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

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

Neuron 2470 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1877. Min Act: -0.0002

Data Index: 656510 (C4 (Web Text))

Max Activating Token Index: 89

Text #1

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1736. Min Act: -0.0002

Data Index: 443110 (C4 (Web Text))

Max Activating Token Index: 660

Text #2

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1610. Min Act: -0.0001

Data Index: 974941 (C4 (Web Text))

Max Activating Token Index: 651

Text #3

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1623. Min Act: -0.0002

Data Index: 682925 (C4 (Web Text))

Max Activating Token Index: 720

Text #4

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1515. Min Act: -0.0002

Data Index: 1128748 (C4 (Web Text))

Max Activating Token Index: 661

Text #5

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1551. Min Act: -0.0002

Data Index: 1230916 (C4 (Web Text))

Max Activating Token Index: 616

Text #6

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1490. Min Act: -0.0001

Data Index: 897766 (C4 (Web Text))

Max Activating Token Index: 801

Text #7

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1442. Min Act: -0.0002

Data Index: 1009427 (C4 (Web Text))

Max Activating Token Index: 539

Text #8

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1345. Min Act: -0.0002

Data Index: 1097344 (C4 (Web Text))

Max Activating Token Index: 679

Text #9

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1286. Min Act: -0.0002

Data Index: 123977 (C4 (Web Text))

Max Activating Token Index: 367

Text #10

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1375. Min Act: -0.0002

Data Index: 315886 (C4 (Web Text))

Max Activating Token Index: 439

Text #11

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1126. Min Act: -0.0002

Data Index: 118532 (C4 (Web Text))

Max Activating Token Index: 883

Text #12

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1234. Min Act: -0.0001

Data Index: 1142709 (C4 (Web Text))

Max Activating Token Index: 816

Text #13

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1115. Min Act: -0.0001

Data Index: 1475579 (Python Code)

Max Activating Token Index: 965

Text #14

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1149. Min Act: -0.0002

Data Index: 1626876 (Python Code)

Max Activating Token Index: 533

Text #15

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.0967. Min Act: -0.0002

Data Index: 776861 (C4 (Web Text))

Max Activating Token Index: 360

Text #16

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1083. Min Act: -0.0001

Data Index: 1387338 (Python Code)

Max Activating Token Index: 931

Text #17

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1166. Min Act: -0.0001

Data Index: 1495793 (Python Code)

Max Activating Token Index: 719

Text #18

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1162. Min Act: -0.0002

Data Index: 1130242 (C4 (Web Text))

Max Activating Token Index: 737

Text #19

Max Range: 0.1877. Min Range: -0.1877

Max Act: 0.1162. Min Act: -0.0001

Data Index: 1650546 (Python Code)

Max Activating Token Index: 292