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

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

Neuron 3150 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1395. Min Act: -0.0000

Data Index: 576567 (C4 (Web Text))

Max Activating Token Index: 516

Text #1

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1245. Min Act: -0.0000

Data Index: 1307443 (C4 (Web Text))

Max Activating Token Index: 148

Text #2

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1229. Min Act: -0.0000

Data Index: 390232 (C4 (Web Text))

Max Activating Token Index: 657

Text #3

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1187. Min Act: -0.0000

Data Index: 229603 (C4 (Web Text))

Max Activating Token Index: 637

Text #4

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1128. Min Act: -0.0000

Data Index: 1341413 (C4 (Web Text))

Max Activating Token Index: 422

Text #5

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1146. Min Act: -0.0000

Data Index: 356555 (C4 (Web Text))

Max Activating Token Index: 434

Text #6

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1085. Min Act: -0.0001

Data Index: 906088 (C4 (Web Text))

Max Activating Token Index: 517

Text #7

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1100. Min Act: -0.0000

Data Index: 400032 (C4 (Web Text))

Max Activating Token Index: 604

Text #8

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1101. Min Act: -0.0000

Data Index: 1053533 (C4 (Web Text))

Max Activating Token Index: 891

Text #9

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1098. Min Act: -0.0000

Data Index: 399705 (C4 (Web Text))

Max Activating Token Index: 980

Text #10

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1082. Min Act: -0.0001

Data Index: 1214148 (C4 (Web Text))

Max Activating Token Index: 149

Text #11

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1079. Min Act: -0.0001

Data Index: 1640701 (Python Code)

Max Activating Token Index: 325

Text #12

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1093. Min Act: -0.0000

Data Index: 340667 (C4 (Web Text))

Max Activating Token Index: 987

Text #13

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1073. Min Act: -0.0000

Data Index: 388859 (C4 (Web Text))

Max Activating Token Index: 16

Text #14

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1100. Min Act: -0.0000

Data Index: 908969 (C4 (Web Text))

Max Activating Token Index: 1020

Text #15

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1043. Min Act: -0.0001

Data Index: 315515 (C4 (Web Text))

Max Activating Token Index: 978

Text #16

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1045. Min Act: -0.0000

Data Index: 1556001 (Python Code)

Max Activating Token Index: 862

Text #17

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.0962. Min Act: -0.0001

Data Index: 775711 (C4 (Web Text))

Max Activating Token Index: 743

Text #18

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1055. Min Act: -0.0000

Data Index: 73545 (C4 (Web Text))

Max Activating Token Index: 47

Text #19

Max Range: 0.1395. Min Range: -0.1395

Max Act: 0.1030. Min Act: -0.0000

Data Index: 227758 (C4 (Web Text))

Max Activating Token Index: 24