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

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

Neuron 2010 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: 0.0272. Min Range: -0.0272

Max Act: 0.0272. Min Act: -0.0006

Data Index: 622913 (C4 (Web Text))

Max Activating Token Index: 176

Text #1

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0272. Min Act: -0.0006

Data Index: 827050 (C4 (Web Text))

Max Activating Token Index: 244

Text #2

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0248. Min Act: -0.0006

Data Index: 882068 (C4 (Web Text))

Max Activating Token Index: 215

Text #3

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0239. Min Act: -0.0006

Data Index: 980177 (C4 (Web Text))

Max Activating Token Index: 314

Text #4

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0233. Min Act: -0.0006

Data Index: 906014 (C4 (Web Text))

Max Activating Token Index: 944

Text #5

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0231. Min Act: -0.0006

Data Index: 632525 (C4 (Web Text))

Max Activating Token Index: 706

Text #6

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0235. Min Act: -0.0006

Data Index: 310772 (C4 (Web Text))

Max Activating Token Index: 123

Text #7

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0236. Min Act: -0.0006

Data Index: 1230792 (C4 (Web Text))

Max Activating Token Index: 369

Text #8

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0233. Min Act: -0.0006

Data Index: 220053 (C4 (Web Text))

Max Activating Token Index: 777

Text #9

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0226. Min Act: -0.0006

Data Index: 381282 (C4 (Web Text))

Max Activating Token Index: 260

Text #10

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0223. Min Act: -0.0006

Data Index: 784730 (C4 (Web Text))

Max Activating Token Index: 869

Text #11

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0218. Min Act: -0.0006

Data Index: 993983 (C4 (Web Text))

Max Activating Token Index: 503

Text #12

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0213. Min Act: -0.0006

Data Index: 725173 (C4 (Web Text))

Max Activating Token Index: 501

Text #13

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0233. Min Act: -0.0006

Data Index: 1192058 (C4 (Web Text))

Max Activating Token Index: 52

Text #14

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0222. Min Act: -0.0006

Data Index: 787028 (C4 (Web Text))

Max Activating Token Index: 1000

Text #15

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0207. Min Act: -0.0006

Data Index: 730185 (C4 (Web Text))

Max Activating Token Index: 1011

Text #16

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0221. Min Act: -0.0006

Data Index: 584472 (C4 (Web Text))

Max Activating Token Index: 55

Text #17

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0224. Min Act: -0.0006

Data Index: 690664 (C4 (Web Text))

Max Activating Token Index: 535

Text #18

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0219. Min Act: -0.0006

Data Index: 104927 (C4 (Web Text))

Max Activating Token Index: 1002

Text #19

Max Range: 0.0272. Min Range: -0.0272

Max Act: 0.0217. Min Act: -0.0006

Data Index: 3627 (C4 (Web Text))

Max Activating Token Index: 560