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

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

Neuron 1929 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0949. Min Act: -0.0001

Data Index: 1090987 (C4 (Web Text))

Max Activating Token Index: 875

Text #1

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0949. Min Act: -0.0001

Data Index: 1362011 (Python Code)

Max Activating Token Index: 56

Text #2

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0901. Min Act: -0.0001

Data Index: 868406 (C4 (Web Text))

Max Activating Token Index: 825

Text #3

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0826. Min Act: -0.0001

Data Index: 753427 (C4 (Web Text))

Max Activating Token Index: 613

Text #4

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0759. Min Act: -0.0001

Data Index: 1463150 (Python Code)

Max Activating Token Index: 411

Text #5

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0835. Min Act: -0.0001

Data Index: 760266 (C4 (Web Text))

Max Activating Token Index: 699

Text #6

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0776. Min Act: -0.0001

Data Index: 742097 (C4 (Web Text))

Max Activating Token Index: 476

Text #7

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0726. Min Act: -0.0001

Data Index: 1518821 (Python Code)

Max Activating Token Index: 396

Text #8

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0734. Min Act: -0.0001

Data Index: 1065968 (C4 (Web Text))

Max Activating Token Index: 133

Text #9

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0713. Min Act: -0.0001

Data Index: 840469 (C4 (Web Text))

Max Activating Token Index: 643

Text #10

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0729. Min Act: -0.0001

Data Index: 524048 (C4 (Web Text))

Max Activating Token Index: 992

Text #11

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0769. Min Act: -0.0001

Data Index: 501300 (C4 (Web Text))

Max Activating Token Index: 890

Text #12

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0680. Min Act: -0.0001

Data Index: 684533 (C4 (Web Text))

Max Activating Token Index: 632

Text #13

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0691. Min Act: -0.0001

Data Index: 458013 (C4 (Web Text))

Max Activating Token Index: 979

Text #14

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0630. Min Act: -0.0001

Data Index: 1573425 (Python Code)

Max Activating Token Index: 472

Text #15

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0629. Min Act: -0.0001

Data Index: 921923 (C4 (Web Text))

Max Activating Token Index: 884

Text #16

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0654. Min Act: -0.0001

Data Index: 1322948 (C4 (Web Text))

Max Activating Token Index: 810

Text #17

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0621. Min Act: -0.0001

Data Index: 1617458 (Python Code)

Max Activating Token Index: 975

Text #18

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0644. Min Act: -0.0001

Data Index: 1543097 (Python Code)

Max Activating Token Index: 992

Text #19

Max Range: 0.0949. Min Range: -0.0949

Max Act: 0.0647. Min Act: -0.0001

Data Index: 509105 (C4 (Web Text))

Max Activating Token Index: 732