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

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

Neuron 1234 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.2842. Min Act: -0.0001

Data Index: 1298437 (C4 (Web Text))

Max Activating Token Index: 50

Text #1

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.2402. Min Act: -0.0001

Data Index: 1197882 (C4 (Web Text))

Max Activating Token Index: 32

Text #2

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.2280. Min Act: -0.0000

Data Index: 175259 (C4 (Web Text))

Max Activating Token Index: 21

Text #3

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.2125. Min Act: -0.0000

Data Index: 994706 (C4 (Web Text))

Max Activating Token Index: 136

Text #4

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.2044. Min Act: -0.0001

Data Index: 561795 (C4 (Web Text))

Max Activating Token Index: 53

Text #5

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.2040. Min Act: -0.0001

Data Index: 1186651 (C4 (Web Text))

Max Activating Token Index: 115

Text #6

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.2096. Min Act: -0.0001

Data Index: 1138118 (C4 (Web Text))

Max Activating Token Index: 100

Text #7

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1888. Min Act: -0.0000

Data Index: 838005 (C4 (Web Text))

Max Activating Token Index: 162

Text #8

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1876. Min Act: -0.0001

Data Index: 686269 (C4 (Web Text))

Max Activating Token Index: 169

Text #9

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1925. Min Act: -0.0000

Data Index: 1235014 (C4 (Web Text))

Max Activating Token Index: 96

Text #10

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1743. Min Act: -0.0001

Data Index: 1075335 (C4 (Web Text))

Max Activating Token Index: 214

Text #11

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1889. Min Act: -0.0001

Data Index: 919523 (C4 (Web Text))

Max Activating Token Index: 208

Text #12

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1753. Min Act: -0.0001

Data Index: 1654634 (Python Code)

Max Activating Token Index: 227

Text #13

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1628. Min Act: -0.0001

Data Index: 1075972 (C4 (Web Text))

Max Activating Token Index: 737

Text #14

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1861. Min Act: -0.0000

Data Index: 913616 (C4 (Web Text))

Max Activating Token Index: 57

Text #15

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1807. Min Act: -0.0001

Data Index: 691710 (C4 (Web Text))

Max Activating Token Index: 66

Text #16

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1586. Min Act: -0.0001

Data Index: 1542699 (Python Code)

Max Activating Token Index: 620

Text #17

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1630. Min Act: -0.0000

Data Index: 1038630 (C4 (Web Text))

Max Activating Token Index: 848

Text #18

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1718. Min Act: -0.0001

Data Index: 1644361 (Python Code)

Max Activating Token Index: 540

Text #19

Max Range: 0.2842. Min Range: -0.2842

Max Act: 0.1431. Min Act: -0.0001

Data Index: 1550589 (Python Code)

Max Activating Token Index: 599