Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4802 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 2.3420. Min Range: -2.3420

Max Act: 2.3420. Min Act: 0.0000

Data Index: 130485 (C4 (Web Text))

Max Activating Token Index: 474

Text #1

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.7914. Min Act: -0.0000

Data Index: 1402689 (Python Code)

Max Activating Token Index: 1014

Text #2

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.8424. Min Act: -0.0000

Data Index: 707306 (C4 (Web Text))

Max Activating Token Index: 32

Text #3

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.3568. Min Act: -0.0000

Data Index: 63794 (C4 (Web Text))

Max Activating Token Index: 122

Text #4

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.3139. Min Act: -0.0000

Data Index: 1357482 (C4 (Web Text))

Max Activating Token Index: 878

Text #5

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.2680. Min Act: -0.0000

Data Index: 823220 (C4 (Web Text))

Max Activating Token Index: 99

Text #6

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.3527. Min Act: -0.0000

Data Index: 1054868 (C4 (Web Text))

Max Activating Token Index: 69

Text #7

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.2162. Min Act: -0.0000

Data Index: 96689 (C4 (Web Text))

Max Activating Token Index: 938

Text #8

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.1902. Min Act: -0.0000

Data Index: 739978 (C4 (Web Text))

Max Activating Token Index: 277

Text #9

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.2614. Min Act: 0.0000

Data Index: 732047 (C4 (Web Text))

Max Activating Token Index: 737

Text #10

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.2062. Min Act: -0.0000

Data Index: 1560467 (Python Code)

Max Activating Token Index: 832

Text #11

Max Range: 2.3420. Min Range: -2.3420

Max Act: 0.9503. Min Act: -0.0000

Data Index: 1013856 (C4 (Web Text))

Max Activating Token Index: 881

Text #12

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.1441. Min Act: -0.0000

Data Index: 270655 (C4 (Web Text))

Max Activating Token Index: 534

Text #13

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.0398. Min Act: -0.0000

Data Index: 1631872 (Python Code)

Max Activating Token Index: 281

Text #14

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.0388. Min Act: 0.0000

Data Index: 674554 (C4 (Web Text))

Max Activating Token Index: 924

Text #15

Max Range: 2.3420. Min Range: -2.3420

Max Act: 0.9892. Min Act: -0.0000

Data Index: 337709 (C4 (Web Text))

Max Activating Token Index: 977

Text #16

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.1708. Min Act: -0.0000

Data Index: 1337387 (C4 (Web Text))

Max Activating Token Index: 928

Text #17

Max Range: 2.3420. Min Range: -2.3420

Max Act: 0.9305. Min Act: -0.0000

Data Index: 534116 (C4 (Web Text))

Max Activating Token Index: 435

Text #18

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.0181. Min Act: -0.0000

Data Index: 704510 (C4 (Web Text))

Max Activating Token Index: 444

Text #19

Max Range: 2.3420. Min Range: -2.3420

Max Act: 1.0736. Min Act: -0.0000

Data Index: 255986 (C4 (Web Text))

Max Activating Token Index: 417