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

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

Neuron 2559 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3234. Min Range: -1.3234

Max Act: 1.3125. Min Act: -0.0000

Data Index: 1579010 (Python Code)

Max Activating Token Index: 164

Text #1

Max Range: 1.3234. Min Range: -1.3234

Max Act: 1.3234. Min Act: -0.0000

Data Index: 1634831 (Python Code)

Max Activating Token Index: 224

Text #2

Max Range: 1.3234. Min Range: -1.3234

Max Act: 1.2047. Min Act: -0.0000

Data Index: 1190548 (C4 (Web Text))

Max Activating Token Index: 842

Text #3

Max Range: 1.3234. Min Range: -1.3234

Max Act: 1.2058. Min Act: -0.0000

Data Index: 1619851 (Python Code)

Max Activating Token Index: 489

Text #4

Max Range: 1.3234. Min Range: -1.3234

Max Act: 1.0103. Min Act: -0.0000

Data Index: 936908 (C4 (Web Text))

Max Activating Token Index: 266

Text #5

Max Range: 1.3234. Min Range: -1.3234

Max Act: 1.1045. Min Act: -0.0000

Data Index: 1631829 (Python Code)

Max Activating Token Index: 176

Text #6

Max Range: 1.3234. Min Range: -1.3234

Max Act: 1.0959. Min Act: -0.0000

Data Index: 1576957 (Python Code)

Max Activating Token Index: 496

Text #7

Max Range: 1.3234. Min Range: -1.3234

Max Act: 1.0633. Min Act: -0.0000

Data Index: 14314 (C4 (Web Text))

Max Activating Token Index: 55

Text #8

Max Range: 1.3234. Min Range: -1.3234

Max Act: 1.0334. Min Act: -0.0000

Data Index: 1582827 (Python Code)

Max Activating Token Index: 46

Text #9

Max Range: 1.3234. Min Range: -1.3234

Max Act: 0.9923. Min Act: -0.0000

Data Index: 1600995 (Python Code)

Max Activating Token Index: 920

Text #10

Max Range: 1.3234. Min Range: -1.3234

Max Act: 0.9505. Min Act: -0.0000

Data Index: 1439118 (Python Code)

Max Activating Token Index: 818

Text #11

Max Range: 1.3234. Min Range: -1.3234

Max Act: 0.9883. Min Act: -0.0000

Data Index: 1434390 (Python Code)

Max Activating Token Index: 46

Text #12

Max Range: 1.3234. Min Range: -1.3234

Max Act: 0.9002. Min Act: -0.0000

Data Index: 1507783 (Python Code)

Max Activating Token Index: 33

Text #13

Max Range: 1.3234. Min Range: -1.3234

Max Act: 1.0037. Min Act: -0.0000

Data Index: 352983 (C4 (Web Text))

Max Activating Token Index: 252

Text #14

Max Range: 1.3234. Min Range: -1.3234

Max Act: 0.9672. Min Act: -0.0000

Data Index: 1365540 (Python Code)

Max Activating Token Index: 47

Text #15

Max Range: 1.3234. Min Range: -1.3234

Max Act: 0.9297. Min Act: -0.0000

Data Index: 1257198 (C4 (Web Text))

Max Activating Token Index: 36

Text #16

Max Range: 1.3234. Min Range: -1.3234

Max Act: 0.9823. Min Act: -0.0000

Data Index: 185253 (C4 (Web Text))

Max Activating Token Index: 979

Text #17

Max Range: 1.3234. Min Range: -1.3234

Max Act: 0.9259. Min Act: -0.0000

Data Index: 720944 (C4 (Web Text))

Max Activating Token Index: 1012

Text #18

Max Range: 1.3234. Min Range: -1.3234

Max Act: 0.9834. Min Act: -0.0000

Data Index: 882744 (C4 (Web Text))

Max Activating Token Index: 464

Text #19

Max Range: 1.3234. Min Range: -1.3234

Max Act: 0.9321. Min Act: -0.0000

Data Index: 50688 (C4 (Web Text))

Max Activating Token Index: 47