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

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

Neuron 3075 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0995. Min Act: -0.0000

Data Index: 166698 (C4 (Web Text))

Max Activating Token Index: 926

Text #1

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0877. Min Act: -0.0000

Data Index: 858391 (C4 (Web Text))

Max Activating Token Index: 53

Text #2

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0830. Min Act: -0.0001

Data Index: 186687 (C4 (Web Text))

Max Activating Token Index: 425

Text #3

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0742. Min Act: -0.0000

Data Index: 863827 (C4 (Web Text))

Max Activating Token Index: 228

Text #4

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0741. Min Act: -0.0000

Data Index: 307749 (C4 (Web Text))

Max Activating Token Index: 936

Text #5

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0725. Min Act: -0.0000

Data Index: 1137882 (C4 (Web Text))

Max Activating Token Index: 931

Text #6

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0735. Min Act: -0.0001

Data Index: 1455032 (Python Code)

Max Activating Token Index: 415

Text #7

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0721. Min Act: -0.0001

Data Index: 1144562 (C4 (Web Text))

Max Activating Token Index: 769

Text #8

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0734. Min Act: -0.0001

Data Index: 1645086 (Python Code)

Max Activating Token Index: 47

Text #9

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0693. Min Act: -0.0001

Data Index: 1457311 (Python Code)

Max Activating Token Index: 261

Text #10

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0684. Min Act: -0.0000

Data Index: 1320823 (C4 (Web Text))

Max Activating Token Index: 542

Text #11

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0653. Min Act: -0.0000

Data Index: 1137083 (C4 (Web Text))

Max Activating Token Index: 308

Text #12

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0592. Min Act: -0.0000

Data Index: 477235 (C4 (Web Text))

Max Activating Token Index: 477

Text #13

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0617. Min Act: -0.0000

Data Index: 956462 (C4 (Web Text))

Max Activating Token Index: 980

Text #14

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0652. Min Act: -0.0000

Data Index: 818378 (C4 (Web Text))

Max Activating Token Index: 122

Text #15

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0600. Min Act: -0.0000

Data Index: 1128 (C4 (Web Text))

Max Activating Token Index: 382

Text #16

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0599. Min Act: -0.0000

Data Index: 438551 (C4 (Web Text))

Max Activating Token Index: 345

Text #17

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0629. Min Act: -0.0001

Data Index: 1546922 (Python Code)

Max Activating Token Index: 1002

Text #18

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0603. Min Act: -0.0000

Data Index: 1345965 (C4 (Web Text))

Max Activating Token Index: 877

Text #19

Max Range: 0.0995. Min Range: -0.0995

Max Act: 0.0653. Min Act: -0.0000

Data Index: 1215110 (C4 (Web Text))

Max Activating Token Index: 154