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

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

Neuron 386 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.1121. Min Act: -0.0000

Data Index: 873568 (C4 (Web Text))

Max Activating Token Index: 611

Text #1

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.1040. Min Act: -0.0000

Data Index: 1053284 (C4 (Web Text))

Max Activating Token Index: 241

Text #2

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.1031. Min Act: -0.0000

Data Index: 62838 (C4 (Web Text))

Max Activating Token Index: 200

Text #3

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0889. Min Act: -0.0000

Data Index: 443818 (C4 (Web Text))

Max Activating Token Index: 880

Text #4

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0809. Min Act: -0.0000

Data Index: 846007 (C4 (Web Text))

Max Activating Token Index: 520

Text #5

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0789. Min Act: -0.0000

Data Index: 260609 (C4 (Web Text))

Max Activating Token Index: 607

Text #6

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0782. Min Act: -0.0000

Data Index: 968378 (C4 (Web Text))

Max Activating Token Index: 386

Text #7

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0706. Min Act: -0.0000

Data Index: 564803 (C4 (Web Text))

Max Activating Token Index: 727

Text #8

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0732. Min Act: -0.0000

Data Index: 604891 (C4 (Web Text))

Max Activating Token Index: 970

Text #9

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0702. Min Act: -0.0000

Data Index: 1368444 (Python Code)

Max Activating Token Index: 908

Text #10

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0679. Min Act: -0.0000

Data Index: 1240857 (C4 (Web Text))

Max Activating Token Index: 899

Text #11

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0696. Min Act: -0.0000

Data Index: 595210 (C4 (Web Text))

Max Activating Token Index: 662

Text #12

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0699. Min Act: -0.0000

Data Index: 468735 (C4 (Web Text))

Max Activating Token Index: 402

Text #13

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0695. Min Act: -0.0000

Data Index: 91662 (C4 (Web Text))

Max Activating Token Index: 833

Text #14

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0646. Min Act: -0.0000

Data Index: 1363698 (Python Code)

Max Activating Token Index: 995

Text #15

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0668. Min Act: 0.0000

Data Index: 955468 (C4 (Web Text))

Max Activating Token Index: 899

Text #16

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0694. Min Act: -0.0000

Data Index: 103210 (C4 (Web Text))

Max Activating Token Index: 193

Text #17

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0662. Min Act: -0.0000

Data Index: 1322418 (C4 (Web Text))

Max Activating Token Index: 335

Text #18

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0667. Min Act: -0.0000

Data Index: 424304 (C4 (Web Text))

Max Activating Token Index: 748

Text #19

Max Range: 0.1121. Min Range: -0.1121

Max Act: 0.0682. Min Act: -0.0000

Data Index: 754120 (C4 (Web Text))

Max Activating Token Index: 1000