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

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

Neuron 972 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: 0.0223. Min Range: -0.0223

Max Act: 0.0223. Min Act: -0.0006

Data Index: 1410416 (Python Code)

Max Activating Token Index: 84

Text #1

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0220. Min Act: -0.0006

Data Index: 252572 (C4 (Web Text))

Max Activating Token Index: 241

Text #2

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0207. Min Act: -0.0006

Data Index: 207821 (C4 (Web Text))

Max Activating Token Index: 93

Text #3

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0198. Min Act: -0.0006

Data Index: 17648 (C4 (Web Text))

Max Activating Token Index: 477

Text #4

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0205. Min Act: -0.0006

Data Index: 438478 (C4 (Web Text))

Max Activating Token Index: 138

Text #5

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0199. Min Act: -0.0006

Data Index: 251333 (C4 (Web Text))

Max Activating Token Index: 40

Text #6

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0195. Min Act: -0.0006

Data Index: 380232 (C4 (Web Text))

Max Activating Token Index: 939

Text #7

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0199. Min Act: -0.0006

Data Index: 1178271 (C4 (Web Text))

Max Activating Token Index: 188

Text #8

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0195. Min Act: -0.0006

Data Index: 702945 (C4 (Web Text))

Max Activating Token Index: 357

Text #9

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0194. Min Act: -0.0006

Data Index: 898816 (C4 (Web Text))

Max Activating Token Index: 255

Text #10

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0190. Min Act: -0.0006

Data Index: 1496604 (Python Code)

Max Activating Token Index: 584

Text #11

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0197. Min Act: -0.0006

Data Index: 1197364 (C4 (Web Text))

Max Activating Token Index: 268

Text #12

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0196. Min Act: -0.0006

Data Index: 665662 (C4 (Web Text))

Max Activating Token Index: 807

Text #13

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0190. Min Act: -0.0006

Data Index: 1143727 (C4 (Web Text))

Max Activating Token Index: 17

Text #14

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0190. Min Act: -0.0006

Data Index: 702302 (C4 (Web Text))

Max Activating Token Index: 53

Text #15

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0185. Min Act: -0.0006

Data Index: 920745 (C4 (Web Text))

Max Activating Token Index: 631

Text #16

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0193. Min Act: -0.0006

Data Index: 1492278 (Python Code)

Max Activating Token Index: 389

Text #17

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0173. Min Act: -0.0006

Data Index: 436411 (C4 (Web Text))

Max Activating Token Index: 981

Text #18

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0189. Min Act: -0.0006

Data Index: 924493 (C4 (Web Text))

Max Activating Token Index: 783

Text #19

Max Range: 0.0223. Min Range: -0.0223

Max Act: 0.0189. Min Act: -0.0006

Data Index: 1006842 (C4 (Web Text))

Max Activating Token Index: 490