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

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

Neuron 1492 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: 0.2446. Min Range: -0.2446

Max Act: 0.2446. Min Act: -0.0002

Data Index: 565541 (C4 (Web Text))

Max Activating Token Index: 831

Text #1

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1910. Min Act: -0.0001

Data Index: 375158 (C4 (Web Text))

Max Activating Token Index: 931

Text #2

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1790. Min Act: -0.0002

Data Index: 1196996 (C4 (Web Text))

Max Activating Token Index: 343

Text #3

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1847. Min Act: -0.0002

Data Index: 1486748 (Python Code)

Max Activating Token Index: 613

Text #4

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1706. Min Act: -0.0002

Data Index: 1606687 (Python Code)

Max Activating Token Index: 923

Text #5

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1755. Min Act: -0.0002

Data Index: 730338 (C4 (Web Text))

Max Activating Token Index: 959

Text #6

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1576. Min Act: -0.0002

Data Index: 1519462 (Python Code)

Max Activating Token Index: 349

Text #7

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1740. Min Act: -0.0002

Data Index: 678817 (C4 (Web Text))

Max Activating Token Index: 879

Text #8

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1712. Min Act: -0.0002

Data Index: 667351 (C4 (Web Text))

Max Activating Token Index: 735

Text #9

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1724. Min Act: -0.0002

Data Index: 673457 (C4 (Web Text))

Max Activating Token Index: 966

Text #10

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1675. Min Act: -0.0002

Data Index: 495826 (C4 (Web Text))

Max Activating Token Index: 550

Text #11

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1713. Min Act: -0.0002

Data Index: 215676 (C4 (Web Text))

Max Activating Token Index: 478

Text #12

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1702. Min Act: -0.0002

Data Index: 1269454 (C4 (Web Text))

Max Activating Token Index: 460

Text #13

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1657. Min Act: -0.0002

Data Index: 637305 (C4 (Web Text))

Max Activating Token Index: 646

Text #14

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1621. Min Act: -0.0002

Data Index: 1542713 (Python Code)

Max Activating Token Index: 423

Text #15

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1623. Min Act: -0.0002

Data Index: 363996 (C4 (Web Text))

Max Activating Token Index: 359

Text #16

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1614. Min Act: -0.0002

Data Index: 846699 (C4 (Web Text))

Max Activating Token Index: 84

Text #17

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1559. Min Act: -0.0002

Data Index: 350738 (C4 (Web Text))

Max Activating Token Index: 605

Text #18

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1661. Min Act: -0.0002

Data Index: 949039 (C4 (Web Text))

Max Activating Token Index: 247

Text #19

Max Range: 0.2446. Min Range: -0.2446

Max Act: 0.1678. Min Act: -0.0002

Data Index: 1380412 (Python Code)

Max Activating Token Index: 223