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

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

Neuron 3087 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.9856. Min Act: -0.0002

Data Index: 1274994 (C4 (Web Text))

Max Activating Token Index: 62

Text #1

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.8121. Min Act: -0.0002

Data Index: 499444 (C4 (Web Text))

Max Activating Token Index: 656

Text #2

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.9812. Min Act: -0.0002

Data Index: 1244708 (C4 (Web Text))

Max Activating Token Index: 995

Text #3

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.9390. Min Act: -0.0002

Data Index: 975512 (C4 (Web Text))

Max Activating Token Index: 6

Text #4

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.6888. Min Act: -0.0002

Data Index: 233960 (C4 (Web Text))

Max Activating Token Index: 219

Text #5

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.6349. Min Act: -0.0002

Data Index: 1054924 (C4 (Web Text))

Max Activating Token Index: 796

Text #6

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.8629. Min Act: -0.0002

Data Index: 1142895 (C4 (Web Text))

Max Activating Token Index: 275

Text #7

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.7360. Min Act: -0.0002

Data Index: 459712 (C4 (Web Text))

Max Activating Token Index: 539

Text #8

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.7245. Min Act: -0.0002

Data Index: 354508 (C4 (Web Text))

Max Activating Token Index: 725

Text #9

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.7220. Min Act: -0.0002

Data Index: 421704 (C4 (Web Text))

Max Activating Token Index: 247

Text #10

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.8202. Min Act: -0.0002

Data Index: 634175 (C4 (Web Text))

Max Activating Token Index: 366

Text #11

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.7158. Min Act: -0.0002

Data Index: 32198 (C4 (Web Text))

Max Activating Token Index: 89

Text #12

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.8263. Min Act: -0.0002

Data Index: 248698 (C4 (Web Text))

Max Activating Token Index: 97

Text #13

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.6912. Min Act: -0.0002

Data Index: 495912 (C4 (Web Text))

Max Activating Token Index: 220

Text #14

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.7310. Min Act: -0.0002

Data Index: 632109 (C4 (Web Text))

Max Activating Token Index: 323

Text #15

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.8851. Min Act: -0.0002

Data Index: 147986 (C4 (Web Text))

Max Activating Token Index: 262

Text #16

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.7314. Min Act: -0.0002

Data Index: 1253291 (C4 (Web Text))

Max Activating Token Index: 393

Text #17

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.6252. Min Act: -0.0002

Data Index: 1257771 (C4 (Web Text))

Max Activating Token Index: 453

Text #18

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.6031. Min Act: -0.0002

Data Index: 1135202 (C4 (Web Text))

Max Activating Token Index: 1021

Text #19

Max Range: 1.9856. Min Range: -1.9856

Max Act: 1.6285. Min Act: -0.0002

Data Index: 1303576 (C4 (Web Text))

Max Activating Token Index: 727