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

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

Neuron 1768 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1588. Min Act: -0.0000

Data Index: 539154 (C4 (Web Text))

Max Activating Token Index: 918

Text #1

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1648. Min Act: -0.0000

Data Index: 1352042 (C4 (Web Text))

Max Activating Token Index: 421

Text #2

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1675. Min Act: -0.0000

Data Index: 272628 (C4 (Web Text))

Max Activating Token Index: 682

Text #3

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1541. Min Act: 0.0002

Data Index: 780499 (C4 (Web Text))

Max Activating Token Index: 1001

Text #4

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1514. Min Act: -0.0000

Data Index: 798906 (C4 (Web Text))

Max Activating Token Index: 106

Text #5

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1453. Min Act: -0.0000

Data Index: 1204287 (C4 (Web Text))

Max Activating Token Index: 47

Text #6

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1449. Min Act: 0.0001

Data Index: 1535454 (Python Code)

Max Activating Token Index: 501

Text #7

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1414. Min Act: -0.0000

Data Index: 1195309 (C4 (Web Text))

Max Activating Token Index: 907

Text #8

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1415. Min Act: -0.0000

Data Index: 834279 (C4 (Web Text))

Max Activating Token Index: 824

Text #9

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1360. Min Act: -0.0000

Data Index: 565451 (C4 (Web Text))

Max Activating Token Index: 863

Text #10

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1382. Min Act: 0.0002

Data Index: 549887 (C4 (Web Text))

Max Activating Token Index: 263

Text #11

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1404. Min Act: -0.0000

Data Index: 946539 (C4 (Web Text))

Max Activating Token Index: 324

Text #12

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1315. Min Act: -0.0000

Data Index: 652087 (C4 (Web Text))

Max Activating Token Index: 490

Text #13

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1277. Min Act: 0.0000

Data Index: 1478622 (Python Code)

Max Activating Token Index: 370

Text #14

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1320. Min Act: 0.0001

Data Index: 8009 (C4 (Web Text))

Max Activating Token Index: 877

Text #15

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1344. Min Act: -0.0000

Data Index: 250716 (C4 (Web Text))

Max Activating Token Index: 70

Text #16

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1331. Min Act: -0.0000

Data Index: 13720 (C4 (Web Text))

Max Activating Token Index: 337

Text #17

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1266. Min Act: -0.0000

Data Index: 345476 (C4 (Web Text))

Max Activating Token Index: 123

Text #18

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1230. Min Act: -0.0000

Data Index: 808858 (C4 (Web Text))

Max Activating Token Index: 383

Text #19

Max Range: 0.1675. Min Range: -0.1675

Max Act: 0.1262. Min Act: 0.0001

Data Index: 325263 (C4 (Web Text))

Max Activating Token Index: 493