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

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

Neuron 1526 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0558. Min Act: -0.0000

Data Index: 620883 (C4 (Web Text))

Max Activating Token Index: 319

Text #1

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0571. Min Act: -0.0000

Data Index: 582398 (C4 (Web Text))

Max Activating Token Index: 710

Text #2

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0462. Min Act: -0.0001

Data Index: 385145 (C4 (Web Text))

Max Activating Token Index: 863

Text #3

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0448. Min Act: -0.0000

Data Index: 1177398 (C4 (Web Text))

Max Activating Token Index: 34

Text #4

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0471. Min Act: -0.0000

Data Index: 917134 (C4 (Web Text))

Max Activating Token Index: 516

Text #5

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0448. Min Act: -0.0000

Data Index: 1467513 (Python Code)

Max Activating Token Index: 789

Text #6

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0463. Min Act: -0.0000

Data Index: 709068 (C4 (Web Text))

Max Activating Token Index: 157

Text #7

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0443. Min Act: -0.0000

Data Index: 1140611 (C4 (Web Text))

Max Activating Token Index: 967

Text #8

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0421. Min Act: -0.0001

Data Index: 1429422 (Python Code)

Max Activating Token Index: 562

Text #9

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0422. Min Act: -0.0001

Data Index: 1533884 (Python Code)

Max Activating Token Index: 413

Text #10

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0373. Min Act: -0.0001

Data Index: 618186 (C4 (Web Text))

Max Activating Token Index: 580

Text #11

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0421. Min Act: -0.0001

Data Index: 1586421 (Python Code)

Max Activating Token Index: 711

Text #12

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0435. Min Act: -0.0001

Data Index: 1138469 (C4 (Web Text))

Max Activating Token Index: 204

Text #13

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0433. Min Act: -0.0001

Data Index: 1462096 (Python Code)

Max Activating Token Index: 412

Text #14

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0447. Min Act: -0.0001

Data Index: 1446005 (Python Code)

Max Activating Token Index: 256

Text #15

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0430. Min Act: -0.0000

Data Index: 1486226 (Python Code)

Max Activating Token Index: 35

Text #16

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0416. Min Act: -0.0001

Data Index: 1526564 (Python Code)

Max Activating Token Index: 150

Text #17

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0428. Min Act: -0.0001

Data Index: 1306187 (C4 (Web Text))

Max Activating Token Index: 561

Text #18

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0412. Min Act: -0.0001

Data Index: 1629072 (Python Code)

Max Activating Token Index: 99

Text #19

Max Range: 0.0571. Min Range: -0.0571

Max Act: 0.0421. Min Act: -0.0000

Data Index: 716964 (C4 (Web Text))

Max Activating Token Index: 643