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

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

Neuron 5008 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: 0.1366. Min Range: -0.1366

Max Act: 0.1299. Min Act: -0.0001

Data Index: 562108 (C4 (Web Text))

Max Activating Token Index: 22

Text #1

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1366. Min Act: -0.0001

Data Index: 463456 (C4 (Web Text))

Max Activating Token Index: 343

Text #2

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1295. Min Act: -0.0001

Data Index: 78894 (C4 (Web Text))

Max Activating Token Index: 45

Text #3

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1314. Min Act: -0.0001

Data Index: 1008987 (C4 (Web Text))

Max Activating Token Index: 38

Text #4

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1344. Min Act: -0.0001

Data Index: 452072 (C4 (Web Text))

Max Activating Token Index: 269

Text #5

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1331. Min Act: -0.0001

Data Index: 52092 (C4 (Web Text))

Max Activating Token Index: 153

Text #6

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1294. Min Act: -0.0001

Data Index: 420755 (C4 (Web Text))

Max Activating Token Index: 461

Text #7

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1206. Min Act: -0.0001

Data Index: 177663 (C4 (Web Text))

Max Activating Token Index: 592

Text #8

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1223. Min Act: -0.0001

Data Index: 519461 (C4 (Web Text))

Max Activating Token Index: 281

Text #9

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1359. Min Act: -0.0001

Data Index: 683872 (C4 (Web Text))

Max Activating Token Index: 112

Text #10

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1186. Min Act: -0.0001

Data Index: 299787 (C4 (Web Text))

Max Activating Token Index: 206

Text #11

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1291. Min Act: -0.0001

Data Index: 149988 (C4 (Web Text))

Max Activating Token Index: 369

Text #12

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1244. Min Act: -0.0001

Data Index: 666375 (C4 (Web Text))

Max Activating Token Index: 124

Text #13

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1320. Min Act: -0.0001

Data Index: 1006651 (C4 (Web Text))

Max Activating Token Index: 643

Text #14

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1129. Min Act: -0.0001

Data Index: 1510234 (Python Code)

Max Activating Token Index: 869

Text #15

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1284. Min Act: -0.0001

Data Index: 1196683 (C4 (Web Text))

Max Activating Token Index: 126

Text #16

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1177. Min Act: -0.0001

Data Index: 65942 (C4 (Web Text))

Max Activating Token Index: 956

Text #17

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1228. Min Act: -0.0001

Data Index: 754509 (C4 (Web Text))

Max Activating Token Index: 882

Text #18

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1224. Min Act: -0.0001

Data Index: 1203188 (C4 (Web Text))

Max Activating Token Index: 24

Text #19

Max Range: 0.1366. Min Range: -0.1366

Max Act: 0.1229. Min Act: -0.0001

Data Index: 13414 (C4 (Web Text))

Max Activating Token Index: 52