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

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

Neuron 575 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1702. Min Act: -0.0001

Data Index: 113981 (C4 (Web Text))

Max Activating Token Index: 118

Text #1

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1723. Min Act: -0.0001

Data Index: 138158 (C4 (Web Text))

Max Activating Token Index: 21

Text #2

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1596. Min Act: -0.0001

Data Index: 975060 (C4 (Web Text))

Max Activating Token Index: 59

Text #3

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1517. Min Act: -0.0002

Data Index: 1307214 (C4 (Web Text))

Max Activating Token Index: 11

Text #4

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1482. Min Act: -0.0001

Data Index: 153673 (C4 (Web Text))

Max Activating Token Index: 60

Text #5

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1399. Min Act: -0.0001

Data Index: 1072854 (C4 (Web Text))

Max Activating Token Index: 136

Text #6

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1334. Min Act: -0.0002

Data Index: 237923 (C4 (Web Text))

Max Activating Token Index: 10

Text #7

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1331. Min Act: -0.0001

Data Index: 1199785 (C4 (Web Text))

Max Activating Token Index: 55

Text #8

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1320. Min Act: -0.0001

Data Index: 357931 (C4 (Web Text))

Max Activating Token Index: 857

Text #9

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1299. Min Act: 0.0002

Data Index: 74356 (C4 (Web Text))

Max Activating Token Index: 81

Text #10

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1271. Min Act: 0.0002

Data Index: 90221 (C4 (Web Text))

Max Activating Token Index: 154

Text #11

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1231. Min Act: -0.0001

Data Index: 20427 (C4 (Web Text))

Max Activating Token Index: 113

Text #12

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1269. Min Act: -0.0000

Data Index: 1079950 (C4 (Web Text))

Max Activating Token Index: 154

Text #13

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1249. Min Act: -0.0001

Data Index: 89868 (C4 (Web Text))

Max Activating Token Index: 18

Text #14

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1251. Min Act: -0.0001

Data Index: 228219 (C4 (Web Text))

Max Activating Token Index: 97

Text #15

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1200. Min Act: 0.0001

Data Index: 86847 (C4 (Web Text))

Max Activating Token Index: 30

Text #16

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1223. Min Act: -0.0002

Data Index: 265184 (C4 (Web Text))

Max Activating Token Index: 15

Text #17

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1213. Min Act: -0.0002

Data Index: 806694 (C4 (Web Text))

Max Activating Token Index: 766

Text #18

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1228. Min Act: -0.0000

Data Index: 914364 (C4 (Web Text))

Max Activating Token Index: 34

Text #19

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1183. Min Act: -0.0001

Data Index: 430006 (C4 (Web Text))

Max Activating Token Index: 65