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

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

Neuron 4779 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.2495. Min Act: -0.0000

Data Index: 801272 (C4 (Web Text))

Max Activating Token Index: 635

Text #1

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1920. Min Act: -0.0000

Data Index: 582516 (C4 (Web Text))

Max Activating Token Index: 410

Text #2

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.2030. Min Act: -0.0000

Data Index: 476770 (C4 (Web Text))

Max Activating Token Index: 589

Text #3

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1749. Min Act: -0.0000

Data Index: 91216 (C4 (Web Text))

Max Activating Token Index: 454

Text #4

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1731. Min Act: -0.0000

Data Index: 1167792 (C4 (Web Text))

Max Activating Token Index: 103

Text #5

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1687. Min Act: -0.0000

Data Index: 1642875 (Python Code)

Max Activating Token Index: 603

Text #6

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1598. Min Act: -0.0000

Data Index: 1282358 (C4 (Web Text))

Max Activating Token Index: 780

Text #7

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1573. Min Act: -0.0000

Data Index: 815685 (C4 (Web Text))

Max Activating Token Index: 103

Text #8

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1640. Min Act: -0.0000

Data Index: 851110 (C4 (Web Text))

Max Activating Token Index: 449

Text #9

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1608. Min Act: -0.0000

Data Index: 1080074 (C4 (Web Text))

Max Activating Token Index: 258

Text #10

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1412. Min Act: -0.0000

Data Index: 711858 (C4 (Web Text))

Max Activating Token Index: 685

Text #11

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1615. Min Act: -0.0000

Data Index: 715364 (C4 (Web Text))

Max Activating Token Index: 100

Text #12

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1611. Min Act: -0.0000

Data Index: 607957 (C4 (Web Text))

Max Activating Token Index: 693

Text #13

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1712. Min Act: -0.0000

Data Index: 1503292 (Python Code)

Max Activating Token Index: 159

Text #14

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1526. Min Act: -0.0000

Data Index: 247214 (C4 (Web Text))

Max Activating Token Index: 156

Text #15

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1578. Min Act: -0.0000

Data Index: 1024390 (C4 (Web Text))

Max Activating Token Index: 175

Text #16

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1686. Min Act: -0.0000

Data Index: 1486916 (Python Code)

Max Activating Token Index: 191

Text #17

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1555. Min Act: -0.0000

Data Index: 329600 (C4 (Web Text))

Max Activating Token Index: 621

Text #18

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1483. Min Act: -0.0000

Data Index: 929595 (C4 (Web Text))

Max Activating Token Index: 37

Text #19

Max Range: 0.2495. Min Range: -0.2495

Max Act: 0.1326. Min Act: -0.0000

Data Index: 918905 (C4 (Web Text))

Max Activating Token Index: 1002