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

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

Neuron 3477 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.8444. Min Range: -4.8444

Max Act: 4.8444. Min Act: -0.0000

Data Index: 219657 (C4 (Web Text))

Max Activating Token Index: 150

Text #1

Max Range: 4.8444. Min Range: -4.8444

Max Act: 4.1723. Min Act: -0.0000

Data Index: 999817 (C4 (Web Text))

Max Activating Token Index: 106

Text #2

Max Range: 4.8444. Min Range: -4.8444

Max Act: 4.5249. Min Act: -0.0000

Data Index: 645041 (C4 (Web Text))

Max Activating Token Index: 193

Text #3

Max Range: 4.8444. Min Range: -4.8444

Max Act: 4.4112. Min Act: -0.0000

Data Index: 1328905 (C4 (Web Text))

Max Activating Token Index: 103

Text #4

Max Range: 4.8444. Min Range: -4.8444

Max Act: 4.2383. Min Act: -0.0000

Data Index: 1070620 (C4 (Web Text))

Max Activating Token Index: 784

Text #5

Max Range: 4.8444. Min Range: -4.8444

Max Act: 4.0716. Min Act: -0.0000

Data Index: 1225398 (C4 (Web Text))

Max Activating Token Index: 252

Text #6

Max Range: 4.8444. Min Range: -4.8444

Max Act: 4.2286. Min Act: -0.0000

Data Index: 1068274 (C4 (Web Text))

Max Activating Token Index: 372

Text #7

Max Range: 4.8444. Min Range: -4.8444

Max Act: 4.3705. Min Act: -0.0000

Data Index: 1256504 (C4 (Web Text))

Max Activating Token Index: 335

Text #8

Max Range: 4.8444. Min Range: -4.8444

Max Act: 3.7642. Min Act: -0.0000

Data Index: 88229 (C4 (Web Text))

Max Activating Token Index: 719

Text #9

Max Range: 4.8444. Min Range: -4.8444

Max Act: 3.9202. Min Act: -0.0000

Data Index: 94071 (C4 (Web Text))

Max Activating Token Index: 391

Text #10

Max Range: 4.8444. Min Range: -4.8444

Max Act: 3.7976. Min Act: -0.0000

Data Index: 750319 (C4 (Web Text))

Max Activating Token Index: 275

Text #11

Max Range: 4.8444. Min Range: -4.8444

Max Act: 4.0859. Min Act: -0.0000

Data Index: 809944 (C4 (Web Text))

Max Activating Token Index: 281

Text #12

Max Range: 4.8444. Min Range: -4.8444

Max Act: 4.3248. Min Act: -0.0000

Data Index: 1255011 (C4 (Web Text))

Max Activating Token Index: 840

Text #13

Max Range: 4.8444. Min Range: -4.8444

Max Act: 3.9445. Min Act: -0.0000

Data Index: 660232 (C4 (Web Text))

Max Activating Token Index: 800

Text #14

Max Range: 4.8444. Min Range: -4.8444

Max Act: 3.5781. Min Act: -0.0000

Data Index: 808019 (C4 (Web Text))

Max Activating Token Index: 222

Text #15

Max Range: 4.8444. Min Range: -4.8444

Max Act: 3.2807. Min Act: -0.0000

Data Index: 786260 (C4 (Web Text))

Max Activating Token Index: 217

Text #16

Max Range: 4.8444. Min Range: -4.8444

Max Act: 3.8194. Min Act: -0.0000

Data Index: 1281561 (C4 (Web Text))

Max Activating Token Index: 346

Text #17

Max Range: 4.8444. Min Range: -4.8444

Max Act: 3.5758. Min Act: -0.0000

Data Index: 291785 (C4 (Web Text))

Max Activating Token Index: 731

Text #18

Max Range: 4.8444. Min Range: -4.8444

Max Act: 3.9167. Min Act: -0.0000

Data Index: 1188620 (C4 (Web Text))

Max Activating Token Index: 356

Text #19

Max Range: 4.8444. Min Range: -4.8444

Max Act: 3.6710. Min Act: -0.0000

Data Index: 812357 (C4 (Web Text))

Max Activating Token Index: 571