Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1452 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1606. Min Act: -0.0002

Data Index: 1581872 (Python Code)

Max Activating Token Index: 676

Text #1

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1664. Min Act: -0.0002

Data Index: 1610091 (Python Code)

Max Activating Token Index: 743

Text #2

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1662. Min Act: -0.0002

Data Index: 1141867 (C4 (Web Text))

Max Activating Token Index: 832

Text #3

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1546. Min Act: -0.0002

Data Index: 1389440 (Python Code)

Max Activating Token Index: 792

Text #4

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1583. Min Act: -0.0002

Data Index: 1501457 (Python Code)

Max Activating Token Index: 737

Text #5

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1734. Min Act: -0.0002

Data Index: 1471248 (Python Code)

Max Activating Token Index: 562

Text #6

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1499. Min Act: -0.0002

Data Index: 422526 (C4 (Web Text))

Max Activating Token Index: 921

Text #7

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1573. Min Act: -0.0002

Data Index: 844770 (C4 (Web Text))

Max Activating Token Index: 940

Text #8

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1450. Min Act: -0.0002

Data Index: 562764 (C4 (Web Text))

Max Activating Token Index: 1000

Text #9

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1482. Min Act: -0.0002

Data Index: 705830 (C4 (Web Text))

Max Activating Token Index: 914

Text #10

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1488. Min Act: -0.0002

Data Index: 452825 (C4 (Web Text))

Max Activating Token Index: 1011

Text #11

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1500. Min Act: -0.0002

Data Index: 1508175 (Python Code)

Max Activating Token Index: 337

Text #12

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1434. Min Act: -0.0002

Data Index: 1445028 (Python Code)

Max Activating Token Index: 131

Text #13

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1376. Min Act: -0.0002

Data Index: 945530 (C4 (Web Text))

Max Activating Token Index: 927

Text #14

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1520. Min Act: -0.0002

Data Index: 1557929 (Python Code)

Max Activating Token Index: 757

Text #15

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1419. Min Act: -0.0002

Data Index: 1436416 (Python Code)

Max Activating Token Index: 280

Text #16

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1416. Min Act: -0.0002

Data Index: 1106784 (C4 (Web Text))

Max Activating Token Index: 905

Text #17

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1580. Min Act: -0.0002

Data Index: 1569766 (Python Code)

Max Activating Token Index: 929

Text #18

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1534. Min Act: -0.0002

Data Index: 1163991 (C4 (Web Text))

Max Activating Token Index: 973

Text #19

Max Range: 0.1734. Min Range: -0.1734

Max Act: 0.1446. Min Act: -0.0002

Data Index: 6039 (C4 (Web Text))

Max Activating Token Index: 735