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

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

Neuron 1714 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: 0.0789. Min Range: -0.0789

Max Act: 0.0766. Min Act: -0.0001

Data Index: 817625 (C4 (Web Text))

Max Activating Token Index: 162

Text #1

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0789. Min Act: -0.0001

Data Index: 502994 (C4 (Web Text))

Max Activating Token Index: 162

Text #2

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0685. Min Act: -0.0001

Data Index: 856617 (C4 (Web Text))

Max Activating Token Index: 231

Text #3

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0611. Min Act: -0.0002

Data Index: 89955 (C4 (Web Text))

Max Activating Token Index: 349

Text #4

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0539. Min Act: -0.0002

Data Index: 671873 (C4 (Web Text))

Max Activating Token Index: 915

Text #5

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0551. Min Act: -0.0001

Data Index: 557398 (C4 (Web Text))

Max Activating Token Index: 446

Text #6

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0561. Min Act: -0.0001

Data Index: 1042745 (C4 (Web Text))

Max Activating Token Index: 1011

Text #7

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0521. Min Act: -0.0001

Data Index: 1246761 (C4 (Web Text))

Max Activating Token Index: 184

Text #8

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0567. Min Act: -0.0002

Data Index: 568407 (C4 (Web Text))

Max Activating Token Index: 88

Text #9

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0558. Min Act: -0.0002

Data Index: 360677 (C4 (Web Text))

Max Activating Token Index: 556

Text #10

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0510. Min Act: -0.0002

Data Index: 257098 (C4 (Web Text))

Max Activating Token Index: 144

Text #11

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0485. Min Act: -0.0002

Data Index: 1140925 (C4 (Web Text))

Max Activating Token Index: 203

Text #12

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0524. Min Act: -0.0002

Data Index: 1353229 (C4 (Web Text))

Max Activating Token Index: 925

Text #13

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0549. Min Act: -0.0002

Data Index: 806705 (C4 (Web Text))

Max Activating Token Index: 828

Text #14

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0552. Min Act: -0.0002

Data Index: 1082933 (C4 (Web Text))

Max Activating Token Index: 928

Text #15

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0523. Min Act: -0.0002

Data Index: 172428 (C4 (Web Text))

Max Activating Token Index: 212

Text #16

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0528. Min Act: -0.0002

Data Index: 225331 (C4 (Web Text))

Max Activating Token Index: 259

Text #17

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0518. Min Act: -0.0002

Data Index: 1084290 (C4 (Web Text))

Max Activating Token Index: 190

Text #18

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0501. Min Act: -0.0002

Data Index: 102282 (C4 (Web Text))

Max Activating Token Index: 584

Text #19

Max Range: 0.0789. Min Range: -0.0789

Max Act: 0.0512. Min Act: -0.0002

Data Index: 1180294 (C4 (Web Text))

Max Activating Token Index: 600