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

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

Neuron 1427 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: 1.7542. Min Range: -1.7542

Max Act: 1.7306. Min Act: -0.0001

Data Index: 1506756 (Python Code)

Max Activating Token Index: 32

Text #1

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.7098. Min Act: -0.0001

Data Index: 182786 (C4 (Web Text))

Max Activating Token Index: 556

Text #2

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.5942. Min Act: -0.0001

Data Index: 854128 (C4 (Web Text))

Max Activating Token Index: 593

Text #3

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.6609. Min Act: -0.0001

Data Index: 584134 (C4 (Web Text))

Max Activating Token Index: 844

Text #4

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.5702. Min Act: -0.0001

Data Index: 1420895 (Python Code)

Max Activating Token Index: 771

Text #5

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.7542. Min Act: -0.0001

Data Index: 1110189 (C4 (Web Text))

Max Activating Token Index: 689

Text #6

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.5572. Min Act: -0.0000

Data Index: 317065 (C4 (Web Text))

Max Activating Token Index: 191

Text #7

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.3750. Min Act: -0.0001

Data Index: 713326 (C4 (Web Text))

Max Activating Token Index: 727

Text #8

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.5074. Min Act: -0.0000

Data Index: 1406066 (Python Code)

Max Activating Token Index: 116

Text #9

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.3851. Min Act: -0.0001

Data Index: 756587 (C4 (Web Text))

Max Activating Token Index: 869

Text #10

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.6251. Min Act: -0.0001

Data Index: 235795 (C4 (Web Text))

Max Activating Token Index: 879

Text #11

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.4643. Min Act: -0.0001

Data Index: 311609 (C4 (Web Text))

Max Activating Token Index: 290

Text #12

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.3847. Min Act: -0.0001

Data Index: 426565 (C4 (Web Text))

Max Activating Token Index: 695

Text #13

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.4521. Min Act: -0.0001

Data Index: 1531642 (Python Code)

Max Activating Token Index: 150

Text #14

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.4079. Min Act: -0.0001

Data Index: 1595987 (Python Code)

Max Activating Token Index: 587

Text #15

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.4868. Min Act: -0.0001

Data Index: 1378185 (Python Code)

Max Activating Token Index: 939

Text #16

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.4719. Min Act: -0.0001

Data Index: 1486705 (Python Code)

Max Activating Token Index: 617

Text #17

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.3991. Min Act: -0.0001

Data Index: 1088736 (C4 (Web Text))

Max Activating Token Index: 842

Text #18

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.3868. Min Act: -0.0001

Data Index: 1171601 (C4 (Web Text))

Max Activating Token Index: 602

Text #19

Max Range: 1.7542. Min Range: -1.7542

Max Act: 1.3445. Min Act: -0.0001

Data Index: 1082175 (C4 (Web Text))

Max Activating Token Index: 687