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

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

Neuron 4526 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 6.3376. Min Range: -6.3376

Max Act: 6.3087. Min Act: -0.0000

Data Index: 736691 (C4 (Web Text))

Max Activating Token Index: 624

Text #1

Max Range: 6.3376. Min Range: -6.3376

Max Act: 6.0388. Min Act: -0.0000

Data Index: 315957 (C4 (Web Text))

Max Activating Token Index: 119

Text #2

Max Range: 6.3376. Min Range: -6.3376

Max Act: 6.2982. Min Act: -0.0000

Data Index: 648866 (C4 (Web Text))

Max Activating Token Index: 365

Text #3

Max Range: 6.3376. Min Range: -6.3376

Max Act: 6.3376. Min Act: -0.0000

Data Index: 1016088 (C4 (Web Text))

Max Activating Token Index: 745

Text #4

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.8947. Min Act: -0.0000

Data Index: 1080341 (C4 (Web Text))

Max Activating Token Index: 557

Text #5

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.8489. Min Act: -0.0000

Data Index: 981919 (C4 (Web Text))

Max Activating Token Index: 388

Text #6

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.5441. Min Act: -0.0000

Data Index: 642518 (C4 (Web Text))

Max Activating Token Index: 825

Text #7

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.7206. Min Act: -0.0000

Data Index: 979865 (C4 (Web Text))

Max Activating Token Index: 97

Text #8

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.4200. Min Act: -0.0000

Data Index: 678096 (C4 (Web Text))

Max Activating Token Index: 875

Text #9

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.3222. Min Act: -0.0000

Data Index: 991300 (C4 (Web Text))

Max Activating Token Index: 662

Text #10

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.5078. Min Act: -0.0000

Data Index: 1018137 (C4 (Web Text))

Max Activating Token Index: 130

Text #11

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.2518. Min Act: -0.0000

Data Index: 777223 (C4 (Web Text))

Max Activating Token Index: 369

Text #12

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.1392. Min Act: -0.0000

Data Index: 268970 (C4 (Web Text))

Max Activating Token Index: 794

Text #13

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.4180. Min Act: -0.0000

Data Index: 201858 (C4 (Web Text))

Max Activating Token Index: 12

Text #14

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.3699. Min Act: -0.0000

Data Index: 547564 (C4 (Web Text))

Max Activating Token Index: 712

Text #15

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.3371. Min Act: -0.0000

Data Index: 1071548 (C4 (Web Text))

Max Activating Token Index: 357

Text #16

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.1642. Min Act: -0.0000

Data Index: 223805 (C4 (Web Text))

Max Activating Token Index: 605

Text #17

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.2740. Min Act: -0.0000

Data Index: 137245 (C4 (Web Text))

Max Activating Token Index: 806

Text #18

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.5156. Min Act: -0.0000

Data Index: 832477 (C4 (Web Text))

Max Activating Token Index: 671

Text #19

Max Range: 6.3376. Min Range: -6.3376

Max Act: 5.0793. Min Act: -0.0000

Data Index: 396737 (C4 (Web Text))

Max Activating Token Index: 980