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

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

Neuron 2711 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1698. Min Act: -0.0001

Data Index: 1429591 (Python Code)

Max Activating Token Index: 328

Text #1

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1415. Min Act: -0.0001

Data Index: 787530 (C4 (Web Text))

Max Activating Token Index: 822

Text #2

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1456. Min Act: -0.0001

Data Index: 380894 (C4 (Web Text))

Max Activating Token Index: 925

Text #3

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1439. Min Act: -0.0001

Data Index: 938593 (C4 (Web Text))

Max Activating Token Index: 25

Text #4

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1423. Min Act: -0.0001

Data Index: 224213 (C4 (Web Text))

Max Activating Token Index: 864

Text #5

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1405. Min Act: -0.0001

Data Index: 229821 (C4 (Web Text))

Max Activating Token Index: 640

Text #6

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1373. Min Act: -0.0001

Data Index: 727188 (C4 (Web Text))

Max Activating Token Index: 912

Text #7

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1349. Min Act: -0.0001

Data Index: 886520 (C4 (Web Text))

Max Activating Token Index: 388

Text #8

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1262. Min Act: -0.0000

Data Index: 1479565 (Python Code)

Max Activating Token Index: 50

Text #9

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1296. Min Act: -0.0001

Data Index: 77393 (C4 (Web Text))

Max Activating Token Index: 693

Text #10

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1278. Min Act: -0.0001

Data Index: 107927 (C4 (Web Text))

Max Activating Token Index: 32

Text #11

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1275. Min Act: -0.0001

Data Index: 1061854 (C4 (Web Text))

Max Activating Token Index: 143

Text #12

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1267. Min Act: -0.0001

Data Index: 1084885 (C4 (Web Text))

Max Activating Token Index: 474

Text #13

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1166. Min Act: -0.0001

Data Index: 1573660 (Python Code)

Max Activating Token Index: 729

Text #14

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1217. Min Act: -0.0001

Data Index: 1445233 (Python Code)

Max Activating Token Index: 689

Text #15

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1126. Min Act: -0.0001

Data Index: 18668 (C4 (Web Text))

Max Activating Token Index: 455

Text #16

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1049. Min Act: -0.0001

Data Index: 1403696 (Python Code)

Max Activating Token Index: 632

Text #17

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1206. Min Act: -0.0001

Data Index: 18998 (C4 (Web Text))

Max Activating Token Index: 944

Text #18

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1139. Min Act: -0.0001

Data Index: 965342 (C4 (Web Text))

Max Activating Token Index: 445

Text #19

Max Range: 0.1698. Min Range: -0.1698

Max Act: 0.1186. Min Act: -0.0001

Data Index: 568995 (C4 (Web Text))

Max Activating Token Index: 625