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

Dataset: The Pile

Neuron 198 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.2640. Min Range: -7.2640

Max Act: 7.2640. Min Act: -0.0000

Data Index: 1392630 (The Pile)

Max Activating Token Index: 343

Text #1

Max Range: 7.2640. Min Range: -7.2640

Max Act: 6.1912. Min Act: -0.0000

Data Index: 250477 (The Pile)

Max Activating Token Index: 810

Text #2

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.8616. Min Act: -0.0000

Data Index: 1300403 (The Pile)

Max Activating Token Index: 9

Text #3

Max Range: 7.2640. Min Range: -7.2640

Max Act: 4.9980. Min Act: -0.0000

Data Index: 724644 (The Pile)

Max Activating Token Index: 973

Text #4

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.1062. Min Act: -0.0000

Data Index: 911840 (The Pile)

Max Activating Token Index: 222

Text #5

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.4259. Min Act: -0.0000

Data Index: 67884 (The Pile)

Max Activating Token Index: 876

Text #6

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.1988. Min Act: -0.0000

Data Index: 890636 (The Pile)

Max Activating Token Index: 514

Text #7

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.3928. Min Act: -0.0000

Data Index: 406605 (The Pile)

Max Activating Token Index: 719

Text #8

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.3443. Min Act: -0.0000

Data Index: 842952 (The Pile)

Max Activating Token Index: 908

Text #9

Max Range: 7.2640. Min Range: -7.2640

Max Act: 4.9554. Min Act: -0.0000

Data Index: 1919413 (The Pile)

Max Activating Token Index: 744

Text #10

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.2878. Min Act: -0.0000

Data Index: 1786011 (The Pile)

Max Activating Token Index: 1009

Text #11

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.0598. Min Act: 0.0001

Data Index: 381420 (The Pile)

Max Activating Token Index: 254

Text #12

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.0495. Min Act: -0.0000

Data Index: 944829 (The Pile)

Max Activating Token Index: 517

Text #13

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.0541. Min Act: -0.0000

Data Index: 385647 (The Pile)

Max Activating Token Index: 384

Text #14

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.1971. Min Act: -0.0000

Data Index: 1598233 (The Pile)

Max Activating Token Index: 93

Text #15

Max Range: 7.2640. Min Range: -7.2640

Max Act: 4.8217. Min Act: -0.0000

Data Index: 952530 (The Pile)

Max Activating Token Index: 828

Text #16

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.0529. Min Act: -0.0000

Data Index: 1211623 (The Pile)

Max Activating Token Index: 29

Text #17

Max Range: 7.2640. Min Range: -7.2640

Max Act: 4.6748. Min Act: -0.0000

Data Index: 96892 (The Pile)

Max Activating Token Index: 968

Text #18

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.0078. Min Act: 0.0000

Data Index: 1276548 (The Pile)

Max Activating Token Index: 64

Text #19

Max Range: 7.2640. Min Range: -7.2640

Max Act: 5.1776. Min Act: -0.0000

Data Index: 1041935 (The Pile)

Max Activating Token Index: 766