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

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

Neuron 478 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 6.9531. Min Range: -6.9531

Max Act: 6.9531. Min Act: -0.0000

Data Index: 1254902 (C4 (Web Text))

Max Activating Token Index: 94

Text #1

Max Range: 6.9531. Min Range: -6.9531

Max Act: 6.5614. Min Act: -0.0000

Data Index: 296772 (C4 (Web Text))

Max Activating Token Index: 87

Text #2

Max Range: 6.9531. Min Range: -6.9531

Max Act: 6.2174. Min Act: -0.0000

Data Index: 129998 (C4 (Web Text))

Max Activating Token Index: 105

Text #3

Max Range: 6.9531. Min Range: -6.9531

Max Act: 5.8791. Min Act: -0.0000

Data Index: 1133460 (C4 (Web Text))

Max Activating Token Index: 196

Text #4

Max Range: 6.9531. Min Range: -6.9531

Max Act: 5.6599. Min Act: -0.0000

Data Index: 387566 (C4 (Web Text))

Max Activating Token Index: 438

Text #5

Max Range: 6.9531. Min Range: -6.9531

Max Act: 5.8483. Min Act: -0.0000

Data Index: 954408 (C4 (Web Text))

Max Activating Token Index: 17

Text #6

Max Range: 6.9531. Min Range: -6.9531

Max Act: 5.1906. Min Act: -0.0000

Data Index: 1478204 (Python Code)

Max Activating Token Index: 551

Text #7

Max Range: 6.9531. Min Range: -6.9531

Max Act: 5.2754. Min Act: -0.0000

Data Index: 207760 (C4 (Web Text))

Max Activating Token Index: 273

Text #8

Max Range: 6.9531. Min Range: -6.9531

Max Act: 5.0446. Min Act: -0.0000

Data Index: 20172 (C4 (Web Text))

Max Activating Token Index: 225

Text #9

Max Range: 6.9531. Min Range: -6.9531

Max Act: 5.0857. Min Act: -0.0000

Data Index: 720251 (C4 (Web Text))

Max Activating Token Index: 779

Text #10

Max Range: 6.9531. Min Range: -6.9531

Max Act: 4.5370. Min Act: -0.0000

Data Index: 81622 (C4 (Web Text))

Max Activating Token Index: 828

Text #11

Max Range: 6.9531. Min Range: -6.9531

Max Act: 4.8208. Min Act: -0.0000

Data Index: 1250863 (C4 (Web Text))

Max Activating Token Index: 283

Text #12

Max Range: 6.9531. Min Range: -6.9531

Max Act: 4.8966. Min Act: -0.0000

Data Index: 300588 (C4 (Web Text))

Max Activating Token Index: 332

Text #13

Max Range: 6.9531. Min Range: -6.9531

Max Act: 4.4232. Min Act: -0.0000

Data Index: 481710 (C4 (Web Text))

Max Activating Token Index: 200

Text #14

Max Range: 6.9531. Min Range: -6.9531

Max Act: 4.4390. Min Act: -0.0000

Data Index: 540268 (C4 (Web Text))

Max Activating Token Index: 743

Text #15

Max Range: 6.9531. Min Range: -6.9531

Max Act: 4.4428. Min Act: -0.0000

Data Index: 16503 (C4 (Web Text))

Max Activating Token Index: 42

Text #16

Max Range: 6.9531. Min Range: -6.9531

Max Act: 4.1008. Min Act: -0.0000

Data Index: 23602 (C4 (Web Text))

Max Activating Token Index: 210

Text #17

Max Range: 6.9531. Min Range: -6.9531

Max Act: 4.0599. Min Act: -0.0000

Data Index: 1231791 (C4 (Web Text))

Max Activating Token Index: 760

Text #18

Max Range: 6.9531. Min Range: -6.9531

Max Act: 3.6013. Min Act: -0.0000

Data Index: 871077 (C4 (Web Text))

Max Activating Token Index: 680

Text #19

Max Range: 6.9531. Min Range: -6.9531

Max Act: 4.3096. Min Act: -0.0000

Data Index: 1047928 (C4 (Web Text))

Max Activating Token Index: 123