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

Dataset: The Pile

Neuron 2590 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: 9.2957. Min Range: -9.2957

Max Act: 8.8857. Min Act: -0.0000

Data Index: 1217568 (The Pile)

Max Activating Token Index: 247

Text #1

Max Range: 9.2957. Min Range: -9.2957

Max Act: 9.2957. Min Act: -0.0000

Data Index: 72305 (The Pile)

Max Activating Token Index: 1007

Text #2

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.9763. Min Act: -0.0000

Data Index: 1687155 (The Pile)

Max Activating Token Index: 320

Text #3

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.9227. Min Act: -0.0000

Data Index: 829028 (The Pile)

Max Activating Token Index: 774

Text #4

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.9752. Min Act: -0.0000

Data Index: 576283 (The Pile)

Max Activating Token Index: 789

Text #5

Max Range: 9.2957. Min Range: -9.2957

Max Act: 9.1252. Min Act: -0.0000

Data Index: 980617 (The Pile)

Max Activating Token Index: 224

Text #6

Max Range: 9.2957. Min Range: -9.2957

Max Act: 9.0751. Min Act: -0.0000

Data Index: 97306 (The Pile)

Max Activating Token Index: 868

Text #7

Max Range: 9.2957. Min Range: -9.2957

Max Act: 9.0139. Min Act: -0.0000

Data Index: 32934 (The Pile)

Max Activating Token Index: 754

Text #8

Max Range: 9.2957. Min Range: -9.2957

Max Act: 9.0996. Min Act: -0.0000

Data Index: 1042011 (The Pile)

Max Activating Token Index: 1023

Text #9

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.9337. Min Act: -0.0000

Data Index: 1847166 (The Pile)

Max Activating Token Index: 987

Text #10

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.4261. Min Act: -0.0000

Data Index: 172715 (The Pile)

Max Activating Token Index: 775

Text #11

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.7320. Min Act: -0.0000

Data Index: 1842243 (The Pile)

Max Activating Token Index: 10

Text #12

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.7457. Min Act: -0.0000

Data Index: 1591959 (The Pile)

Max Activating Token Index: 400

Text #13

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.4138. Min Act: -0.0000

Data Index: 709349 (The Pile)

Max Activating Token Index: 419

Text #14

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.7987. Min Act: -0.0000

Data Index: 1250127 (The Pile)

Max Activating Token Index: 1012

Text #15

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.4736. Min Act: -0.0000

Data Index: 331042 (The Pile)

Max Activating Token Index: 505

Text #16

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.5759. Min Act: -0.0000

Data Index: 869441 (The Pile)

Max Activating Token Index: 920

Text #17

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.3486. Min Act: -0.0000

Data Index: 217821 (The Pile)

Max Activating Token Index: 31

Text #18

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.3504. Min Act: -0.0000

Data Index: 183678 (The Pile)

Max Activating Token Index: 819

Text #19

Max Range: 9.2957. Min Range: -9.2957

Max Act: 8.0512. Min Act: -0.0000

Data Index: 921972 (The Pile)

Max Activating Token Index: 848