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

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

Neuron 4053 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: 0.1659. Min Range: -0.1659

Max Act: 0.1659. Min Act: -0.0001

Data Index: 872975 (C4 (Web Text))

Max Activating Token Index: 779

Text #1

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1342. Min Act: -0.0001

Data Index: 406774 (C4 (Web Text))

Max Activating Token Index: 906

Text #2

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1354. Min Act: -0.0001

Data Index: 26219 (C4 (Web Text))

Max Activating Token Index: 748

Text #3

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1234. Min Act: -0.0001

Data Index: 430041 (C4 (Web Text))

Max Activating Token Index: 798

Text #4

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1301. Min Act: -0.0001

Data Index: 692322 (C4 (Web Text))

Max Activating Token Index: 31

Text #5

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1446. Min Act: -0.0001

Data Index: 945765 (C4 (Web Text))

Max Activating Token Index: 675

Text #6

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1301. Min Act: -0.0001

Data Index: 500696 (C4 (Web Text))

Max Activating Token Index: 51

Text #7

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1260. Min Act: -0.0001

Data Index: 1610378 (Python Code)

Max Activating Token Index: 44

Text #8

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1255. Min Act: -0.0001

Data Index: 149073 (C4 (Web Text))

Max Activating Token Index: 469

Text #9

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1310. Min Act: -0.0001

Data Index: 1530776 (Python Code)

Max Activating Token Index: 5

Text #10

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1223. Min Act: -0.0001

Data Index: 671881 (C4 (Web Text))

Max Activating Token Index: 703

Text #11

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1225. Min Act: -0.0001

Data Index: 1029010 (C4 (Web Text))

Max Activating Token Index: 886

Text #12

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1183. Min Act: -0.0001

Data Index: 1231659 (C4 (Web Text))

Max Activating Token Index: 982

Text #13

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1248. Min Act: -0.0001

Data Index: 1017587 (C4 (Web Text))

Max Activating Token Index: 10

Text #14

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1247. Min Act: -0.0001

Data Index: 1649485 (Python Code)

Max Activating Token Index: 13

Text #15

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1243. Min Act: -0.0001

Data Index: 724604 (C4 (Web Text))

Max Activating Token Index: 497

Text #16

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1234. Min Act: -0.0001

Data Index: 1532547 (Python Code)

Max Activating Token Index: 162

Text #17

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1214. Min Act: -0.0001

Data Index: 1100598 (C4 (Web Text))

Max Activating Token Index: 965

Text #18

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1083. Min Act: -0.0001

Data Index: 513877 (C4 (Web Text))

Max Activating Token Index: 1020

Text #19

Max Range: 0.1659. Min Range: -0.1659

Max Act: 0.1082. Min Act: -0.0001

Data Index: 1498061 (Python Code)

Max Activating Token Index: 244