Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4630 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.2438. Min Range: -1.2438

Max Act: 1.2438. Min Act: -0.0001

Data Index: 1098739 (C4 (Web Text))

Max Activating Token Index: 123

Text #1

Max Range: 1.2438. Min Range: -1.2438

Max Act: 1.0200. Min Act: -0.0001

Data Index: 760053 (C4 (Web Text))

Max Activating Token Index: 563

Text #2

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.9643. Min Act: -0.0001

Data Index: 720599 (C4 (Web Text))

Max Activating Token Index: 649

Text #3

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.8452. Min Act: -0.0001

Data Index: 937754 (C4 (Web Text))

Max Activating Token Index: 422

Text #4

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.8292. Min Act: -0.0001

Data Index: 855024 (C4 (Web Text))

Max Activating Token Index: 102

Text #5

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.7710. Min Act: -0.0001

Data Index: 1195769 (C4 (Web Text))

Max Activating Token Index: 363

Text #6

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.8278. Min Act: -0.0001

Data Index: 46798 (C4 (Web Text))

Max Activating Token Index: 209

Text #7

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.7182. Min Act: -0.0001

Data Index: 1059496 (C4 (Web Text))

Max Activating Token Index: 430

Text #8

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.6485. Min Act: -0.0001

Data Index: 270631 (C4 (Web Text))

Max Activating Token Index: 162

Text #9

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.7491. Min Act: -0.0001

Data Index: 1153252 (C4 (Web Text))

Max Activating Token Index: 833

Text #10

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.7204. Min Act: -0.0001

Data Index: 237823 (C4 (Web Text))

Max Activating Token Index: 885

Text #11

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.7383. Min Act: -0.0001

Data Index: 516138 (C4 (Web Text))

Max Activating Token Index: 665

Text #12

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.7634. Min Act: -0.0001

Data Index: 996517 (C4 (Web Text))

Max Activating Token Index: 23

Text #13

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.7161. Min Act: -0.0001

Data Index: 435274 (C4 (Web Text))

Max Activating Token Index: 352

Text #14

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.6838. Min Act: -0.0001

Data Index: 275548 (C4 (Web Text))

Max Activating Token Index: 226

Text #15

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.6625. Min Act: -0.0001

Data Index: 6289 (C4 (Web Text))

Max Activating Token Index: 711

Text #16

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.6235. Min Act: -0.0001

Data Index: 991903 (C4 (Web Text))

Max Activating Token Index: 704

Text #17

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.6066. Min Act: -0.0001

Data Index: 329622 (C4 (Web Text))

Max Activating Token Index: 157

Text #18

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.6296. Min Act: -0.0001

Data Index: 975835 (C4 (Web Text))

Max Activating Token Index: 163

Text #19

Max Range: 1.2438. Min Range: -1.2438

Max Act: 0.5855. Min Act: -0.0001

Data Index: 287360 (C4 (Web Text))

Max Activating Token Index: 129