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

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

Neuron 2052 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0867. Min Act: -0.0001

Data Index: 1508946 (Python Code)

Max Activating Token Index: 672

Text #1

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.1036. Min Act: -0.0001

Data Index: 230638 (C4 (Web Text))

Max Activating Token Index: 208

Text #2

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0993. Min Act: -0.0001

Data Index: 392807 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0879. Min Act: -0.0001

Data Index: 1356436 (C4 (Web Text))

Max Activating Token Index: 21

Text #4

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.1067. Min Act: -0.0001

Data Index: 333438 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0951. Min Act: -0.0001

Data Index: 603894 (C4 (Web Text))

Max Activating Token Index: 186

Text #6

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0951. Min Act: -0.0001

Data Index: 369368 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0973. Min Act: -0.0001

Data Index: 1291295 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.1026. Min Act: -0.0001

Data Index: 898267 (C4 (Web Text))

Max Activating Token Index: 931

Text #9

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0989. Min Act: -0.0001

Data Index: 1589007 (Python Code)

Max Activating Token Index: 134

Text #10

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0732. Min Act: -0.0001

Data Index: 774031 (C4 (Web Text))

Max Activating Token Index: 758

Text #11

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0996. Min Act: -0.0001

Data Index: 903959 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0870. Min Act: -0.0001

Data Index: 917080 (C4 (Web Text))

Max Activating Token Index: 698

Text #13

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0904. Min Act: -0.0001

Data Index: 457363 (C4 (Web Text))

Max Activating Token Index: 143

Text #14

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0862. Min Act: -0.0001

Data Index: 1158105 (C4 (Web Text))

Max Activating Token Index: 598

Text #15

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0937. Min Act: -0.0001

Data Index: 772049 (C4 (Web Text))

Max Activating Token Index: 166

Text #16

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0879. Min Act: -0.0001

Data Index: 142185 (C4 (Web Text))

Max Activating Token Index: 15

Text #17

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0794. Min Act: -0.0001

Data Index: 1068754 (C4 (Web Text))

Max Activating Token Index: 668

Text #18

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0740. Min Act: -0.0001

Data Index: 1384375 (Python Code)

Max Activating Token Index: 1000

Text #19

Max Range: 0.1067. Min Range: -0.1067

Max Act: 0.0843. Min Act: -0.0001

Data Index: 441986 (C4 (Web Text))

Max Activating Token Index: 238