. Min Act: -0.0000

Data Index: 1304823 (C4 (Web Text))

Max Activating Token Index: 733

Text #13

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.5799. Min Act: -0.0000

Data Index: 1150623 (C4 (Web Text))

Max Activating Token Index: 713

Text #14

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.5868. Min Act: -0.0000

Data Index: 518056 (C4 (Web Text))

Max Activating Token Index: 780

Text #15

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.4221. Min Act: -0.0000

Data Index: 911987 (C4 (Web Text))

Max Activating Token Index: 763

Text #16

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.5864. Min Act: -0.0000

Data Index: 944016 (C4 (Web Text))

Max Activating Token Index: 788

Text #17

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.7394. Min Act: -0.0000

Data Index: 1335349 (C4 (Web Text))

Max Activating Token Index: 910

Text #18

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.4185. Min Act: -0.0000

Data Index: 45992 (C4 (Web Text))

Max Activating Token Index: 645

Text #19

Max Range: 4.8072. Min Range: -4.8072

Max Act: 4.4357. Min Act: -0.0000

Data Index: 381538 (C4 (Web Text))

Max Activating Token Index: 459