Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1072 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0306. Min Act: -0.0003

Data Index: 1567574 (Python Code)

Max Activating Token Index: 656

Text #1

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0271. Min Act: -0.0003

Data Index: 1100214 (C4 (Web Text))

Max Activating Token Index: 989

Text #2

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0274. Min Act: -0.0003

Data Index: 246574 (C4 (Web Text))

Max Activating Token Index: 251

Text #3

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0281. Min Act: -0.0003

Data Index: 1457313 (Python Code)

Max Activating Token Index: 883

Text #4

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0284. Min Act: -0.0003

Data Index: 1007123 (C4 (Web Text))

Max Activating Token Index: 920

Text #5

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0287. Min Act: -0.0003

Data Index: 240538 (C4 (Web Text))

Max Activating Token Index: 176

Text #6

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0269. Min Act: -0.0003

Data Index: 1565335 (Python Code)

Max Activating Token Index: 896

Text #7

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0267. Min Act: -0.0003

Data Index: 1464428 (Python Code)

Max Activating Token Index: 599

Text #8

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0251. Min Act: -0.0003

Data Index: 966223 (C4 (Web Text))

Max Activating Token Index: 236

Text #9

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0244. Min Act: -0.0003

Data Index: 1495049 (Python Code)

Max Activating Token Index: 204

Text #10

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0271. Min Act: -0.0003

Data Index: 1201660 (C4 (Web Text))

Max Activating Token Index: 589

Text #11

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0254. Min Act: -0.0003

Data Index: 1629159 (Python Code)

Max Activating Token Index: 582

Text #12

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0220. Min Act: -0.0003

Data Index: 693016 (C4 (Web Text))

Max Activating Token Index: 808

Text #13

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0236. Min Act: -0.0003

Data Index: 863558 (C4 (Web Text))

Max Activating Token Index: 387

Text #14

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0239. Min Act: -0.0003

Data Index: 1199761 (C4 (Web Text))

Max Activating Token Index: 712

Text #15

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0219. Min Act: -0.0003

Data Index: 22958 (C4 (Web Text))

Max Activating Token Index: 334

Text #16

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0232. Min Act: -0.0003

Data Index: 1265097 (C4 (Web Text))

Max Activating Token Index: 443

Text #17

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0225. Min Act: -0.0003

Data Index: 1065717 (C4 (Web Text))

Max Activating Token Index: 714

Text #18

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0246. Min Act: -0.0003

Data Index: 1443290 (Python Code)

Max Activating Token Index: 792

Text #19

Max Range: 0.0306. Min Range: -0.0306

Max Act: 0.0243. Min Act: -0.0003

Data Index: 1496657 (Python Code)

Max Activating Token Index: 550