Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: The Pile

Neuron 581 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l-pile')

Text #0

Max Range: 6.1686. Min Range: -6.1686

Max Act: 6.1686. Min Act: -0.0000

Data Index: 92629 (The Pile)

Max Activating Token Index: 595

Text #1

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.9608. Min Act: -0.0000

Data Index: 1195651 (The Pile)

Max Activating Token Index: 100

Text #2

Max Range: 6.1686. Min Range: -6.1686

Max Act: 6.0163. Min Act: -0.0000

Data Index: 1968624 (The Pile)

Max Activating Token Index: 817

Text #3

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.7681. Min Act: -0.0000

Data Index: 1449817 (The Pile)

Max Activating Token Index: 776

Text #4

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.6450. Min Act: -0.0000

Data Index: 920020 (The Pile)

Max Activating Token Index: 368

Text #5

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.7977. Min Act: -0.0000

Data Index: 920141 (The Pile)

Max Activating Token Index: 944

Text #6

Max Range: 6.1686. Min Range: -6.1686

Max Act: 6.1372. Min Act: -0.0000

Data Index: 1200912 (The Pile)

Max Activating Token Index: 710

Text #7

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.4420. Min Act: -0.0000

Data Index: 1430285 (The Pile)

Max Activating Token Index: 797

Text #8

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.7650. Min Act: -0.0000

Data Index: 264698 (The Pile)

Max Activating Token Index: 803

Text #9

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.5540. Min Act: -0.0000

Data Index: 698784 (The Pile)

Max Activating Token Index: 572

Text #10

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.5256. Min Act: -0.0000

Data Index: 932582 (The Pile)

Max Activating Token Index: 715

Text #11

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.6473. Min Act: -0.0000

Data Index: 243714 (The Pile)

Max Activating Token Index: 826

Text #12

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.4573. Min Act: -0.0000

Data Index: 1549514 (The Pile)

Max Activating Token Index: 641

Text #13

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.2676. Min Act: -0.0000

Data Index: 1010533 (The Pile)

Max Activating Token Index: 271

Text #14

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.5706. Min Act: -0.0000

Data Index: 578164 (The Pile)

Max Activating Token Index: 894

Text #15

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.5540. Min Act: -0.0000

Data Index: 381176 (The Pile)

Max Activating Token Index: 118

Text #16

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.5618. Min Act: -0.0000

Data Index: 608561 (The Pile)

Max Activating Token Index: 451

Text #17

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.3000. Min Act: -0.0000

Data Index: 534771 (The Pile)

Max Activating Token Index: 442

Text #18

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.4677. Min Act: -0.0000

Data Index: 1235513 (The Pile)

Max Activating Token Index: 620

Text #19

Max Range: 6.1686. Min Range: -6.1686

Max Act: 5.6480. Min Act: -0.0000

Data Index: 1658940 (The Pile)

Max Activating Token Index: 781