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

Dataset: The Pile

Neuron 630 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: 7.2695. Min Range: -7.2695

Max Act: 7.0407. Min Act: -0.0000

Data Index: 1992773 (The Pile)

Max Activating Token Index: 551

Text #1

Max Range: 7.2695. Min Range: -7.2695

Max Act: 7.2695. Min Act: -0.0000

Data Index: 66249 (The Pile)

Max Activating Token Index: 585

Text #2

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.7542. Min Act: -0.0000

Data Index: 1179371 (The Pile)

Max Activating Token Index: 708

Text #3

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.6760. Min Act: -0.0000

Data Index: 1341885 (The Pile)

Max Activating Token Index: 208

Text #4

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.6272. Min Act: -0.0000

Data Index: 1621806 (The Pile)

Max Activating Token Index: 960

Text #5

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.4717. Min Act: -0.0000

Data Index: 1854851 (The Pile)

Max Activating Token Index: 734

Text #6

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.3513. Min Act: -0.0000

Data Index: 618855 (The Pile)

Max Activating Token Index: 806

Text #7

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.1533. Min Act: -0.0000

Data Index: 240156 (The Pile)

Max Activating Token Index: 746

Text #8

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.2199. Min Act: -0.0000

Data Index: 565770 (The Pile)

Max Activating Token Index: 796

Text #9

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.1435. Min Act: -0.0000

Data Index: 1352244 (The Pile)

Max Activating Token Index: 700

Text #10

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.1508. Min Act: -0.0000

Data Index: 1359858 (The Pile)

Max Activating Token Index: 189

Text #11

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.0032. Min Act: -0.0000

Data Index: 1107437 (The Pile)

Max Activating Token Index: 700

Text #12

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.0929. Min Act: -0.0000

Data Index: 1811035 (The Pile)

Max Activating Token Index: 311

Text #13

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.2743. Min Act: -0.0000

Data Index: 1213033 (The Pile)

Max Activating Token Index: 472

Text #14

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.0836. Min Act: -0.0000

Data Index: 581947 (The Pile)

Max Activating Token Index: 327

Text #15

Max Range: 7.2695. Min Range: -7.2695

Max Act: 5.8703. Min Act: -0.0000

Data Index: 382159 (The Pile)

Max Activating Token Index: 467

Text #16

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.2068. Min Act: -0.0000

Data Index: 174322 (The Pile)

Max Activating Token Index: 139

Text #17

Max Range: 7.2695. Min Range: -7.2695

Max Act: 5.5971. Min Act: -0.0000

Data Index: 199845 (The Pile)

Max Activating Token Index: 928

Text #18

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.1280. Min Act: -0.0000

Data Index: 1205981 (The Pile)

Max Activating Token Index: 882

Text #19

Max Range: 7.2695. Min Range: -7.2695

Max Act: 6.1015. Min Act: -0.0000

Data Index: 376235 (The Pile)

Max Activating Token Index: 195