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

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

Neuron 3959 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2499. Min Act: -0.0000

Data Index: 248695 (C4 (Web Text))

Max Activating Token Index: 675

Text #1

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2172. Min Act: -0.0001

Data Index: 496252 (C4 (Web Text))

Max Activating Token Index: 876

Text #2

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2337. Min Act: -0.0000

Data Index: 676954 (C4 (Web Text))

Max Activating Token Index: 929

Text #3

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2177. Min Act: -0.0000

Data Index: 915839 (C4 (Web Text))

Max Activating Token Index: 781

Text #4

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.1627. Min Act: -0.0000

Data Index: 1608245 (Python Code)

Max Activating Token Index: 770

Text #5

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2121. Min Act: -0.0000

Data Index: 226286 (C4 (Web Text))

Max Activating Token Index: 916

Text #6

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.1987. Min Act: -0.0000

Data Index: 39983 (C4 (Web Text))

Max Activating Token Index: 924

Text #7

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.1923. Min Act: -0.0000

Data Index: 483968 (C4 (Web Text))

Max Activating Token Index: 728

Text #8

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.1896. Min Act: -0.0001

Data Index: 708793 (C4 (Web Text))

Max Activating Token Index: 731

Text #9

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2196. Min Act: -0.0000

Data Index: 717386 (C4 (Web Text))

Max Activating Token Index: 888

Text #10

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.1844. Min Act: -0.0000

Data Index: 441525 (C4 (Web Text))

Max Activating Token Index: 233

Text #11

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2027. Min Act: -0.0001

Data Index: 1497478 (Python Code)

Max Activating Token Index: 286

Text #12

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.1996. Min Act: -0.0000

Data Index: 1152790 (C4 (Web Text))

Max Activating Token Index: 510

Text #13

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.1996. Min Act: -0.0000

Data Index: 386501 (C4 (Web Text))

Max Activating Token Index: 823

Text #14

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2093. Min Act: -0.0000

Data Index: 1421180 (Python Code)

Max Activating Token Index: 688

Text #15

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2012. Min Act: -0.0001

Data Index: 1190138 (C4 (Web Text))

Max Activating Token Index: 323

Text #16

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.1977. Min Act: -0.0001

Data Index: 1214180 (C4 (Web Text))

Max Activating Token Index: 924

Text #17

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2005. Min Act: -0.0001

Data Index: 654923 (C4 (Web Text))

Max Activating Token Index: 914

Text #18

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.1790. Min Act: -0.0001

Data Index: 250567 (C4 (Web Text))

Max Activating Token Index: 853

Text #19

Max Range: 0.2499. Min Range: -0.2499

Max Act: 0.2012. Min Act: -0.0000

Data Index: 1162688 (C4 (Web Text))

Max Activating Token Index: 208