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

Dataset: The Pile

Neuron 1026 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: 8.7701. Min Range: -8.7701

Max Act: 8.7543. Min Act: -0.0000

Data Index: 1681675 (The Pile)

Max Activating Token Index: 387

Text #1

Max Range: 8.7701. Min Range: -8.7701

Max Act: 8.5398. Min Act: -0.0000

Data Index: 183280 (The Pile)

Max Activating Token Index: 491

Text #2

Max Range: 8.7701. Min Range: -8.7701

Max Act: 8.4433. Min Act: -0.0000

Data Index: 498402 (The Pile)

Max Activating Token Index: 676

Text #3

Max Range: 8.7701. Min Range: -8.7701

Max Act: 8.7701. Min Act: -0.0000

Data Index: 1676694 (The Pile)

Max Activating Token Index: 911

Text #4

Max Range: 8.7701. Min Range: -8.7701

Max Act: 8.4169. Min Act: -0.0000

Data Index: 190459 (The Pile)

Max Activating Token Index: 350

Text #5

Max Range: 8.7701. Min Range: -8.7701

Max Act: 8.1874. Min Act: -0.0000

Data Index: 323535 (The Pile)

Max Activating Token Index: 439

Text #6

Max Range: 8.7701. Min Range: -8.7701

Max Act: 8.6473. Min Act: -0.0000

Data Index: 1648478 (The Pile)

Max Activating Token Index: 653

Text #7

Max Range: 8.7701. Min Range: -8.7701

Max Act: 7.9738. Min Act: -0.0000

Data Index: 1293399 (The Pile)

Max Activating Token Index: 375

Text #8

Max Range: 8.7701. Min Range: -8.7701

Max Act: 8.1036. Min Act: -0.0000

Data Index: 574783 (The Pile)

Max Activating Token Index: 323

Text #9

Max Range: 8.7701. Min Range: -8.7701

Max Act: 7.7176. Min Act: -0.0000

Data Index: 714241 (The Pile)

Max Activating Token Index: 739

Text #10

Max Range: 8.7701. Min Range: -8.7701

Max Act: 7.7534. Min Act: -0.0000

Data Index: 223692 (The Pile)

Max Activating Token Index: 708

Text #11

Max Range: 8.7701. Min Range: -8.7701

Max Act: 6.6391. Min Act: -0.0000

Data Index: 1750320 (The Pile)

Max Activating Token Index: 853

Text #12

Max Range: 8.7701. Min Range: -8.7701

Max Act: 8.0518. Min Act: -0.0000

Data Index: 723032 (The Pile)

Max Activating Token Index: 572

Text #13

Max Range: 8.7701. Min Range: -8.7701

Max Act: 8.0391. Min Act: -0.0000

Data Index: 262332 (The Pile)

Max Activating Token Index: 767

Text #14

Max Range: 8.7701. Min Range: -8.7701

Max Act: 7.9671. Min Act: -0.0000

Data Index: 1380585 (The Pile)

Max Activating Token Index: 873

Text #15

Max Range: 8.7701. Min Range: -8.7701

Max Act: 7.8869. Min Act: -0.0000

Data Index: 1879911 (The Pile)

Max Activating Token Index: 642

Text #16

Max Range: 8.7701. Min Range: -8.7701

Max Act: 8.0622. Min Act: -0.0000

Data Index: 1235657 (The Pile)

Max Activating Token Index: 701

Text #17

Max Range: 8.7701. Min Range: -8.7701

Max Act: 7.5702. Min Act: -0.0000

Data Index: 187089 (The Pile)

Max Activating Token Index: 683

Text #18

Max Range: 8.7701. Min Range: -8.7701

Max Act: 7.8843. Min Act: -0.0000

Data Index: 236368 (The Pile)

Max Activating Token Index: 515

Text #19

Max Range: 8.7701. Min Range: -8.7701

Max Act: 7.5622. Min Act: -0.0000

Data Index: 194525 (The Pile)

Max Activating Token Index: 522