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

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

Neuron 3371 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0337. Min Act: -0.0001

Data Index: 935574 (C4 (Web Text))

Max Activating Token Index: 250

Text #1

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0306. Min Act: -0.0001

Data Index: 30845 (C4 (Web Text))

Max Activating Token Index: 906

Text #2

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0313. Min Act: -0.0001

Data Index: 1636633 (Python Code)

Max Activating Token Index: 237

Text #3

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0283. Min Act: -0.0001

Data Index: 1241912 (C4 (Web Text))

Max Activating Token Index: 806

Text #4

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0268. Min Act: -0.0001

Data Index: 490989 (C4 (Web Text))

Max Activating Token Index: 597

Text #5

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0272. Min Act: -0.0001

Data Index: 1428964 (Python Code)

Max Activating Token Index: 344

Text #6

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0277. Min Act: -0.0001

Data Index: 1217549 (C4 (Web Text))

Max Activating Token Index: 142

Text #7

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0269. Min Act: -0.0001

Data Index: 339516 (C4 (Web Text))

Max Activating Token Index: 690

Text #8

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0265. Min Act: -0.0001

Data Index: 1010967 (C4 (Web Text))

Max Activating Token Index: 423

Text #9

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0259. Min Act: -0.0001

Data Index: 780870 (C4 (Web Text))

Max Activating Token Index: 702

Text #10

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0243. Min Act: -0.0001

Data Index: 1510486 (Python Code)

Max Activating Token Index: 164

Text #11

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0221. Min Act: -0.0001

Data Index: 321008 (C4 (Web Text))

Max Activating Token Index: 613

Text #12

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0225. Min Act: -0.0001

Data Index: 900489 (C4 (Web Text))

Max Activating Token Index: 136

Text #13

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0238. Min Act: -0.0001

Data Index: 1507456 (Python Code)

Max Activating Token Index: 601

Text #14

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0214. Min Act: -0.0001

Data Index: 793740 (C4 (Web Text))

Max Activating Token Index: 761

Text #15

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0236. Min Act: -0.0001

Data Index: 20608 (C4 (Web Text))

Max Activating Token Index: 758

Text #16

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0240. Min Act: -0.0001

Data Index: 1293740 (C4 (Web Text))

Max Activating Token Index: 133

Text #17

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0233. Min Act: -0.0001

Data Index: 1591987 (Python Code)

Max Activating Token Index: 318

Text #18

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0242. Min Act: -0.0001

Data Index: 1544298 (Python Code)

Max Activating Token Index: 976

Text #19

Max Range: 0.0337. Min Range: -0.0337

Max Act: 0.0246. Min Act: -0.0001

Data Index: 416564 (C4 (Web Text))

Max Activating Token Index: 108