Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 893 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.2320. Min Act: -0.0000

Data Index: 449278 (C4 (Web Text))

Max Activating Token Index: 1018

Text #1

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.2181. Min Act: -0.0000

Data Index: 770199 (C4 (Web Text))

Max Activating Token Index: 39

Text #2

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.2019. Min Act: -0.0000

Data Index: 174995 (C4 (Web Text))

Max Activating Token Index: 37

Text #3

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1845. Min Act: -0.0000

Data Index: 702947 (C4 (Web Text))

Max Activating Token Index: 1011

Text #4

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1817. Min Act: -0.0000

Data Index: 161513 (C4 (Web Text))

Max Activating Token Index: 528

Text #5

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1888. Min Act: -0.0001

Data Index: 295797 (C4 (Web Text))

Max Activating Token Index: 63

Text #6

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1870. Min Act: -0.0000

Data Index: 740580 (C4 (Web Text))

Max Activating Token Index: 472

Text #7

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1774. Min Act: -0.0000

Data Index: 877077 (C4 (Web Text))

Max Activating Token Index: 922

Text #8

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1680. Min Act: -0.0000

Data Index: 696732 (C4 (Web Text))

Max Activating Token Index: 933

Text #9

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1685. Min Act: -0.0000

Data Index: 550542 (C4 (Web Text))

Max Activating Token Index: 587

Text #10

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1639. Min Act: -0.0001

Data Index: 1167742 (C4 (Web Text))

Max Activating Token Index: 136

Text #11

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1692. Min Act: -0.0000

Data Index: 608283 (C4 (Web Text))

Max Activating Token Index: 270

Text #12

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1675. Min Act: -0.0000

Data Index: 520869 (C4 (Web Text))

Max Activating Token Index: 720

Text #13

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1560. Min Act: -0.0001

Data Index: 1626837 (Python Code)

Max Activating Token Index: 343

Text #14

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1596. Min Act: -0.0000

Data Index: 330274 (C4 (Web Text))

Max Activating Token Index: 122

Text #15

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1607. Min Act: -0.0001

Data Index: 309483 (C4 (Web Text))

Max Activating Token Index: 842

Text #16

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1571. Min Act: -0.0000

Data Index: 73477 (C4 (Web Text))

Max Activating Token Index: 263

Text #17

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1521. Min Act: -0.0000

Data Index: 1279814 (C4 (Web Text))

Max Activating Token Index: 1003

Text #18

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1655. Min Act: -0.0000

Data Index: 1254862 (C4 (Web Text))

Max Activating Token Index: 185

Text #19

Max Range: 0.2320. Min Range: -0.2320

Max Act: 0.1599. Min Act: -0.0000

Data Index: 1149096 (C4 (Web Text))

Max Activating Token Index: 135