Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 703 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0302. Min Act: -0.0006

Data Index: 14171 (C4 (Web Text))

Max Activating Token Index: 754

Text #1

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0282. Min Act: -0.0006

Data Index: 445755 (C4 (Web Text))

Max Activating Token Index: 978

Text #2

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0264. Min Act: -0.0006

Data Index: 511056 (C4 (Web Text))

Max Activating Token Index: 514

Text #3

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0259. Min Act: -0.0006

Data Index: 984833 (C4 (Web Text))

Max Activating Token Index: 7

Text #4

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0252. Min Act: -0.0006

Data Index: 109693 (C4 (Web Text))

Max Activating Token Index: 620

Text #5

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0261. Min Act: -0.0006

Data Index: 875730 (C4 (Web Text))

Max Activating Token Index: 892

Text #6

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0245. Min Act: -0.0006

Data Index: 288277 (C4 (Web Text))

Max Activating Token Index: 691

Text #7

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0245. Min Act: -0.0006

Data Index: 651967 (C4 (Web Text))

Max Activating Token Index: 146

Text #8

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0248. Min Act: -0.0006

Data Index: 342588 (C4 (Web Text))

Max Activating Token Index: 795

Text #9

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0230. Min Act: -0.0006

Data Index: 70432 (C4 (Web Text))

Max Activating Token Index: 1021

Text #10

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0251. Min Act: -0.0006

Data Index: 621918 (C4 (Web Text))

Max Activating Token Index: 1004

Text #11

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0239. Min Act: -0.0006

Data Index: 346835 (C4 (Web Text))

Max Activating Token Index: 575

Text #12

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0228. Min Act: -0.0006

Data Index: 778204 (C4 (Web Text))

Max Activating Token Index: 212

Text #13

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0231. Min Act: -0.0006

Data Index: 589040 (C4 (Web Text))

Max Activating Token Index: 598

Text #14

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0251. Min Act: -0.0006

Data Index: 544581 (C4 (Web Text))

Max Activating Token Index: 846

Text #15

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0227. Min Act: -0.0006

Data Index: 789092 (C4 (Web Text))

Max Activating Token Index: 614

Text #16

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0221. Min Act: -0.0006

Data Index: 1343372 (C4 (Web Text))

Max Activating Token Index: 223

Text #17

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0229. Min Act: -0.0006

Data Index: 313851 (C4 (Web Text))

Max Activating Token Index: 99

Text #18

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0214. Min Act: -0.0005

Data Index: 1354219 (C4 (Web Text))

Max Activating Token Index: 721

Text #19

Max Range: 0.0302. Min Range: -0.0302

Max Act: 0.0223. Min Act: -0.0006

Data Index: 576290 (C4 (Web Text))

Max Activating Token Index: 13