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

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

Neuron 605 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1885. Min Range: -5.1885

Max Act: 5.1885. Min Act: -0.0001

Data Index: 1134070 (C4 (Web Text))

Max Activating Token Index: 816

Text #1

Max Range: 5.1885. Min Range: -5.1885

Max Act: 4.6216. Min Act: -0.0001

Data Index: 493871 (C4 (Web Text))

Max Activating Token Index: 734

Text #2

Max Range: 5.1885. Min Range: -5.1885

Max Act: 4.2237. Min Act: -0.0001

Data Index: 1071455 (C4 (Web Text))

Max Activating Token Index: 615

Text #3

Max Range: 5.1885. Min Range: -5.1885

Max Act: 4.5951. Min Act: -0.0001

Data Index: 1036106 (C4 (Web Text))

Max Activating Token Index: 975

Text #4

Max Range: 5.1885. Min Range: -5.1885

Max Act: 4.2087. Min Act: -0.0001

Data Index: 574438 (C4 (Web Text))

Max Activating Token Index: 335

Text #5

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.7682. Min Act: -0.0001

Data Index: 330901 (C4 (Web Text))

Max Activating Token Index: 414

Text #6

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.6354. Min Act: -0.0001

Data Index: 729941 (C4 (Web Text))

Max Activating Token Index: 190

Text #7

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.9208. Min Act: -0.0001

Data Index: 547029 (C4 (Web Text))

Max Activating Token Index: 930

Text #8

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.8895. Min Act: -0.0001

Data Index: 477340 (C4 (Web Text))

Max Activating Token Index: 77

Text #9

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.8781. Min Act: -0.0001

Data Index: 525766 (C4 (Web Text))

Max Activating Token Index: 463

Text #10

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.9656. Min Act: -0.0001

Data Index: 1254712 (C4 (Web Text))

Max Activating Token Index: 813

Text #11

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.6364. Min Act: -0.0001

Data Index: 366786 (C4 (Web Text))

Max Activating Token Index: 452

Text #12

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.8937. Min Act: -0.0001

Data Index: 640288 (C4 (Web Text))

Max Activating Token Index: 1014

Text #13

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.6010. Min Act: -0.0001

Data Index: 774954 (C4 (Web Text))

Max Activating Token Index: 49

Text #14

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.3092. Min Act: -0.0001

Data Index: 139275 (C4 (Web Text))

Max Activating Token Index: 190

Text #15

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.5433. Min Act: -0.0001

Data Index: 220751 (C4 (Web Text))

Max Activating Token Index: 610

Text #16

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.4906. Min Act: -0.0001

Data Index: 1210439 (C4 (Web Text))

Max Activating Token Index: 376

Text #17

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.3652. Min Act: -0.0001

Data Index: 778974 (C4 (Web Text))

Max Activating Token Index: 242

Text #18

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.4900. Min Act: -0.0001

Data Index: 477952 (C4 (Web Text))

Max Activating Token Index: 210

Text #19

Max Range: 5.1885. Min Range: -5.1885

Max Act: 3.4291. Min Act: -0.0001

Data Index: 531087 (C4 (Web Text))

Max Activating Token Index: 130