Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3425 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1948. Min Act: -0.0001

Data Index: 197242 (C4 (Web Text))

Max Activating Token Index: 427

Text #1

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1856. Min Act: -0.0001

Data Index: 1601183 (Python Code)

Max Activating Token Index: 232

Text #2

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1869. Min Act: -0.0001

Data Index: 288417 (C4 (Web Text))

Max Activating Token Index: 899

Text #3

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1758. Min Act: -0.0001

Data Index: 952210 (C4 (Web Text))

Max Activating Token Index: 887

Text #4

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1716. Min Act: -0.0001

Data Index: 763250 (C4 (Web Text))

Max Activating Token Index: 77

Text #5

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1736. Min Act: -0.0001

Data Index: 436763 (C4 (Web Text))

Max Activating Token Index: 743

Text #6

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1732. Min Act: -0.0001

Data Index: 1632009 (Python Code)

Max Activating Token Index: 628

Text #7

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1805. Min Act: -0.0001

Data Index: 786509 (C4 (Web Text))

Max Activating Token Index: 386

Text #8

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1772. Min Act: -0.0001

Data Index: 1569615 (Python Code)

Max Activating Token Index: 563

Text #9

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1725. Min Act: -0.0001

Data Index: 970175 (C4 (Web Text))

Max Activating Token Index: 259

Text #10

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1693. Min Act: -0.0001

Data Index: 64581 (C4 (Web Text))

Max Activating Token Index: 457

Text #11

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1706. Min Act: -0.0001

Data Index: 1304711 (C4 (Web Text))

Max Activating Token Index: 928

Text #12

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1693. Min Act: -0.0001

Data Index: 1104398 (C4 (Web Text))

Max Activating Token Index: 86

Text #13

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1514. Min Act: -0.0001

Data Index: 1400197 (Python Code)

Max Activating Token Index: 1010

Text #14

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1566. Min Act: -0.0001

Data Index: 1654283 (Python Code)

Max Activating Token Index: 257

Text #15

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1710. Min Act: -0.0001

Data Index: 1413548 (Python Code)

Max Activating Token Index: 496

Text #16

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1609. Min Act: -0.0001

Data Index: 1028166 (C4 (Web Text))

Max Activating Token Index: 953

Text #17

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1478. Min Act: -0.0001

Data Index: 1579604 (Python Code)

Max Activating Token Index: 989

Text #18

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1718. Min Act: -0.0001

Data Index: 21867 (C4 (Web Text))

Max Activating Token Index: 684

Text #19

Max Range: 0.1948. Min Range: -0.1948

Max Act: 0.1638. Min Act: -0.0001

Data Index: 771523 (C4 (Web Text))

Max Activating Token Index: 281