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

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

Neuron 56 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: 0.0268. Min Range: -0.0268

Max Act: 0.0268. Min Act: -0.0006

Data Index: 219509 (C4 (Web Text))

Max Activating Token Index: 65

Text #1

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0231. Min Act: -0.0005

Data Index: 1142109 (C4 (Web Text))

Max Activating Token Index: 325

Text #2

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0235. Min Act: -0.0006

Data Index: 392315 (C4 (Web Text))

Max Activating Token Index: 444

Text #3

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0226. Min Act: -0.0006

Data Index: 479391 (C4 (Web Text))

Max Activating Token Index: 189

Text #4

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0216. Min Act: -0.0006

Data Index: 227133 (C4 (Web Text))

Max Activating Token Index: 763

Text #5

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0204. Min Act: -0.0006

Data Index: 1276313 (C4 (Web Text))

Max Activating Token Index: 31

Text #6

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0203. Min Act: -0.0006

Data Index: 178082 (C4 (Web Text))

Max Activating Token Index: 30

Text #7

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0203. Min Act: -0.0006

Data Index: 1206348 (C4 (Web Text))

Max Activating Token Index: 148

Text #8

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0199. Min Act: -0.0006

Data Index: 569920 (C4 (Web Text))

Max Activating Token Index: 557

Text #9

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0197. Min Act: -0.0006

Data Index: 542994 (C4 (Web Text))

Max Activating Token Index: 352

Text #10

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0188. Min Act: -0.0006

Data Index: 225214 (C4 (Web Text))

Max Activating Token Index: 157

Text #11

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0191. Min Act: -0.0006

Data Index: 695922 (C4 (Web Text))

Max Activating Token Index: 340

Text #12

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0195. Min Act: -0.0006

Data Index: 104095 (C4 (Web Text))

Max Activating Token Index: 46

Text #13

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0195. Min Act: -0.0006

Data Index: 852285 (C4 (Web Text))

Max Activating Token Index: 196

Text #14

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0196. Min Act: -0.0006

Data Index: 663332 (C4 (Web Text))

Max Activating Token Index: 397

Text #15

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0202. Min Act: -0.0006

Data Index: 588767 (C4 (Web Text))

Max Activating Token Index: 936

Text #16

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0190. Min Act: -0.0006

Data Index: 300594 (C4 (Web Text))

Max Activating Token Index: 252

Text #17

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0184. Min Act: -0.0006

Data Index: 1173915 (C4 (Web Text))

Max Activating Token Index: 244

Text #18

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0184. Min Act: -0.0006

Data Index: 542249 (C4 (Web Text))

Max Activating Token Index: 342

Text #19

Max Range: 0.0268. Min Range: -0.0268

Max Act: 0.0187. Min Act: -0.0006

Data Index: 104423 (C4 (Web Text))

Max Activating Token Index: 296