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

Dataset: The Pile

Neuron 4004 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: 8.4021. Min Range: -8.4021

Max Act: 8.1961. Min Act: -0.0000

Data Index: 637695 (The Pile)

Max Activating Token Index: 226

Text #1

Max Range: 8.4021. Min Range: -8.4021

Max Act: 8.3512. Min Act: -0.0000

Data Index: 1334025 (The Pile)

Max Activating Token Index: 134

Text #2

Max Range: 8.4021. Min Range: -8.4021

Max Act: 8.4021. Min Act: -0.0000

Data Index: 173450 (The Pile)

Max Activating Token Index: 46

Text #3

Max Range: 8.4021. Min Range: -8.4021

Max Act: 8.2736. Min Act: -0.0000

Data Index: 749708 (The Pile)

Max Activating Token Index: 93

Text #4

Max Range: 8.4021. Min Range: -8.4021

Max Act: 8.2241. Min Act: -0.0000

Data Index: 1915576 (The Pile)

Max Activating Token Index: 298

Text #5

Max Range: 8.4021. Min Range: -8.4021

Max Act: 8.1840. Min Act: -0.0000

Data Index: 499466 (The Pile)

Max Activating Token Index: 451

Text #6

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.9478. Min Act: -0.0000

Data Index: 852611 (The Pile)

Max Activating Token Index: 262

Text #7

Max Range: 8.4021. Min Range: -8.4021

Max Act: 8.0539. Min Act: -0.0000

Data Index: 765558 (The Pile)

Max Activating Token Index: 694

Text #8

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.8970. Min Act: -0.0000

Data Index: 1838277 (The Pile)

Max Activating Token Index: 19

Text #9

Max Range: 8.4021. Min Range: -8.4021

Max Act: 8.1500. Min Act: -0.0000

Data Index: 1479182 (The Pile)

Max Activating Token Index: 250

Text #10

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.6222. Min Act: -0.0000

Data Index: 1179371 (The Pile)

Max Activating Token Index: 782

Text #11

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.7690. Min Act: -0.0000

Data Index: 118105 (The Pile)

Max Activating Token Index: 338

Text #12

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.9524. Min Act: -0.0000

Data Index: 1503955 (The Pile)

Max Activating Token Index: 9

Text #13

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.9424. Min Act: -0.0000

Data Index: 214018 (The Pile)

Max Activating Token Index: 634

Text #14

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.8468. Min Act: -0.0000

Data Index: 1554729 (The Pile)

Max Activating Token Index: 414

Text #15

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.8367. Min Act: -0.0000

Data Index: 1797961 (The Pile)

Max Activating Token Index: 150

Text #16

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.6418. Min Act: -0.0000

Data Index: 1161853 (The Pile)

Max Activating Token Index: 96

Text #17

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.7965. Min Act: -0.0000

Data Index: 810430 (The Pile)

Max Activating Token Index: 169

Text #18

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.5782. Min Act: -0.0000

Data Index: 631892 (The Pile)

Max Activating Token Index: 139

Text #19

Max Range: 8.4021. Min Range: -8.4021

Max Act: 7.7867. Min Act: -0.0000

Data Index: 902471 (The Pile)

Max Activating Token Index: 18