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

Dataset: The Pile

Neuron 2496 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 10.0898. Min Range: -10.0898

Max Act: 10.0898. Min Act: -0.0000

Data Index: 903812 (The Pile)

Max Activating Token Index: 120

Text #1

Max Range: 10.0898. Min Range: -10.0898

Max Act: 8.7898. Min Act: -0.0000

Data Index: 1635987 (The Pile)

Max Activating Token Index: 129

Text #2

Max Range: 10.0898. Min Range: -10.0898

Max Act: 8.5937. Min Act: -0.0000

Data Index: 1850010 (The Pile)

Max Activating Token Index: 554

Text #3

Max Range: 10.0898. Min Range: -10.0898

Max Act: 7.9810. Min Act: -0.0000

Data Index: 969442 (The Pile)

Max Activating Token Index: 936

Text #4

Max Range: 10.0898. Min Range: -10.0898

Max Act: 7.4632. Min Act: -0.0000

Data Index: 1421572 (The Pile)

Max Activating Token Index: 412

Text #5

Max Range: 10.0898. Min Range: -10.0898

Max Act: 7.4789. Min Act: -0.0000

Data Index: 215626 (The Pile)

Max Activating Token Index: 59

Text #6

Max Range: 10.0898. Min Range: -10.0898

Max Act: 7.5017. Min Act: -0.0000

Data Index: 1985123 (The Pile)

Max Activating Token Index: 679

Text #7

Max Range: 10.0898. Min Range: -10.0898

Max Act: 7.1966. Min Act: -0.0000

Data Index: 706213 (The Pile)

Max Activating Token Index: 29

Text #8

Max Range: 10.0898. Min Range: -10.0898

Max Act: 7.1837. Min Act: -0.0000

Data Index: 579656 (The Pile)

Max Activating Token Index: 846

Text #9

Max Range: 10.0898. Min Range: -10.0898

Max Act: 7.0292. Min Act: -0.0000

Data Index: 1969312 (The Pile)

Max Activating Token Index: 26

Text #10

Max Range: 10.0898. Min Range: -10.0898

Max Act: 6.4487. Min Act: -0.0000

Data Index: 1794851 (The Pile)

Max Activating Token Index: 608

Text #11

Max Range: 10.0898. Min Range: -10.0898

Max Act: 7.1512. Min Act: -0.0000

Data Index: 265080 (The Pile)

Max Activating Token Index: 366

Text #12

Max Range: 10.0898. Min Range: -10.0898

Max Act: 6.9433. Min Act: -0.0000

Data Index: 1357763 (The Pile)

Max Activating Token Index: 271

Text #13

Max Range: 10.0898. Min Range: -10.0898

Max Act: 7.0127. Min Act: -0.0000

Data Index: 1850037 (The Pile)

Max Activating Token Index: 640

Text #14

Max Range: 10.0898. Min Range: -10.0898

Max Act: 6.7169. Min Act: -0.0000

Data Index: 1279623 (The Pile)

Max Activating Token Index: 290

Text #15

Max Range: 10.0898. Min Range: -10.0898

Max Act: 6.4893. Min Act: -0.0000

Data Index: 519394 (The Pile)

Max Activating Token Index: 580

Text #16

Max Range: 10.0898. Min Range: -10.0898

Max Act: 6.6243. Min Act: -0.0000

Data Index: 387123 (The Pile)

Max Activating Token Index: 453

Text #17

Max Range: 10.0898. Min Range: -10.0898

Max Act: 6.6364. Min Act: -0.0000

Data Index: 1320159 (The Pile)

Max Activating Token Index: 27

Text #18

Max Range: 10.0898. Min Range: -10.0898

Max Act: 6.3806. Min Act: -0.0000

Data Index: 601997 (The Pile)

Max Activating Token Index: 90

Text #19

Max Range: 10.0898. Min Range: -10.0898

Max Act: 6.3302. Min Act: -0.0000

Data Index: 57492 (The Pile)

Max Activating Token Index: 700