Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 3065 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1414. Min Act: -0.0001

Data Index: 209901 (C4 (Web Text))

Max Activating Token Index: 533

Text #1

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1340. Min Act: -0.0001

Data Index: 708151 (C4 (Web Text))

Max Activating Token Index: 978

Text #2

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1348. Min Act: -0.0001

Data Index: 761867 (C4 (Web Text))

Max Activating Token Index: 687

Text #3

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1198. Min Act: -0.0001

Data Index: 896636 (C4 (Web Text))

Max Activating Token Index: 879

Text #4

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1145. Min Act: -0.0001

Data Index: 1048345 (C4 (Web Text))

Max Activating Token Index: 686

Text #5

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1107. Min Act: -0.0001

Data Index: 1568521 (Python Code)

Max Activating Token Index: 42

Text #6

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1170. Min Act: -0.0001

Data Index: 1079659 (C4 (Web Text))

Max Activating Token Index: 683

Text #7

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1102. Min Act: -0.0001

Data Index: 1300260 (C4 (Web Text))

Max Activating Token Index: 927

Text #8

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1089. Min Act: -0.0001

Data Index: 430136 (C4 (Web Text))

Max Activating Token Index: 900

Text #9

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1171. Min Act: -0.0001

Data Index: 968316 (C4 (Web Text))

Max Activating Token Index: 680

Text #10

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1063. Min Act: -0.0001

Data Index: 577203 (C4 (Web Text))

Max Activating Token Index: 461

Text #11

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1086. Min Act: -0.0001

Data Index: 216092 (C4 (Web Text))

Max Activating Token Index: 1012

Text #12

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1107. Min Act: -0.0001

Data Index: 1013812 (C4 (Web Text))

Max Activating Token Index: 1019

Text #13

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1089. Min Act: -0.0001

Data Index: 253505 (C4 (Web Text))

Max Activating Token Index: 462

Text #14

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1132. Min Act: -0.0001

Data Index: 1590201 (Python Code)

Max Activating Token Index: 54

Text #15

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1087. Min Act: -0.0001

Data Index: 742622 (C4 (Web Text))

Max Activating Token Index: 965

Text #16

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1017. Min Act: -0.0001

Data Index: 1533659 (Python Code)

Max Activating Token Index: 216

Text #17

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1102. Min Act: -0.0001

Data Index: 253758 (C4 (Web Text))

Max Activating Token Index: 760

Text #18

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1055. Min Act: -0.0001

Data Index: 1378032 (Python Code)

Max Activating Token Index: 137

Text #19

Max Range: 0.1414. Min Range: -0.1414

Max Act: 0.1060. Min Act: -0.0001

Data Index: 240959 (C4 (Web Text))

Max Activating Token Index: 138