Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 129 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0759. Min Act: -0.0002

Data Index: 1646835 (Python Code)

Max Activating Token Index: 45

Text #1

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0650. Min Act: -0.0002

Data Index: 338833 (C4 (Web Text))

Max Activating Token Index: 708

Text #2

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0640. Min Act: -0.0002

Data Index: 926796 (C4 (Web Text))

Max Activating Token Index: 131

Text #3

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0638. Min Act: -0.0002

Data Index: 1614216 (Python Code)

Max Activating Token Index: 422

Text #4

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0597. Min Act: -0.0002

Data Index: 916747 (C4 (Web Text))

Max Activating Token Index: 358

Text #5

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0601. Min Act: -0.0002

Data Index: 509644 (C4 (Web Text))

Max Activating Token Index: 294

Text #6

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0586. Min Act: -0.0000

Data Index: 1649360 (Python Code)

Max Activating Token Index: 21

Text #7

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0607. Min Act: -0.0002

Data Index: 483382 (C4 (Web Text))

Max Activating Token Index: 180

Text #8

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0598. Min Act: -0.0002

Data Index: 837019 (C4 (Web Text))

Max Activating Token Index: 521

Text #9

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0573. Min Act: -0.0002

Data Index: 232220 (C4 (Web Text))

Max Activating Token Index: 708

Text #10

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0576. Min Act: -0.0001

Data Index: 1295598 (C4 (Web Text))

Max Activating Token Index: 378

Text #11

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0583. Min Act: -0.0002

Data Index: 665830 (C4 (Web Text))

Max Activating Token Index: 241

Text #12

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0581. Min Act: -0.0001

Data Index: 1005619 (C4 (Web Text))

Max Activating Token Index: 213

Text #13

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0580. Min Act: -0.0002

Data Index: 966384 (C4 (Web Text))

Max Activating Token Index: 790

Text #14

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0567. Min Act: -0.0002

Data Index: 303426 (C4 (Web Text))

Max Activating Token Index: 64

Text #15

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0600. Min Act: -0.0002

Data Index: 987877 (C4 (Web Text))

Max Activating Token Index: 769

Text #16

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0564. Min Act: -0.0002

Data Index: 89529 (C4 (Web Text))

Max Activating Token Index: 123

Text #17

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0552. Min Act: -0.0002

Data Index: 1142541 (C4 (Web Text))

Max Activating Token Index: 661

Text #18

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0550. Min Act: -0.0002

Data Index: 1346081 (C4 (Web Text))

Max Activating Token Index: 252

Text #19

Max Range: 0.0759. Min Range: -0.0759

Max Act: 0.0552. Min Act: -0.0001

Data Index: 1463226 (Python Code)

Max Activating Token Index: 524