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

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

Neuron 4908 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3558. Min Range: -1.3558

Max Act: 1.3558. Min Act: -0.0000

Data Index: 1008963 (C4 (Web Text))

Max Activating Token Index: 458

Text #1

Max Range: 1.3558. Min Range: -1.3558

Max Act: 1.1531. Min Act: -0.0000

Data Index: 233454 (C4 (Web Text))

Max Activating Token Index: 687

Text #2

Max Range: 1.3558. Min Range: -1.3558

Max Act: 1.1128. Min Act: -0.0000

Data Index: 42294 (C4 (Web Text))

Max Activating Token Index: 197

Text #3

Max Range: 1.3558. Min Range: -1.3558

Max Act: 1.0452. Min Act: -0.0000

Data Index: 1183075 (C4 (Web Text))

Max Activating Token Index: 910

Text #4

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.9600. Min Act: -0.0000

Data Index: 1312732 (C4 (Web Text))

Max Activating Token Index: 905

Text #5

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.9974. Min Act: -0.0000

Data Index: 677834 (C4 (Web Text))

Max Activating Token Index: 95

Text #6

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.9918. Min Act: -0.0000

Data Index: 930898 (C4 (Web Text))

Max Activating Token Index: 561

Text #7

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.9364. Min Act: -0.0000

Data Index: 1029742 (C4 (Web Text))

Max Activating Token Index: 93

Text #8

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.9167. Min Act: -0.0000

Data Index: 566333 (C4 (Web Text))

Max Activating Token Index: 47

Text #9

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.8543. Min Act: -0.0000

Data Index: 907744 (C4 (Web Text))

Max Activating Token Index: 842

Text #10

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.8750. Min Act: -0.0000

Data Index: 297238 (C4 (Web Text))

Max Activating Token Index: 465

Text #11

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.8944. Min Act: -0.0000

Data Index: 965517 (C4 (Web Text))

Max Activating Token Index: 126

Text #12

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.8548. Min Act: -0.0000

Data Index: 521535 (C4 (Web Text))

Max Activating Token Index: 890

Text #13

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.8631. Min Act: -0.0000

Data Index: 1361471 (Python Code)

Max Activating Token Index: 165

Text #14

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.8303. Min Act: -0.0000

Data Index: 1642669 (Python Code)

Max Activating Token Index: 957

Text #15

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.8195. Min Act: -0.0000

Data Index: 512175 (C4 (Web Text))

Max Activating Token Index: 28

Text #16

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.7867. Min Act: -0.0000

Data Index: 323587 (C4 (Web Text))

Max Activating Token Index: 240

Text #17

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.7591. Min Act: -0.0000

Data Index: 1412583 (Python Code)

Max Activating Token Index: 725

Text #18

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.7474. Min Act: -0.0000

Data Index: 1556246 (Python Code)

Max Activating Token Index: 978

Text #19

Max Range: 1.3558. Min Range: -1.3558

Max Act: 0.7693. Min Act: -0.0000

Data Index: 1572021 (Python Code)

Max Activating Token Index: 341