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

Dataset: The Pile

Neuron 4071 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: 3.5609. Min Range: -3.5609

Max Act: 3.1449. Min Act: -0.0000

Data Index: 587210 (The Pile)

Max Activating Token Index: 714

Text #1

Max Range: 3.5609. Min Range: -3.5609

Max Act: 3.4230. Min Act: -0.0000

Data Index: 195277 (The Pile)

Max Activating Token Index: 296

Text #2

Max Range: 3.5609. Min Range: -3.5609

Max Act: 3.0793. Min Act: -0.0000

Data Index: 46672 (The Pile)

Max Activating Token Index: 329

Text #3

Max Range: 3.5609. Min Range: -3.5609

Max Act: 3.5609. Min Act: -0.0000

Data Index: 586412 (The Pile)

Max Activating Token Index: 725

Text #4

Max Range: 3.5609. Min Range: -3.5609

Max Act: 3.1680. Min Act: -0.0000

Data Index: 490585 (The Pile)

Max Activating Token Index: 763

Text #5

Max Range: 3.5609. Min Range: -3.5609

Max Act: 3.2112. Min Act: -0.0000

Data Index: 389286 (The Pile)

Max Activating Token Index: 855

Text #6

Max Range: 3.5609. Min Range: -3.5609

Max Act: 3.0090. Min Act: -0.0000

Data Index: 1983688 (The Pile)

Max Activating Token Index: 383

Text #7

Max Range: 3.5609. Min Range: -3.5609

Max Act: 3.2464. Min Act: -0.0000

Data Index: 692331 (The Pile)

Max Activating Token Index: 868

Text #8

Max Range: 3.5609. Min Range: -3.5609

Max Act: 2.4543. Min Act: -0.0000

Data Index: 364758 (The Pile)

Max Activating Token Index: 969

Text #9

Max Range: 3.5609. Min Range: -3.5609

Max Act: 3.0309. Min Act: -0.0000

Data Index: 1764723 (The Pile)

Max Activating Token Index: 786

Text #10

Max Range: 3.5609. Min Range: -3.5609

Max Act: 2.6981. Min Act: -0.0000

Data Index: 1139055 (The Pile)

Max Activating Token Index: 605

Text #11

Max Range: 3.5609. Min Range: -3.5609

Max Act: 2.8253. Min Act: -0.0001

Data Index: 1635176 (The Pile)

Max Activating Token Index: 430

Text #12

Max Range: 3.5609. Min Range: -3.5609

Max Act: 2.9243. Min Act: -0.0000

Data Index: 948636 (The Pile)

Max Activating Token Index: 207

Text #13

Max Range: 3.5609. Min Range: -3.5609

Max Act: 2.4873. Min Act: -0.0000

Data Index: 488888 (The Pile)

Max Activating Token Index: 774

Text #14

Max Range: 3.5609. Min Range: -3.5609

Max Act: 2.7817. Min Act: -0.0000

Data Index: 705593 (The Pile)

Max Activating Token Index: 264

Text #15

Max Range: 3.5609. Min Range: -3.5609

Max Act: 2.7286. Min Act: -0.0000

Data Index: 1767086 (The Pile)

Max Activating Token Index: 219

Text #16

Max Range: 3.5609. Min Range: -3.5609

Max Act: 2.7017. Min Act: -0.0000

Data Index: 1077543 (The Pile)

Max Activating Token Index: 316

Text #17

Max Range: 3.5609. Min Range: -3.5609

Max Act: 2.3983. Min Act: -0.0000

Data Index: 602510 (The Pile)

Max Activating Token Index: 778

Text #18

Max Range: 3.5609. Min Range: -3.5609

Max Act: 2.8124. Min Act: -0.0000

Data Index: 167514 (The Pile)

Max Activating Token Index: 216

Text #19

Max Range: 3.5609. Min Range: -3.5609

Max Act: 3.0223. Min Act: -0.0000

Data Index: 1997598 (The Pile)

Max Activating Token Index: 777