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

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

Neuron 3836 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.8695. Min Act: -0.0000

Data Index: 245848 (C4 (Web Text))

Max Activating Token Index: 810

Text #1

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.7257. Min Act: -0.0000

Data Index: 182969 (C4 (Web Text))

Max Activating Token Index: 1009

Text #2

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.3260. Min Act: -0.0000

Data Index: 1113069 (C4 (Web Text))

Max Activating Token Index: 302

Text #3

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.3813. Min Act: -0.0000

Data Index: 350479 (C4 (Web Text))

Max Activating Token Index: 19

Text #4

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.3295. Min Act: -0.0000

Data Index: 79225 (C4 (Web Text))

Max Activating Token Index: 313

Text #5

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.4705. Min Act: -0.0000

Data Index: 815067 (C4 (Web Text))

Max Activating Token Index: 131

Text #6

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.3090. Min Act: -0.0000

Data Index: 1405829 (Python Code)

Max Activating Token Index: 670

Text #7

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.1142. Min Act: -0.0000

Data Index: 351338 (C4 (Web Text))

Max Activating Token Index: 6

Text #8

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.1067. Min Act: -0.0000

Data Index: 865505 (C4 (Web Text))

Max Activating Token Index: 3

Text #9

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.2361. Min Act: -0.0000

Data Index: 796825 (C4 (Web Text))

Max Activating Token Index: 771

Text #10

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.2354. Min Act: -0.0000

Data Index: 746133 (C4 (Web Text))

Max Activating Token Index: 910

Text #11

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.1978. Min Act: -0.0000

Data Index: 200075 (C4 (Web Text))

Max Activating Token Index: 32

Text #12

Max Range: 3.8695. Min Range: -3.8695

Max Act: 2.9982. Min Act: -0.0000

Data Index: 1214383 (C4 (Web Text))

Max Activating Token Index: 613

Text #13

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.0412. Min Act: -0.0000

Data Index: 2435 (C4 (Web Text))

Max Activating Token Index: 131

Text #14

Max Range: 3.8695. Min Range: -3.8695

Max Act: 3.0077. Min Act: -0.0000

Data Index: 774512 (C4 (Web Text))

Max Activating Token Index: 868

Text #15

Max Range: 3.8695. Min Range: -3.8695

Max Act: 2.9958. Min Act: -0.0000

Data Index: 604747 (C4 (Web Text))

Max Activating Token Index: 806

Text #16

Max Range: 3.8695. Min Range: -3.8695

Max Act: 2.8281. Min Act: -0.0000

Data Index: 1318390 (C4 (Web Text))

Max Activating Token Index: 556

Text #17

Max Range: 3.8695. Min Range: -3.8695

Max Act: 2.9601. Min Act: -0.0000

Data Index: 1345704 (C4 (Web Text))

Max Activating Token Index: 40

Text #18

Max Range: 3.8695. Min Range: -3.8695

Max Act: 2.9516. Min Act: -0.0000

Data Index: 956764 (C4 (Web Text))

Max Activating Token Index: 851

Text #19

Max Range: 3.8695. Min Range: -3.8695

Max Act: 2.9402. Min Act: -0.0000

Data Index: 296509 (C4 (Web Text))

Max Activating Token Index: 2