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

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

Neuron 198 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: 0.2630. Min Range: -0.2630

Max Act: 0.2630. Min Act: -0.0002

Data Index: 687767 (C4 (Web Text))

Max Activating Token Index: 61

Text #1

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2531. Min Act: -0.0000

Data Index: 871977 (C4 (Web Text))

Max Activating Token Index: 588

Text #2

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2530. Min Act: -0.0001

Data Index: 700781 (C4 (Web Text))

Max Activating Token Index: 236

Text #3

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2425. Min Act: -0.0000

Data Index: 1314296 (C4 (Web Text))

Max Activating Token Index: 285

Text #4

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2527. Min Act: -0.0001

Data Index: 467221 (C4 (Web Text))

Max Activating Token Index: 331

Text #5

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2420. Min Act: -0.0001

Data Index: 845675 (C4 (Web Text))

Max Activating Token Index: 338

Text #6

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2381. Min Act: -0.0002

Data Index: 389449 (C4 (Web Text))

Max Activating Token Index: 103

Text #7

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2294. Min Act: -0.0001

Data Index: 1254567 (C4 (Web Text))

Max Activating Token Index: 49

Text #8

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2343. Min Act: -0.0001

Data Index: 83902 (C4 (Web Text))

Max Activating Token Index: 165

Text #9

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2304. Min Act: -0.0002

Data Index: 591460 (C4 (Web Text))

Max Activating Token Index: 101

Text #10

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2285. Min Act: -0.0002

Data Index: 193959 (C4 (Web Text))

Max Activating Token Index: 24

Text #11

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2267. Min Act: -0.0002

Data Index: 1146706 (C4 (Web Text))

Max Activating Token Index: 621

Text #12

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2178. Min Act: -0.0001

Data Index: 625173 (C4 (Web Text))

Max Activating Token Index: 360

Text #13

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2247. Min Act: -0.0001

Data Index: 1357582 (C4 (Web Text))

Max Activating Token Index: 320

Text #14

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2301. Min Act: -0.0000

Data Index: 1079652 (C4 (Web Text))

Max Activating Token Index: 217

Text #15

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2240. Min Act: -0.0002

Data Index: 652111 (C4 (Web Text))

Max Activating Token Index: 367

Text #16

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2224. Min Act: -0.0002

Data Index: 929193 (C4 (Web Text))

Max Activating Token Index: 301

Text #17

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2232. Min Act: -0.0002

Data Index: 1285494 (C4 (Web Text))

Max Activating Token Index: 606

Text #18

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2219. Min Act: -0.0002

Data Index: 781182 (C4 (Web Text))

Max Activating Token Index: 152

Text #19

Max Range: 0.2630. Min Range: -0.2630

Max Act: 0.2152. Min Act: -0.0001

Data Index: 275533 (C4 (Web Text))

Max Activating Token Index: 807