Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: The Pile

Neuron 168 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l-pile')

Text #0

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.6970. Min Act: -0.0000

Data Index: 362330 (The Pile)

Max Activating Token Index: 204

Text #1

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.8794. Min Act: -0.0000

Data Index: 588398 (The Pile)

Max Activating Token Index: 14

Text #2

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.6349. Min Act: -0.0000

Data Index: 1493696 (The Pile)

Max Activating Token Index: 95

Text #3

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.3771. Min Act: -0.0000

Data Index: 1682041 (The Pile)

Max Activating Token Index: 667

Text #4

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.5158. Min Act: -0.0000

Data Index: 411635 (The Pile)

Max Activating Token Index: 70

Text #5

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.3931. Min Act: -0.0000

Data Index: 1835277 (The Pile)

Max Activating Token Index: 111

Text #6

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.5148. Min Act: -0.0000

Data Index: 169739 (The Pile)

Max Activating Token Index: 234

Text #7

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.4893. Min Act: -0.0000

Data Index: 931420 (The Pile)

Max Activating Token Index: 326

Text #8

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.4537. Min Act: -0.0000

Data Index: 102023 (The Pile)

Max Activating Token Index: 102

Text #9

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.4985. Min Act: -0.0000

Data Index: 432866 (The Pile)

Max Activating Token Index: 107

Text #10

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.3974. Min Act: -0.0000

Data Index: 37615 (The Pile)

Max Activating Token Index: 102

Text #11

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.2462. Min Act: -0.0000

Data Index: 1258947 (The Pile)

Max Activating Token Index: 133

Text #12

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.1958. Min Act: -0.0000

Data Index: 1943436 (The Pile)

Max Activating Token Index: 525

Text #13

Max Range: 3.8794. Min Range: -3.8794

Max Act: 2.9785. Min Act: -0.0000

Data Index: 494997 (The Pile)

Max Activating Token Index: 45

Text #14

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.3165. Min Act: -0.0000

Data Index: 474394 (The Pile)

Max Activating Token Index: 133

Text #15

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.3120. Min Act: -0.0000

Data Index: 254896 (The Pile)

Max Activating Token Index: 249

Text #16

Max Range: 3.8794. Min Range: -3.8794

Max Act: 2.9843. Min Act: -0.0000

Data Index: 203729 (The Pile)

Max Activating Token Index: 28

Text #17

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.1040. Min Act: -0.0000

Data Index: 1301901 (The Pile)

Max Activating Token Index: 354

Text #18

Max Range: 3.8794. Min Range: -3.8794

Max Act: 2.9518. Min Act: -0.0000

Data Index: 106761 (The Pile)

Max Activating Token Index: 215

Text #19

Max Range: 3.8794. Min Range: -3.8794

Max Act: 3.3389. Min Act: -0.0000

Data Index: 716914 (The Pile)

Max Activating Token Index: 170