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

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

Neuron 1123 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0011. Min Act: -0.0001

Data Index: 295967 (C4 (Web Text))

Max Activating Token Index: 926

Text #1

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0008. Min Act: -0.0001

Data Index: 845618 (C4 (Web Text))

Max Activating Token Index: 1011

Text #2

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0007. Min Act: -0.0001

Data Index: 479893 (C4 (Web Text))

Max Activating Token Index: 962

Text #3

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0007. Min Act: -0.0001

Data Index: 628345 (C4 (Web Text))

Max Activating Token Index: 808

Text #4

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0006. Min Act: -0.0001

Data Index: 1623476 (Python Code)

Max Activating Token Index: 109

Text #5

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0006. Min Act: -0.0001

Data Index: 32878 (C4 (Web Text))

Max Activating Token Index: 944

Text #6

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0005. Min Act: -0.0001

Data Index: 88725 (C4 (Web Text))

Max Activating Token Index: 824

Text #7

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0006. Min Act: -0.0001

Data Index: 218959 (C4 (Web Text))

Max Activating Token Index: 252

Text #8

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0005. Min Act: -0.0001

Data Index: 538341 (C4 (Web Text))

Max Activating Token Index: 947

Text #9

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0006. Min Act: -0.0001

Data Index: 175000 (C4 (Web Text))

Max Activating Token Index: 980

Text #10

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0006. Min Act: -0.0001

Data Index: 1428272 (Python Code)

Max Activating Token Index: 110

Text #11

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0005. Min Act: -0.0001

Data Index: 453009 (C4 (Web Text))

Max Activating Token Index: 1010

Text #12

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0005. Min Act: -0.0001

Data Index: 64378 (C4 (Web Text))

Max Activating Token Index: 930

Text #13

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0006. Min Act: -0.0001

Data Index: 321289 (C4 (Web Text))

Max Activating Token Index: 132

Text #14

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0005. Min Act: -0.0001

Data Index: 1237774 (C4 (Web Text))

Max Activating Token Index: 998

Text #15

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0005. Min Act: -0.0001

Data Index: 1309930 (C4 (Web Text))

Max Activating Token Index: 965

Text #16

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0005. Min Act: -0.0001

Data Index: 1387065 (Python Code)

Max Activating Token Index: 855

Text #17

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0004. Min Act: -0.0001

Data Index: 1195312 (C4 (Web Text))

Max Activating Token Index: 600

Text #18

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0004. Min Act: -0.0001

Data Index: 493652 (C4 (Web Text))

Max Activating Token Index: 977

Text #19

Max Range: 0.0011. Min Range: -0.0011

Max Act: 0.0005. Min Act: -0.0001

Data Index: 737201 (C4 (Web Text))

Max Activating Token Index: 995