Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3543 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1653. Min Act: -0.0000

Data Index: 513698 (C4 (Web Text))

Max Activating Token Index: 617

Text #1

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1523. Min Act: -0.0000

Data Index: 611485 (C4 (Web Text))

Max Activating Token Index: 438

Text #2

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1546. Min Act: -0.0000

Data Index: 1108530 (C4 (Web Text))

Max Activating Token Index: 441

Text #3

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1636. Min Act: -0.0000

Data Index: 805552 (C4 (Web Text))

Max Activating Token Index: 464

Text #4

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1565. Min Act: -0.0000

Data Index: 1098117 (C4 (Web Text))

Max Activating Token Index: 107

Text #5

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1533. Min Act: -0.0000

Data Index: 1217284 (C4 (Web Text))

Max Activating Token Index: 623

Text #6

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1466. Min Act: -0.0000

Data Index: 489326 (C4 (Web Text))

Max Activating Token Index: 622

Text #7

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1450. Min Act: -0.0000

Data Index: 636099 (C4 (Web Text))

Max Activating Token Index: 107

Text #8

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1383. Min Act: -0.0000

Data Index: 328394 (C4 (Web Text))

Max Activating Token Index: 680

Text #9

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1365. Min Act: -0.0000

Data Index: 1585066 (Python Code)

Max Activating Token Index: 12

Text #10

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1411. Min Act: -0.0000

Data Index: 338475 (C4 (Web Text))

Max Activating Token Index: 834

Text #11

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1361. Min Act: -0.0000

Data Index: 1495233 (Python Code)

Max Activating Token Index: 572

Text #12

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1290. Min Act: -0.0000

Data Index: 719753 (C4 (Web Text))

Max Activating Token Index: 650

Text #13

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1281. Min Act: -0.0000

Data Index: 1563395 (Python Code)

Max Activating Token Index: 100

Text #14

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1358. Min Act: -0.0000

Data Index: 3625 (C4 (Web Text))

Max Activating Token Index: 553

Text #15

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1222. Min Act: -0.0000

Data Index: 888583 (C4 (Web Text))

Max Activating Token Index: 713

Text #16

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1321. Min Act: -0.0000

Data Index: 1650358 (Python Code)

Max Activating Token Index: 798

Text #17

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1233. Min Act: -0.0000

Data Index: 638312 (C4 (Web Text))

Max Activating Token Index: 27

Text #18

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1323. Min Act: -0.0000

Data Index: 514620 (C4 (Web Text))

Max Activating Token Index: 372

Text #19

Max Range: 0.1653. Min Range: -0.1653

Max Act: 0.1302. Min Act: -0.0000

Data Index: 583829 (C4 (Web Text))

Max Activating Token Index: 636