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

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

Neuron 4885 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: 0.2060. Min Range: -0.2060

Max Act: 0.2060. Min Act: -0.0001

Data Index: 13498 (C4 (Web Text))

Max Activating Token Index: 1017

Text #1

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1917. Min Act: -0.0001

Data Index: 1072382 (C4 (Web Text))

Max Activating Token Index: 876

Text #2

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1890. Min Act: -0.0001

Data Index: 1203844 (C4 (Web Text))

Max Activating Token Index: 717

Text #3

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1767. Min Act: -0.0000

Data Index: 319468 (C4 (Web Text))

Max Activating Token Index: 645

Text #4

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1753. Min Act: -0.0000

Data Index: 320358 (C4 (Web Text))

Max Activating Token Index: 754

Text #5

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1852. Min Act: -0.0001

Data Index: 187538 (C4 (Web Text))

Max Activating Token Index: 738

Text #6

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1898. Min Act: -0.0001

Data Index: 1148487 (C4 (Web Text))

Max Activating Token Index: 889

Text #7

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1889. Min Act: -0.0001

Data Index: 783785 (C4 (Web Text))

Max Activating Token Index: 288

Text #8

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1629. Min Act: -0.0001

Data Index: 674521 (C4 (Web Text))

Max Activating Token Index: 870

Text #9

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1686. Min Act: -0.0001

Data Index: 282128 (C4 (Web Text))

Max Activating Token Index: 167

Text #10

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1753. Min Act: -0.0001

Data Index: 1391709 (Python Code)

Max Activating Token Index: 824

Text #11

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1654. Min Act: -0.0001

Data Index: 800883 (C4 (Web Text))

Max Activating Token Index: 248

Text #12

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1773. Min Act: -0.0001

Data Index: 317823 (C4 (Web Text))

Max Activating Token Index: 216

Text #13

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1765. Min Act: -0.0000

Data Index: 186015 (C4 (Web Text))

Max Activating Token Index: 279

Text #14

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1729. Min Act: -0.0000

Data Index: 1106330 (C4 (Web Text))

Max Activating Token Index: 260

Text #15

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1576. Min Act: -0.0001

Data Index: 1032731 (C4 (Web Text))

Max Activating Token Index: 167

Text #16

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1583. Min Act: -0.0001

Data Index: 154330 (C4 (Web Text))

Max Activating Token Index: 826

Text #17

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1576. Min Act: -0.0001

Data Index: 1418292 (Python Code)

Max Activating Token Index: 167

Text #18

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1713. Min Act: -0.0001

Data Index: 1025855 (C4 (Web Text))

Max Activating Token Index: 693

Text #19

Max Range: 0.2060. Min Range: -0.2060

Max Act: 0.1586. Min Act: -0.0001

Data Index: 896376 (C4 (Web Text))

Max Activating Token Index: 751