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

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

Neuron 364 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0742. Min Act: -0.0002

Data Index: 1632076 (Python Code)

Max Activating Token Index: 814

Text #1

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0699. Min Act: -0.0002

Data Index: 776909 (C4 (Web Text))

Max Activating Token Index: 158

Text #2

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0674. Min Act: -0.0002

Data Index: 246118 (C4 (Web Text))

Max Activating Token Index: 662

Text #3

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0594. Min Act: -0.0002

Data Index: 825576 (C4 (Web Text))

Max Activating Token Index: 872

Text #4

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0586. Min Act: -0.0002

Data Index: 278002 (C4 (Web Text))

Max Activating Token Index: 764

Text #5

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0614. Min Act: -0.0002

Data Index: 120256 (C4 (Web Text))

Max Activating Token Index: 505

Text #6

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0578. Min Act: -0.0002

Data Index: 212989 (C4 (Web Text))

Max Activating Token Index: 150

Text #7

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0579. Min Act: -0.0002

Data Index: 587531 (C4 (Web Text))

Max Activating Token Index: 42

Text #8

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0579. Min Act: -0.0002

Data Index: 614992 (C4 (Web Text))

Max Activating Token Index: 830

Text #9

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0566. Min Act: -0.0002

Data Index: 522907 (C4 (Web Text))

Max Activating Token Index: 207

Text #10

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0582. Min Act: -0.0002

Data Index: 229282 (C4 (Web Text))

Max Activating Token Index: 773

Text #11

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0558. Min Act: -0.0002

Data Index: 1633822 (Python Code)

Max Activating Token Index: 979

Text #12

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0554. Min Act: -0.0002

Data Index: 1354796 (C4 (Web Text))

Max Activating Token Index: 673

Text #13

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0553. Min Act: -0.0002

Data Index: 408714 (C4 (Web Text))

Max Activating Token Index: 23

Text #14

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0575. Min Act: -0.0002

Data Index: 1586664 (Python Code)

Max Activating Token Index: 43

Text #15

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0549. Min Act: -0.0002

Data Index: 1600764 (Python Code)

Max Activating Token Index: 592

Text #16

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0538. Min Act: -0.0002

Data Index: 1326538 (C4 (Web Text))

Max Activating Token Index: 86

Text #17

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0534. Min Act: -0.0002

Data Index: 739574 (C4 (Web Text))

Max Activating Token Index: 190

Text #18

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0537. Min Act: -0.0002

Data Index: 1364760 (Python Code)

Max Activating Token Index: 87

Text #19

Max Range: 0.0742. Min Range: -0.0742

Max Act: 0.0513. Min Act: -0.0002

Data Index: 918581 (C4 (Web Text))

Max Activating Token Index: 149