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

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

Neuron 543 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.2406. Min Act: -0.0001

Data Index: 1121932 (C4 (Web Text))

Max Activating Token Index: 884

Text #1

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.2000. Min Act: -0.0001

Data Index: 1372633 (Python Code)

Max Activating Token Index: 735

Text #2

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.2007. Min Act: -0.0001

Data Index: 59985 (C4 (Web Text))

Max Activating Token Index: 228

Text #3

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1950. Min Act: -0.0001

Data Index: 908679 (C4 (Web Text))

Max Activating Token Index: 77

Text #4

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1826. Min Act: -0.0001

Data Index: 886477 (C4 (Web Text))

Max Activating Token Index: 176

Text #5

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1879. Min Act: -0.0001

Data Index: 1632574 (Python Code)

Max Activating Token Index: 472

Text #6

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1734. Min Act: -0.0001

Data Index: 1207906 (C4 (Web Text))

Max Activating Token Index: 275

Text #7

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1715. Min Act: -0.0001

Data Index: 416842 (C4 (Web Text))

Max Activating Token Index: 721

Text #8

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1798. Min Act: -0.0001

Data Index: 1498855 (Python Code)

Max Activating Token Index: 39

Text #9

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1744. Min Act: -0.0001

Data Index: 1197127 (C4 (Web Text))

Max Activating Token Index: 971

Text #10

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.2023. Min Act: -0.0001

Data Index: 203156 (C4 (Web Text))

Max Activating Token Index: 446

Text #11

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1691. Min Act: -0.0001

Data Index: 132840 (C4 (Web Text))

Max Activating Token Index: 510

Text #12

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1472. Min Act: -0.0001

Data Index: 714248 (C4 (Web Text))

Max Activating Token Index: 53

Text #13

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1634. Min Act: -0.0001

Data Index: 1621016 (Python Code)

Max Activating Token Index: 34

Text #14

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1664. Min Act: -0.0001

Data Index: 119862 (C4 (Web Text))

Max Activating Token Index: 638

Text #15

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1643. Min Act: -0.0001

Data Index: 624689 (C4 (Web Text))

Max Activating Token Index: 745

Text #16

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1595. Min Act: -0.0001

Data Index: 950558 (C4 (Web Text))

Max Activating Token Index: 648

Text #17

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1563. Min Act: -0.0001

Data Index: 745160 (C4 (Web Text))

Max Activating Token Index: 451

Text #18

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1576. Min Act: -0.0001

Data Index: 1628593 (Python Code)

Max Activating Token Index: 1001

Text #19

Max Range: 0.2406. Min Range: -0.2406

Max Act: 0.1665. Min Act: -0.0001

Data Index: 14708 (C4 (Web Text))

Max Activating Token Index: 523