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

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

Neuron 3586 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: 0.4071. Min Range: -0.4071

Max Act: 0.4071. Min Act: -0.0000

Data Index: 60186 (C4 (Web Text))

Max Activating Token Index: 200

Text #1

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3892. Min Act: -0.0000

Data Index: 20448 (C4 (Web Text))

Max Activating Token Index: 368

Text #2

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3800. Min Act: -0.0001

Data Index: 241296 (C4 (Web Text))

Max Activating Token Index: 983

Text #3

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3513. Min Act: -0.0001

Data Index: 981221 (C4 (Web Text))

Max Activating Token Index: 381

Text #4

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3403. Min Act: -0.0000

Data Index: 500125 (C4 (Web Text))

Max Activating Token Index: 459

Text #5

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3308. Min Act: -0.0001

Data Index: 147042 (C4 (Web Text))

Max Activating Token Index: 115

Text #6

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3240. Min Act: -0.0000

Data Index: 1086205 (C4 (Web Text))

Max Activating Token Index: 355

Text #7

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3152. Min Act: -0.0001

Data Index: 101669 (C4 (Web Text))

Max Activating Token Index: 266

Text #8

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3393. Min Act: -0.0000

Data Index: 194168 (C4 (Web Text))

Max Activating Token Index: 590

Text #9

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3241. Min Act: -0.0001

Data Index: 1346287 (C4 (Web Text))

Max Activating Token Index: 75

Text #10

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3006. Min Act: -0.0001

Data Index: 1561478 (Python Code)

Max Activating Token Index: 589

Text #11

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3150. Min Act: -0.0001

Data Index: 788400 (C4 (Web Text))

Max Activating Token Index: 439

Text #12

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3083. Min Act: -0.0000

Data Index: 1166865 (C4 (Web Text))

Max Activating Token Index: 970

Text #13

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3080. Min Act: -0.0001

Data Index: 825453 (C4 (Web Text))

Max Activating Token Index: 996

Text #14

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3159. Min Act: -0.0001

Data Index: 100700 (C4 (Web Text))

Max Activating Token Index: 901

Text #15

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3034. Min Act: -0.0001

Data Index: 1317838 (C4 (Web Text))

Max Activating Token Index: 827

Text #16

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3007. Min Act: -0.0000

Data Index: 201248 (C4 (Web Text))

Max Activating Token Index: 273

Text #17

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.3006. Min Act: -0.0001

Data Index: 653960 (C4 (Web Text))

Max Activating Token Index: 170

Text #18

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.2996. Min Act: -0.0001

Data Index: 564183 (C4 (Web Text))

Max Activating Token Index: 831

Text #19

Max Range: 0.4071. Min Range: -0.4071

Max Act: 0.2962. Min Act: -0.0001

Data Index: 1154007 (C4 (Web Text))

Max Activating Token Index: 184