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

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

Neuron 3198 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1697. Min Act: -0.0001

Data Index: 265580 (C4 (Web Text))

Max Activating Token Index: 853

Text #1

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1475. Min Act: -0.0001

Data Index: 785698 (C4 (Web Text))

Max Activating Token Index: 11

Text #2

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1373. Min Act: -0.0001

Data Index: 1262392 (C4 (Web Text))

Max Activating Token Index: 16

Text #3

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1319. Min Act: -0.0001

Data Index: 1315082 (C4 (Web Text))

Max Activating Token Index: 103

Text #4

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1265. Min Act: -0.0001

Data Index: 223113 (C4 (Web Text))

Max Activating Token Index: 728

Text #5

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1296. Min Act: -0.0001

Data Index: 932743 (C4 (Web Text))

Max Activating Token Index: 744

Text #6

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1238. Min Act: -0.0001

Data Index: 852390 (C4 (Web Text))

Max Activating Token Index: 43

Text #7

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1284. Min Act: -0.0001

Data Index: 855265 (C4 (Web Text))

Max Activating Token Index: 12

Text #8

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1260. Min Act: -0.0001

Data Index: 806619 (C4 (Web Text))

Max Activating Token Index: 43

Text #9

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1248. Min Act: -0.0001

Data Index: 902175 (C4 (Web Text))

Max Activating Token Index: 111

Text #10

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1261. Min Act: -0.0001

Data Index: 1128214 (C4 (Web Text))

Max Activating Token Index: 152

Text #11

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1209. Min Act: -0.0001

Data Index: 881335 (C4 (Web Text))

Max Activating Token Index: 278

Text #12

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1145. Min Act: -0.0001

Data Index: 1150638 (C4 (Web Text))

Max Activating Token Index: 106

Text #13

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1121. Min Act: -0.0001

Data Index: 543490 (C4 (Web Text))

Max Activating Token Index: 553

Text #14

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1114. Min Act: -0.0001

Data Index: 1024789 (C4 (Web Text))

Max Activating Token Index: 65

Text #15

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1067. Min Act: -0.0001

Data Index: 94510 (C4 (Web Text))

Max Activating Token Index: 382

Text #16

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1104. Min Act: -0.0001

Data Index: 231264 (C4 (Web Text))

Max Activating Token Index: 48

Text #17

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1125. Min Act: -0.0001

Data Index: 700192 (C4 (Web Text))

Max Activating Token Index: 308

Text #18

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1090. Min Act: -0.0001

Data Index: 728370 (C4 (Web Text))

Max Activating Token Index: 74

Text #19

Max Range: 0.1697. Min Range: -0.1697

Max Act: 0.1045. Min Act: -0.0001

Data Index: 721578 (C4 (Web Text))

Max Activating Token Index: 797