Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1847 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1372. Min Act: -0.0001

Data Index: 652585 (C4 (Web Text))

Max Activating Token Index: 228

Text #1

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1341. Min Act: -0.0001

Data Index: 414983 (C4 (Web Text))

Max Activating Token Index: 409

Text #2

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1169. Min Act: -0.0001

Data Index: 588481 (C4 (Web Text))

Max Activating Token Index: 53

Text #3

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1093. Min Act: -0.0001

Data Index: 254748 (C4 (Web Text))

Max Activating Token Index: 573

Text #4

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1119. Min Act: -0.0001

Data Index: 490744 (C4 (Web Text))

Max Activating Token Index: 419

Text #5

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1113. Min Act: -0.0001

Data Index: 239986 (C4 (Web Text))

Max Activating Token Index: 394

Text #6

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1095. Min Act: -0.0001

Data Index: 1028339 (C4 (Web Text))

Max Activating Token Index: 97

Text #7

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1027. Min Act: -0.0001

Data Index: 586044 (C4 (Web Text))

Max Activating Token Index: 470

Text #8

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1057. Min Act: -0.0001

Data Index: 120388 (C4 (Web Text))

Max Activating Token Index: 1021

Text #9

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1012. Min Act: -0.0001

Data Index: 1269376 (C4 (Web Text))

Max Activating Token Index: 914

Text #10

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.0990. Min Act: -0.0001

Data Index: 106481 (C4 (Web Text))

Max Activating Token Index: 397

Text #11

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.0989. Min Act: -0.0001

Data Index: 220633 (C4 (Web Text))

Max Activating Token Index: 53

Text #12

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1022. Min Act: -0.0001

Data Index: 1525119 (Python Code)

Max Activating Token Index: 24

Text #13

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.0969. Min Act: -0.0001

Data Index: 279826 (C4 (Web Text))

Max Activating Token Index: 207

Text #14

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.0919. Min Act: -0.0001

Data Index: 743842 (C4 (Web Text))

Max Activating Token Index: 757

Text #15

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1018. Min Act: -0.0001

Data Index: 476832 (C4 (Web Text))

Max Activating Token Index: 150

Text #16

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.1022. Min Act: -0.0001

Data Index: 624961 (C4 (Web Text))

Max Activating Token Index: 678

Text #17

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.0910. Min Act: -0.0001

Data Index: 892049 (C4 (Web Text))

Max Activating Token Index: 657

Text #18

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.0969. Min Act: -0.0001

Data Index: 1138271 (C4 (Web Text))

Max Activating Token Index: 955

Text #19

Max Range: 0.1372. Min Range: -0.1372

Max Act: 0.0998. Min Act: -0.0001

Data Index: 477353 (C4 (Web Text))

Max Activating Token Index: 715