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

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

Neuron 4527 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.6901. Min Act: -0.0000

Data Index: 1624878 (Python Code)

Max Activating Token Index: 414

Text #1

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.3871. Min Act: -0.0000

Data Index: 72163 (C4 (Web Text))

Max Activating Token Index: 953

Text #2

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.3768. Min Act: -0.0000

Data Index: 312722 (C4 (Web Text))

Max Activating Token Index: 92

Text #3

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.3087. Min Act: -0.0000

Data Index: 1589135 (Python Code)

Max Activating Token Index: 910

Text #4

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.3002. Min Act: -0.0000

Data Index: 1490460 (Python Code)

Max Activating Token Index: 157

Text #5

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.2686. Min Act: -0.0000

Data Index: 622509 (C4 (Web Text))

Max Activating Token Index: 455

Text #6

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.1663. Min Act: -0.0000

Data Index: 740379 (C4 (Web Text))

Max Activating Token Index: 705

Text #7

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.2324. Min Act: -0.0000

Data Index: 1055393 (C4 (Web Text))

Max Activating Token Index: 890

Text #8

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.2354. Min Act: -0.0000

Data Index: 999626 (C4 (Web Text))

Max Activating Token Index: 508

Text #9

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.2244. Min Act: -0.0000

Data Index: 1398047 (Python Code)

Max Activating Token Index: 895

Text #10

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.1488. Min Act: -0.0000

Data Index: 742766 (C4 (Web Text))

Max Activating Token Index: 730

Text #11

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.1982. Min Act: -0.0000

Data Index: 825907 (C4 (Web Text))

Max Activating Token Index: 323

Text #12

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.1887. Min Act: -0.0000

Data Index: 148434 (C4 (Web Text))

Max Activating Token Index: 423

Text #13

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.1831. Min Act: -0.0000

Data Index: 1573914 (Python Code)

Max Activating Token Index: 415

Text #14

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.0763. Min Act: -0.0000

Data Index: 947811 (C4 (Web Text))

Max Activating Token Index: 143

Text #15

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.1723. Min Act: -0.0000

Data Index: 860708 (C4 (Web Text))

Max Activating Token Index: 213

Text #16

Max Range: 1.6901. Min Range: -1.6901

Max Act: 0.9734. Min Act: -0.0000

Data Index: 1399985 (Python Code)

Max Activating Token Index: 254

Text #17

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.0982. Min Act: -0.0000

Data Index: 1177711 (C4 (Web Text))

Max Activating Token Index: 573

Text #18

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.0878. Min Act: -0.0000

Data Index: 546966 (C4 (Web Text))

Max Activating Token Index: 908

Text #19

Max Range: 1.6901. Min Range: -1.6901

Max Act: 1.0239. Min Act: -0.0000

Data Index: 598391 (C4 (Web Text))

Max Activating Token Index: 133