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

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

Neuron 4889 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.5698. Min Act: -0.0000

Data Index: 1374615 (Python Code)

Max Activating Token Index: 372

Text #1

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.4876. Min Act: -0.0000

Data Index: 933014 (C4 (Web Text))

Max Activating Token Index: 860

Text #2

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.4794. Min Act: -0.0000

Data Index: 1065628 (C4 (Web Text))

Max Activating Token Index: 695

Text #3

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.4273. Min Act: -0.0000

Data Index: 1380774 (Python Code)

Max Activating Token Index: 108

Text #4

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.4290. Min Act: -0.0000

Data Index: 1527204 (Python Code)

Max Activating Token Index: 791

Text #5

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.4218. Min Act: 0.0000

Data Index: 1038775 (C4 (Web Text))

Max Activating Token Index: 792

Text #6

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.4277. Min Act: -0.0000

Data Index: 389055 (C4 (Web Text))

Max Activating Token Index: 616

Text #7

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3989. Min Act: -0.0000

Data Index: 715321 (C4 (Web Text))

Max Activating Token Index: 595

Text #8

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3956. Min Act: -0.0000

Data Index: 1410994 (Python Code)

Max Activating Token Index: 872

Text #9

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3941. Min Act: -0.0000

Data Index: 885243 (C4 (Web Text))

Max Activating Token Index: 761

Text #10

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3795. Min Act: -0.0000

Data Index: 713462 (C4 (Web Text))

Max Activating Token Index: 929

Text #11

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.4125. Min Act: 0.0000

Data Index: 213256 (C4 (Web Text))

Max Activating Token Index: 21

Text #12

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3688. Min Act: -0.0000

Data Index: 1015641 (C4 (Web Text))

Max Activating Token Index: 431

Text #13

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3680. Min Act: -0.0000

Data Index: 1231090 (C4 (Web Text))

Max Activating Token Index: 358

Text #14

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3742. Min Act: -0.0000

Data Index: 459695 (C4 (Web Text))

Max Activating Token Index: 755

Text #15

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.4049. Min Act: -0.0000

Data Index: 1076174 (C4 (Web Text))

Max Activating Token Index: 928

Text #16

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3471. Min Act: -0.0000

Data Index: 1321526 (C4 (Web Text))

Max Activating Token Index: 581

Text #17

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3577. Min Act: -0.0000

Data Index: 57014 (C4 (Web Text))

Max Activating Token Index: 149

Text #18

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3324. Min Act: -0.0000

Data Index: 708780 (C4 (Web Text))

Max Activating Token Index: 11

Text #19

Max Range: 0.5698. Min Range: -0.5698

Max Act: 0.3397. Min Act: 0.0001

Data Index: 184761 (C4 (Web Text))

Max Activating Token Index: 18