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

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

Neuron 2767 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: 0.1689. Min Range: -0.1689

Max Act: 0.1689. Min Act: -0.0001

Data Index: 478202 (C4 (Web Text))

Max Activating Token Index: 188

Text #1

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1365. Min Act: 0.0000

Data Index: 1211350 (C4 (Web Text))

Max Activating Token Index: 520

Text #2

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1592. Min Act: -0.0000

Data Index: 107074 (C4 (Web Text))

Max Activating Token Index: 701

Text #3

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1309. Min Act: -0.0000

Data Index: 240941 (C4 (Web Text))

Max Activating Token Index: 733

Text #4

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1269. Min Act: -0.0000

Data Index: 740780 (C4 (Web Text))

Max Activating Token Index: 395

Text #5

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1269. Min Act: -0.0001

Data Index: 253214 (C4 (Web Text))

Max Activating Token Index: 610

Text #6

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1228. Min Act: -0.0001

Data Index: 293683 (C4 (Web Text))

Max Activating Token Index: 383

Text #7

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1244. Min Act: -0.0000

Data Index: 965640 (C4 (Web Text))

Max Activating Token Index: 390

Text #8

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1141. Min Act: -0.0000

Data Index: 1146854 (C4 (Web Text))

Max Activating Token Index: 532

Text #9

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1212. Min Act: -0.0000

Data Index: 880975 (C4 (Web Text))

Max Activating Token Index: 871

Text #10

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1168. Min Act: -0.0000

Data Index: 906282 (C4 (Web Text))

Max Activating Token Index: 759

Text #11

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1119. Min Act: -0.0000

Data Index: 1101940 (C4 (Web Text))

Max Activating Token Index: 573

Text #12

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1138. Min Act: -0.0000

Data Index: 47617 (C4 (Web Text))

Max Activating Token Index: 516

Text #13

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1054. Min Act: -0.0001

Data Index: 872253 (C4 (Web Text))

Max Activating Token Index: 755

Text #14

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1070. Min Act: -0.0000

Data Index: 504683 (C4 (Web Text))

Max Activating Token Index: 18

Text #15

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1002. Min Act: -0.0000

Data Index: 958102 (C4 (Web Text))

Max Activating Token Index: 99

Text #16

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1023. Min Act: -0.0001

Data Index: 1380227 (Python Code)

Max Activating Token Index: 96

Text #17

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1012. Min Act: -0.0000

Data Index: 479637 (C4 (Web Text))

Max Activating Token Index: 19

Text #18

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.1017. Min Act: -0.0001

Data Index: 635123 (C4 (Web Text))

Max Activating Token Index: 356

Text #19

Max Range: 0.1689. Min Range: -0.1689

Max Act: 0.0984. Min Act: -0.0000

Data Index: 819418 (C4 (Web Text))

Max Activating Token Index: 320