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

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

Neuron 650 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: 0.0271. Min Range: -0.0271

Max Act: 0.0271. Min Act: -0.0006

Data Index: 418039 (C4 (Web Text))

Max Activating Token Index: 1013

Text #1

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0253. Min Act: -0.0005

Data Index: 1580017 (Python Code)

Max Activating Token Index: 716

Text #2

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0245. Min Act: -0.0006

Data Index: 432728 (C4 (Web Text))

Max Activating Token Index: 274

Text #3

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0244. Min Act: -0.0006

Data Index: 445376 (C4 (Web Text))

Max Activating Token Index: 810

Text #4

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0226. Min Act: -0.0006

Data Index: 309018 (C4 (Web Text))

Max Activating Token Index: 176

Text #5

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0216. Min Act: -0.0006

Data Index: 1054241 (C4 (Web Text))

Max Activating Token Index: 1004

Text #6

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0225. Min Act: -0.0006

Data Index: 639947 (C4 (Web Text))

Max Activating Token Index: 878

Text #7

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0230. Min Act: -0.0006

Data Index: 549076 (C4 (Web Text))

Max Activating Token Index: 778

Text #8

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0241. Min Act: -0.0006

Data Index: 1200588 (C4 (Web Text))

Max Activating Token Index: 8

Text #9

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0227. Min Act: -0.0006

Data Index: 1220405 (C4 (Web Text))

Max Activating Token Index: 148

Text #10

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0212. Min Act: -0.0006

Data Index: 875476 (C4 (Web Text))

Max Activating Token Index: 228

Text #11

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0226. Min Act: -0.0006

Data Index: 494966 (C4 (Web Text))

Max Activating Token Index: 88

Text #12

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0235. Min Act: -0.0006

Data Index: 786971 (C4 (Web Text))

Max Activating Token Index: 150

Text #13

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0229. Min Act: -0.0006

Data Index: 1063252 (C4 (Web Text))

Max Activating Token Index: 738

Text #14

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0224. Min Act: -0.0006

Data Index: 866488 (C4 (Web Text))

Max Activating Token Index: 743

Text #15

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0234. Min Act: -0.0006

Data Index: 1200343 (C4 (Web Text))

Max Activating Token Index: 688

Text #16

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0230. Min Act: -0.0006

Data Index: 1547798 (Python Code)

Max Activating Token Index: 72

Text #17

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0224. Min Act: -0.0006

Data Index: 1135263 (C4 (Web Text))

Max Activating Token Index: 905

Text #18

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0225. Min Act: -0.0006

Data Index: 571872 (C4 (Web Text))

Max Activating Token Index: 72

Text #19

Max Range: 0.0271. Min Range: -0.0271

Max Act: 0.0222. Min Act: -0.0006

Data Index: 328700 (C4 (Web Text))

Max Activating Token Index: 971