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

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

Neuron 3136 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.7578. Min Range: -5.7578

Max Act: 5.7578. Min Act: -0.0000

Data Index: 633860 (C4 (Web Text))

Max Activating Token Index: 857

Text #1

Max Range: 5.7578. Min Range: -5.7578

Max Act: 5.4377. Min Act: -0.0000

Data Index: 68311 (C4 (Web Text))

Max Activating Token Index: 377

Text #2

Max Range: 5.7578. Min Range: -5.7578

Max Act: 5.3160. Min Act: -0.0000

Data Index: 401269 (C4 (Web Text))

Max Activating Token Index: 125

Text #3

Max Range: 5.7578. Min Range: -5.7578

Max Act: 5.2559. Min Act: -0.0000

Data Index: 1033425 (C4 (Web Text))

Max Activating Token Index: 580

Text #4

Max Range: 5.7578. Min Range: -5.7578

Max Act: 5.1905. Min Act: -0.0000

Data Index: 903158 (C4 (Web Text))

Max Activating Token Index: 696

Text #5

Max Range: 5.7578. Min Range: -5.7578

Max Act: 5.1667. Min Act: -0.0000

Data Index: 40412 (C4 (Web Text))

Max Activating Token Index: 28

Text #6

Max Range: 5.7578. Min Range: -5.7578

Max Act: 5.1118. Min Act: -0.0000

Data Index: 609719 (C4 (Web Text))

Max Activating Token Index: 937

Text #7

Max Range: 5.7578. Min Range: -5.7578

Max Act: 5.0824. Min Act: -0.0000

Data Index: 303288 (C4 (Web Text))

Max Activating Token Index: 785

Text #8

Max Range: 5.7578. Min Range: -5.7578

Max Act: 5.0553. Min Act: -0.0000

Data Index: 904247 (C4 (Web Text))

Max Activating Token Index: 31

Text #9

Max Range: 5.7578. Min Range: -5.7578

Max Act: 5.1194. Min Act: -0.0000

Data Index: 850398 (C4 (Web Text))

Max Activating Token Index: 74

Text #10

Max Range: 5.7578. Min Range: -5.7578

Max Act: 4.8962. Min Act: -0.0000

Data Index: 577511 (C4 (Web Text))

Max Activating Token Index: 165

Text #11

Max Range: 5.7578. Min Range: -5.7578

Max Act: 4.8448. Min Act: -0.0000

Data Index: 489333 (C4 (Web Text))

Max Activating Token Index: 808

Text #12

Max Range: 5.7578. Min Range: -5.7578

Max Act: 4.7006. Min Act: -0.0000

Data Index: 354968 (C4 (Web Text))

Max Activating Token Index: 151

Text #13

Max Range: 5.7578. Min Range: -5.7578

Max Act: 4.5094. Min Act: -0.0000

Data Index: 532798 (C4 (Web Text))

Max Activating Token Index: 152

Text #14

Max Range: 5.7578. Min Range: -5.7578

Max Act: 4.6758. Min Act: -0.0000

Data Index: 916106 (C4 (Web Text))

Max Activating Token Index: 909

Text #15

Max Range: 5.7578. Min Range: -5.7578

Max Act: 4.6518. Min Act: -0.0000

Data Index: 1224035 (C4 (Web Text))

Max Activating Token Index: 267

Text #16

Max Range: 5.7578. Min Range: -5.7578

Max Act: 4.5559. Min Act: -0.0000

Data Index: 389092 (C4 (Web Text))

Max Activating Token Index: 586

Text #17

Max Range: 5.7578. Min Range: -5.7578

Max Act: 4.3192. Min Act: -0.0000

Data Index: 1601922 (Python Code)

Max Activating Token Index: 823

Text #18

Max Range: 5.7578. Min Range: -5.7578

Max Act: 4.5152. Min Act: -0.0000

Data Index: 1098819 (C4 (Web Text))

Max Activating Token Index: 912

Text #19

Max Range: 5.7578. Min Range: -5.7578

Max Act: 4.4371. Min Act: -0.0000

Data Index: 729417 (C4 (Web Text))

Max Activating Token Index: 669