Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 2019 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1463. Min Act: -0.0001

Data Index: 952557 (C4 (Web Text))

Max Activating Token Index: 619

Text #1

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1508. Min Act: -0.0001

Data Index: 410461 (C4 (Web Text))

Max Activating Token Index: 291

Text #2

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1407. Min Act: -0.0001

Data Index: 24343 (C4 (Web Text))

Max Activating Token Index: 728

Text #3

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1300. Min Act: -0.0001

Data Index: 553706 (C4 (Web Text))

Max Activating Token Index: 158

Text #4

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1353. Min Act: -0.0001

Data Index: 1269793 (C4 (Web Text))

Max Activating Token Index: 864

Text #5

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1306. Min Act: -0.0001

Data Index: 767234 (C4 (Web Text))

Max Activating Token Index: 646

Text #6

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1326. Min Act: -0.0001

Data Index: 820853 (C4 (Web Text))

Max Activating Token Index: 434

Text #7

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1329. Min Act: -0.0001

Data Index: 183146 (C4 (Web Text))

Max Activating Token Index: 844

Text #8

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1320. Min Act: -0.0001

Data Index: 669468 (C4 (Web Text))

Max Activating Token Index: 818

Text #9

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1270. Min Act: -0.0001

Data Index: 1099118 (C4 (Web Text))

Max Activating Token Index: 825

Text #10

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1243. Min Act: -0.0001

Data Index: 978803 (C4 (Web Text))

Max Activating Token Index: 582

Text #11

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1337. Min Act: -0.0001

Data Index: 1186448 (C4 (Web Text))

Max Activating Token Index: 110

Text #12

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1231. Min Act: -0.0001

Data Index: 226100 (C4 (Web Text))

Max Activating Token Index: 494

Text #13

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1224. Min Act: -0.0001

Data Index: 1105184 (C4 (Web Text))

Max Activating Token Index: 43

Text #14

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1257. Min Act: -0.0001

Data Index: 1315398 (C4 (Web Text))

Max Activating Token Index: 454

Text #15

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1208. Min Act: -0.0001

Data Index: 520455 (C4 (Web Text))

Max Activating Token Index: 92

Text #16

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1207. Min Act: -0.0001

Data Index: 270857 (C4 (Web Text))

Max Activating Token Index: 1010

Text #17

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1248. Min Act: -0.0001

Data Index: 1220811 (C4 (Web Text))

Max Activating Token Index: 852

Text #18

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1106. Min Act: -0.0001

Data Index: 403246 (C4 (Web Text))

Max Activating Token Index: 451

Text #19

Max Range: 0.1508. Min Range: -0.1508

Max Act: 0.1219. Min Act: -0.0001

Data Index: 847591 (C4 (Web Text))

Max Activating Token Index: 461