Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 90 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0675. Min Act: -0.0006

Data Index: 405570 (C4 (Web Text))

Max Activating Token Index: 527

Text #1

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0682. Min Act: -0.0005

Data Index: 1294677 (C4 (Web Text))

Max Activating Token Index: 339

Text #2

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0657. Min Act: -0.0006

Data Index: 1219099 (C4 (Web Text))

Max Activating Token Index: 273

Text #3

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0623. Min Act: -0.0006

Data Index: 89582 (C4 (Web Text))

Max Activating Token Index: 611

Text #4

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0624. Min Act: -0.0006

Data Index: 73708 (C4 (Web Text))

Max Activating Token Index: 171

Text #5

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0633. Min Act: -0.0006

Data Index: 1192308 (C4 (Web Text))

Max Activating Token Index: 450

Text #6

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0620. Min Act: -0.0006

Data Index: 120421 (C4 (Web Text))

Max Activating Token Index: 423

Text #7

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0653. Min Act: -0.0006

Data Index: 504327 (C4 (Web Text))

Max Activating Token Index: 889

Text #8

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0633. Min Act: -0.0006

Data Index: 719430 (C4 (Web Text))

Max Activating Token Index: 82

Text #9

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0630. Min Act: -0.0006

Data Index: 609812 (C4 (Web Text))

Max Activating Token Index: 34

Text #10

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0636. Min Act: -0.0006

Data Index: 587869 (C4 (Web Text))

Max Activating Token Index: 681

Text #11

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0615. Min Act: -0.0006

Data Index: 1300544 (C4 (Web Text))

Max Activating Token Index: 585

Text #12

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0589. Min Act: -0.0006

Data Index: 186644 (C4 (Web Text))

Max Activating Token Index: 920

Text #13

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0573. Min Act: -0.0006

Data Index: 818157 (C4 (Web Text))

Max Activating Token Index: 64

Text #14

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0592. Min Act: -0.0006

Data Index: 1321719 (C4 (Web Text))

Max Activating Token Index: 37

Text #15

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0603. Min Act: -0.0006

Data Index: 164613 (C4 (Web Text))

Max Activating Token Index: 496

Text #16

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0600. Min Act: -0.0006

Data Index: 860102 (C4 (Web Text))

Max Activating Token Index: 1022

Text #17

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0596. Min Act: -0.0006

Data Index: 214087 (C4 (Web Text))

Max Activating Token Index: 306

Text #18

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0555. Min Act: -0.0006

Data Index: 1287803 (C4 (Web Text))

Max Activating Token Index: 715

Text #19

Max Range: 0.0682. Min Range: -0.0682

Max Act: 0.0554. Min Act: -0.0006

Data Index: 409456 (C4 (Web Text))

Max Activating Token Index: 668