Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5473 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1435. Min Act: -0.0001

Data Index: 504942 (C4 (Web Text))

Max Activating Token Index: 872

Text #1

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1455. Min Act: -0.0001

Data Index: 876463 (C4 (Web Text))

Max Activating Token Index: 687

Text #2

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1387. Min Act: -0.0001

Data Index: 1559374 (Python Code)

Max Activating Token Index: 15

Text #3

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1276. Min Act: -0.0001

Data Index: 1197138 (C4 (Web Text))

Max Activating Token Index: 925

Text #4

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1261. Min Act: -0.0002

Data Index: 987143 (C4 (Web Text))

Max Activating Token Index: 177

Text #5

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1254. Min Act: -0.0001

Data Index: 66773 (C4 (Web Text))

Max Activating Token Index: 558

Text #6

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1337. Min Act: -0.0001

Data Index: 1075485 (C4 (Web Text))

Max Activating Token Index: 644

Text #7

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1223. Min Act: -0.0001

Data Index: 1483170 (Python Code)

Max Activating Token Index: 34

Text #8

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1387. Min Act: -0.0001

Data Index: 114637 (C4 (Web Text))

Max Activating Token Index: 494

Text #9

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1170. Min Act: -0.0001

Data Index: 15263 (C4 (Web Text))

Max Activating Token Index: 364

Text #10

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1367. Min Act: -0.0001

Data Index: 1031041 (C4 (Web Text))

Max Activating Token Index: 263

Text #11

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1286. Min Act: -0.0001

Data Index: 1561539 (Python Code)

Max Activating Token Index: 22

Text #12

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1184. Min Act: -0.0001

Data Index: 1649674 (Python Code)

Max Activating Token Index: 33

Text #13

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1259. Min Act: -0.0001

Data Index: 1363793 (Python Code)

Max Activating Token Index: 232

Text #14

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1274. Min Act: -0.0001

Data Index: 980557 (C4 (Web Text))

Max Activating Token Index: 879

Text #15

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1220. Min Act: -0.0001

Data Index: 1452980 (Python Code)

Max Activating Token Index: 96

Text #16

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1094. Min Act: -0.0001

Data Index: 1121129 (C4 (Web Text))

Max Activating Token Index: 878

Text #17

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1165. Min Act: -0.0001

Data Index: 1511268 (Python Code)

Max Activating Token Index: 20

Text #18

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1250. Min Act: -0.0001

Data Index: 1420871 (Python Code)

Max Activating Token Index: 191

Text #19

Max Range: 0.1455. Min Range: -0.1455

Max Act: 0.1197. Min Act: -0.0001

Data Index: 1135669 (C4 (Web Text))

Max Activating Token Index: 263