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

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

Neuron 1295 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1525. Min Act: -0.0002

Data Index: 61449 (C4 (Web Text))

Max Activating Token Index: 5

Text #1

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1439. Min Act: -0.0002

Data Index: 1135913 (C4 (Web Text))

Max Activating Token Index: 7

Text #2

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1463. Min Act: -0.0002

Data Index: 1068721 (C4 (Web Text))

Max Activating Token Index: 961

Text #3

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1452. Min Act: -0.0002

Data Index: 553848 (C4 (Web Text))

Max Activating Token Index: 962

Text #4

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1404. Min Act: -0.0002

Data Index: 387783 (C4 (Web Text))

Max Activating Token Index: 494

Text #5

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1362. Min Act: -0.0002

Data Index: 959481 (C4 (Web Text))

Max Activating Token Index: 205

Text #6

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1392. Min Act: -0.0001

Data Index: 1403740 (Python Code)

Max Activating Token Index: 4

Text #7

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1363. Min Act: -0.0001

Data Index: 93277 (C4 (Web Text))

Max Activating Token Index: 14

Text #8

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1295. Min Act: -0.0002

Data Index: 662548 (C4 (Web Text))

Max Activating Token Index: 314

Text #9

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1334. Min Act: -0.0002

Data Index: 1195997 (C4 (Web Text))

Max Activating Token Index: 415

Text #10

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1377. Min Act: -0.0001

Data Index: 1311742 (C4 (Web Text))

Max Activating Token Index: 187

Text #11

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1287. Min Act: -0.0001

Data Index: 1650853 (Python Code)

Max Activating Token Index: 32

Text #12

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1368. Min Act: -0.0002

Data Index: 1041928 (C4 (Web Text))

Max Activating Token Index: 625

Text #13

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1266. Min Act: -0.0001

Data Index: 1112666 (C4 (Web Text))

Max Activating Token Index: 650

Text #14

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1298. Min Act: -0.0002

Data Index: 1253400 (C4 (Web Text))

Max Activating Token Index: 245

Text #15

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1334. Min Act: -0.0002

Data Index: 2185 (C4 (Web Text))

Max Activating Token Index: 614

Text #16

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1327. Min Act: -0.0002

Data Index: 217345 (C4 (Web Text))

Max Activating Token Index: 98

Text #17

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1324. Min Act: -0.0002

Data Index: 371276 (C4 (Web Text))

Max Activating Token Index: 426

Text #18

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1245. Min Act: -0.0001

Data Index: 1502224 (Python Code)

Max Activating Token Index: 32

Text #19

Max Range: 0.1525. Min Range: -0.1525

Max Act: 0.1235. Min Act: -0.0001

Data Index: 905111 (C4 (Web Text))

Max Activating Token Index: 742