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

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

Neuron 4795 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.4308. Min Act: -0.0001

Data Index: 905007 (C4 (Web Text))

Max Activating Token Index: 58

Text #1

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.3563. Min Act: -0.0001

Data Index: 1071936 (C4 (Web Text))

Max Activating Token Index: 95

Text #2

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2774. Min Act: -0.0001

Data Index: 46232 (C4 (Web Text))

Max Activating Token Index: 358

Text #3

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.3727. Min Act: -0.0001

Data Index: 298115 (C4 (Web Text))

Max Activating Token Index: 132

Text #4

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.3342. Min Act: -0.0001

Data Index: 72232 (C4 (Web Text))

Max Activating Token Index: 725

Text #5

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2906. Min Act: -0.0001

Data Index: 63644 (C4 (Web Text))

Max Activating Token Index: 28

Text #6

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.3268. Min Act: -0.0001

Data Index: 1092316 (C4 (Web Text))

Max Activating Token Index: 493

Text #7

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.3771. Min Act: -0.0001

Data Index: 534580 (C4 (Web Text))

Max Activating Token Index: 718

Text #8

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2194. Min Act: -0.0001

Data Index: 96005 (C4 (Web Text))

Max Activating Token Index: 291

Text #9

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2684. Min Act: -0.0001

Data Index: 34278 (C4 (Web Text))

Max Activating Token Index: 696

Text #10

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2511. Min Act: -0.0001

Data Index: 497883 (C4 (Web Text))

Max Activating Token Index: 131

Text #11

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2812. Min Act: -0.0001

Data Index: 76579 (C4 (Web Text))

Max Activating Token Index: 898

Text #12

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2743. Min Act: -0.0001

Data Index: 109930 (C4 (Web Text))

Max Activating Token Index: 744

Text #13

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2595. Min Act: -0.0001

Data Index: 409858 (C4 (Web Text))

Max Activating Token Index: 161

Text #14

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.3010. Min Act: -0.0001

Data Index: 379726 (C4 (Web Text))

Max Activating Token Index: 197

Text #15

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2122. Min Act: -0.0001

Data Index: 907856 (C4 (Web Text))

Max Activating Token Index: 672

Text #16

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2932. Min Act: -0.0001

Data Index: 143725 (C4 (Web Text))

Max Activating Token Index: 431

Text #17

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2446. Min Act: -0.0001

Data Index: 849912 (C4 (Web Text))

Max Activating Token Index: 297

Text #18

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.3245. Min Act: -0.0001

Data Index: 499162 (C4 (Web Text))

Max Activating Token Index: 428

Text #19

Max Range: 0.4308. Min Range: -0.4308

Max Act: 0.2918. Min Act: -0.0001

Data Index: 950661 (C4 (Web Text))

Max Activating Token Index: 130