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

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

Neuron 3568 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1515. Min Act: -0.0002

Data Index: 514471 (C4 (Web Text))

Max Activating Token Index: 787

Text #1

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1668. Min Act: -0.0002

Data Index: 394796 (C4 (Web Text))

Max Activating Token Index: 83

Text #2

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1526. Min Act: -0.0002

Data Index: 899733 (C4 (Web Text))

Max Activating Token Index: 517

Text #3

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1440. Min Act: -0.0002

Data Index: 142597 (C4 (Web Text))

Max Activating Token Index: 638

Text #4

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1514. Min Act: -0.0002

Data Index: 1405502 (Python Code)

Max Activating Token Index: 334

Text #5

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1494. Min Act: -0.0002

Data Index: 1400787 (Python Code)

Max Activating Token Index: 421

Text #6

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1462. Min Act: -0.0002

Data Index: 1090458 (C4 (Web Text))

Max Activating Token Index: 94

Text #7

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1503. Min Act: -0.0002

Data Index: 129089 (C4 (Web Text))

Max Activating Token Index: 121

Text #8

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1435. Min Act: -0.0002

Data Index: 273878 (C4 (Web Text))

Max Activating Token Index: 837

Text #9

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1448. Min Act: -0.0002

Data Index: 1368465 (Python Code)

Max Activating Token Index: 809

Text #10

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1526. Min Act: -0.0002

Data Index: 139489 (C4 (Web Text))

Max Activating Token Index: 3

Text #11

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1323. Min Act: -0.0002

Data Index: 696334 (C4 (Web Text))

Max Activating Token Index: 814

Text #12

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1354. Min Act: -0.0002

Data Index: 1465605 (Python Code)

Max Activating Token Index: 459

Text #13

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1235. Min Act: -0.0002

Data Index: 196116 (C4 (Web Text))

Max Activating Token Index: 812

Text #14

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1413. Min Act: -0.0002

Data Index: 65403 (C4 (Web Text))

Max Activating Token Index: 936

Text #15

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1075. Min Act: -0.0002

Data Index: 1242095 (C4 (Web Text))

Max Activating Token Index: 206

Text #16

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1398. Min Act: -0.0003

Data Index: 810499 (C4 (Web Text))

Max Activating Token Index: 478

Text #17

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1355. Min Act: -0.0002

Data Index: 906258 (C4 (Web Text))

Max Activating Token Index: 755

Text #18

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1274. Min Act: -0.0002

Data Index: 1122371 (C4 (Web Text))

Max Activating Token Index: 864

Text #19

Max Range: 0.1668. Min Range: -0.1668

Max Act: 0.1491. Min Act: -0.0002

Data Index: 373019 (C4 (Web Text))

Max Activating Token Index: 776