Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4756 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1761. Min Act: -0.0000

Data Index: 584770 (C4 (Web Text))

Max Activating Token Index: 187

Text #1

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0742. Min Act: -0.0000

Data Index: 392815 (C4 (Web Text))

Max Activating Token Index: 246

Text #2

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0777. Min Act: -0.0000

Data Index: 322037 (C4 (Web Text))

Max Activating Token Index: 755

Text #3

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0631. Min Act: -0.0000

Data Index: 637484 (C4 (Web Text))

Max Activating Token Index: 755

Text #4

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0692. Min Act: -0.0000

Data Index: 1161199 (C4 (Web Text))

Max Activating Token Index: 955

Text #5

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0665. Min Act: -0.0000

Data Index: 658446 (C4 (Web Text))

Max Activating Token Index: 887

Text #6

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0631. Min Act: -0.0000

Data Index: 759206 (C4 (Web Text))

Max Activating Token Index: 886

Text #7

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0605. Min Act: -0.0000

Data Index: 1203477 (C4 (Web Text))

Max Activating Token Index: 854

Text #8

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0519. Min Act: -0.0000

Data Index: 463503 (C4 (Web Text))

Max Activating Token Index: 417

Text #9

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0603. Min Act: -0.0000

Data Index: 594833 (C4 (Web Text))

Max Activating Token Index: 639

Text #10

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0573. Min Act: -0.0000

Data Index: 172646 (C4 (Web Text))

Max Activating Token Index: 373

Text #11

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0567. Min Act: -0.0000

Data Index: 1182011 (C4 (Web Text))

Max Activating Token Index: 950

Text #12

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0557. Min Act: -0.0000

Data Index: 862592 (C4 (Web Text))

Max Activating Token Index: 425

Text #13

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0553. Min Act: -0.0000

Data Index: 1619355 (Python Code)

Max Activating Token Index: 448

Text #14

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0526. Min Act: -0.0000

Data Index: 809655 (C4 (Web Text))

Max Activating Token Index: 444

Text #15

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0549. Min Act: -0.0000

Data Index: 638817 (C4 (Web Text))

Max Activating Token Index: 160

Text #16

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0549. Min Act: -0.0000

Data Index: 1323177 (C4 (Web Text))

Max Activating Token Index: 869

Text #17

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0563. Min Act: -0.0000

Data Index: 426824 (C4 (Web Text))

Max Activating Token Index: 672

Text #18

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0545. Min Act: -0.0000

Data Index: 1317847 (C4 (Web Text))

Max Activating Token Index: 735

Text #19

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0509. Min Act: -0.0000

Data Index: 890558 (C4 (Web Text))

Max Activating Token Index: 849