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

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

Neuron 755 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0282. Min Act: -0.0006

Data Index: 616903 (C4 (Web Text))

Max Activating Token Index: 856

Text #1

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0236. Min Act: -0.0006

Data Index: 1201502 (C4 (Web Text))

Max Activating Token Index: 959

Text #2

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0207. Min Act: -0.0006

Data Index: 277774 (C4 (Web Text))

Max Activating Token Index: 636

Text #3

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0200. Min Act: -0.0006

Data Index: 1039031 (C4 (Web Text))

Max Activating Token Index: 349

Text #4

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0196. Min Act: -0.0006

Data Index: 1078805 (C4 (Web Text))

Max Activating Token Index: 857

Text #5

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0199. Min Act: -0.0006

Data Index: 1454370 (Python Code)

Max Activating Token Index: 469

Text #6

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0181. Min Act: -0.0006

Data Index: 1409777 (Python Code)

Max Activating Token Index: 67

Text #7

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0193. Min Act: -0.0006

Data Index: 156290 (C4 (Web Text))

Max Activating Token Index: 122

Text #8

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0184. Min Act: -0.0006

Data Index: 496274 (C4 (Web Text))

Max Activating Token Index: 986

Text #9

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0191. Min Act: -0.0006

Data Index: 1310707 (C4 (Web Text))

Max Activating Token Index: 649

Text #10

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0193. Min Act: -0.0006

Data Index: 135878 (C4 (Web Text))

Max Activating Token Index: 891

Text #11

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0177. Min Act: -0.0006

Data Index: 531381 (C4 (Web Text))

Max Activating Token Index: 312

Text #12

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0168. Min Act: -0.0006

Data Index: 312939 (C4 (Web Text))

Max Activating Token Index: 896

Text #13

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0167. Min Act: -0.0005

Data Index: 226206 (C4 (Web Text))

Max Activating Token Index: 952

Text #14

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0175. Min Act: -0.0006

Data Index: 614045 (C4 (Web Text))

Max Activating Token Index: 805

Text #15

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0182. Min Act: -0.0005

Data Index: 1615480 (Python Code)

Max Activating Token Index: 656

Text #16

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0167. Min Act: -0.0005

Data Index: 1365232 (Python Code)

Max Activating Token Index: 841

Text #17

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0173. Min Act: -0.0005

Data Index: 984752 (C4 (Web Text))

Max Activating Token Index: 372

Text #18

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0172. Min Act: -0.0005

Data Index: 1029393 (C4 (Web Text))

Max Activating Token Index: 819

Text #19

Max Range: 0.0282. Min Range: -0.0282

Max Act: 0.0168. Min Act: -0.0006

Data Index: 97049 (C4 (Web Text))

Max Activating Token Index: 871