Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 2851 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0889. Min Act: -0.0001

Data Index: 598813 (C4 (Web Text))

Max Activating Token Index: 29

Text #1

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0806. Min Act: -0.0000

Data Index: 64791 (C4 (Web Text))

Max Activating Token Index: 379

Text #2

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0774. Min Act: -0.0000

Data Index: 147005 (C4 (Web Text))

Max Activating Token Index: 901

Text #3

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0791. Min Act: -0.0000

Data Index: 435666 (C4 (Web Text))

Max Activating Token Index: 69

Text #4

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0824. Min Act: -0.0000

Data Index: 1265384 (C4 (Web Text))

Max Activating Token Index: 878

Text #5

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0749. Min Act: -0.0001

Data Index: 253186 (C4 (Web Text))

Max Activating Token Index: 978

Text #6

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0769. Min Act: -0.0001

Data Index: 169945 (C4 (Web Text))

Max Activating Token Index: 839

Text #7

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0775. Min Act: -0.0000

Data Index: 1509306 (Python Code)

Max Activating Token Index: 865

Text #8

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0691. Min Act: -0.0000

Data Index: 939834 (C4 (Web Text))

Max Activating Token Index: 427

Text #9

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0769. Min Act: -0.0000

Data Index: 1548988 (Python Code)

Max Activating Token Index: 948

Text #10

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0728. Min Act: -0.0000

Data Index: 1315035 (C4 (Web Text))

Max Activating Token Index: 990

Text #11

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0793. Min Act: -0.0000

Data Index: 1640683 (Python Code)

Max Activating Token Index: 104

Text #12

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0668. Min Act: -0.0001

Data Index: 30495 (C4 (Web Text))

Max Activating Token Index: 198

Text #13

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0705. Min Act: -0.0000

Data Index: 637472 (C4 (Web Text))

Max Activating Token Index: 172

Text #14

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0752. Min Act: -0.0000

Data Index: 1475180 (Python Code)

Max Activating Token Index: 60

Text #15

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0760. Min Act: -0.0000

Data Index: 43146 (C4 (Web Text))

Max Activating Token Index: 963

Text #16

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0768. Min Act: -0.0001

Data Index: 394688 (C4 (Web Text))

Max Activating Token Index: 106

Text #17

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0732. Min Act: -0.0000

Data Index: 1522766 (Python Code)

Max Activating Token Index: 401

Text #18

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0731. Min Act: -0.0000

Data Index: 1026916 (C4 (Web Text))

Max Activating Token Index: 637

Text #19

Max Range: 0.0889. Min Range: -0.0889

Max Act: 0.0746. Min Act: -0.0000

Data Index: 1078339 (C4 (Web Text))

Max Activating Token Index: 627