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

Dataset: The Pile

Neuron 4286 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: 7.6960. Min Range: -7.6960

Max Act: 7.6960. Min Act: -0.0000

Data Index: 1949676 (The Pile)

Max Activating Token Index: 566

Text #1

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.5075. Min Act: -0.0000

Data Index: 1638371 (The Pile)

Max Activating Token Index: 952

Text #2

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.5088. Min Act: -0.0000

Data Index: 227464 (The Pile)

Max Activating Token Index: 994

Text #3

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.3268. Min Act: -0.0000

Data Index: 1450512 (The Pile)

Max Activating Token Index: 1002

Text #4

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.4618. Min Act: -0.0000

Data Index: 173325 (The Pile)

Max Activating Token Index: 991

Text #5

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.1602. Min Act: -0.0000

Data Index: 1386817 (The Pile)

Max Activating Token Index: 999

Text #6

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.5017. Min Act: -0.0000

Data Index: 1765747 (The Pile)

Max Activating Token Index: 1018

Text #7

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.1323. Min Act: -0.0000

Data Index: 745555 (The Pile)

Max Activating Token Index: 1014

Text #8

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.3068. Min Act: -0.0000

Data Index: 1192323 (The Pile)

Max Activating Token Index: 1017

Text #9

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.2902. Min Act: -0.0000

Data Index: 617722 (The Pile)

Max Activating Token Index: 983

Text #10

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.1638. Min Act: -0.0000

Data Index: 380121 (The Pile)

Max Activating Token Index: 810

Text #11

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.1546. Min Act: -0.0000

Data Index: 663087 (The Pile)

Max Activating Token Index: 999

Text #12

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.1612. Min Act: -0.0000

Data Index: 1987700 (The Pile)

Max Activating Token Index: 960

Text #13

Max Range: 7.6960. Min Range: -7.6960

Max Act: 6.9767. Min Act: -0.0000

Data Index: 850950 (The Pile)

Max Activating Token Index: 994

Text #14

Max Range: 7.6960. Min Range: -7.6960

Max Act: 6.9647. Min Act: -0.0000

Data Index: 1552598 (The Pile)

Max Activating Token Index: 853

Text #15

Max Range: 7.6960. Min Range: -7.6960

Max Act: 6.8617. Min Act: -0.0000

Data Index: 1145862 (The Pile)

Max Activating Token Index: 751

Text #16

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.0220. Min Act: -0.0000

Data Index: 1533687 (The Pile)

Max Activating Token Index: 756

Text #17

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.0291. Min Act: -0.0000

Data Index: 1084603 (The Pile)

Max Activating Token Index: 1003

Text #18

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.0165. Min Act: -0.0000

Data Index: 738996 (The Pile)

Max Activating Token Index: 316

Text #19

Max Range: 7.6960. Min Range: -7.6960

Max Act: 7.0110. Min Act: -0.0000

Data Index: 1128726 (The Pile)

Max Activating Token Index: 947