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

Dataset: The Pile

Neuron 4032 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: 3.5745. Min Range: -3.5745

Max Act: 3.5745. Min Act: -0.0000

Data Index: 115678 (The Pile)

Max Activating Token Index: 573

Text #1

Max Range: 3.5745. Min Range: -3.5745

Max Act: 3.4461. Min Act: -0.0000

Data Index: 93466 (The Pile)

Max Activating Token Index: 879

Text #2

Max Range: 3.5745. Min Range: -3.5745

Max Act: 3.3173. Min Act: -0.0000

Data Index: 318863 (The Pile)

Max Activating Token Index: 475

Text #3

Max Range: 3.5745. Min Range: -3.5745

Max Act: 3.2845. Min Act: -0.0000

Data Index: 659396 (The Pile)

Max Activating Token Index: 769

Text #4

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.9265. Min Act: -0.0000

Data Index: 977764 (The Pile)

Max Activating Token Index: 921

Text #5

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.9314. Min Act: -0.0000

Data Index: 1427925 (The Pile)

Max Activating Token Index: 456

Text #6

Max Range: 3.5745. Min Range: -3.5745

Max Act: 3.0851. Min Act: -0.0000

Data Index: 1680024 (The Pile)

Max Activating Token Index: 554

Text #7

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.5957. Min Act: -0.0000

Data Index: 742419 (The Pile)

Max Activating Token Index: 767

Text #8

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.9572. Min Act: -0.0000

Data Index: 1750547 (The Pile)

Max Activating Token Index: 749

Text #9

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.7773. Min Act: -0.0000

Data Index: 124129 (The Pile)

Max Activating Token Index: 517

Text #10

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.7292. Min Act: -0.0000

Data Index: 402692 (The Pile)

Max Activating Token Index: 749

Text #11

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.8008. Min Act: -0.0000

Data Index: 1698104 (The Pile)

Max Activating Token Index: 680

Text #12

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.7561. Min Act: -0.0000

Data Index: 1064372 (The Pile)

Max Activating Token Index: 797

Text #13

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.6294. Min Act: -0.0000

Data Index: 1630712 (The Pile)

Max Activating Token Index: 991

Text #14

Max Range: 3.5745. Min Range: -3.5745

Max Act: 3.1065. Min Act: -0.0000

Data Index: 634187 (The Pile)

Max Activating Token Index: 730

Text #15

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.6958. Min Act: -0.0000

Data Index: 1581934 (The Pile)

Max Activating Token Index: 869

Text #16

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.6966. Min Act: -0.0000

Data Index: 429542 (The Pile)

Max Activating Token Index: 858

Text #17

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.6987. Min Act: -0.0000

Data Index: 874805 (The Pile)

Max Activating Token Index: 580

Text #18

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.7573. Min Act: -0.0000

Data Index: 1988882 (The Pile)

Max Activating Token Index: 626

Text #19

Max Range: 3.5745. Min Range: -3.5745

Max Act: 2.5812. Min Act: -0.0000

Data Index: 1873125 (The Pile)

Max Activating Token Index: 878