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

Dataset: The Pile

Neuron 2392 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.9331. Min Act: -0.0000

Data Index: 762818 (The Pile)

Max Activating Token Index: 827

Text #1

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.9732. Min Act: -0.0000

Data Index: 744979 (The Pile)

Max Activating Token Index: 628

Text #2

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8354. Min Act: -0.0000

Data Index: 776604 (The Pile)

Max Activating Token Index: 812

Text #3

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8353. Min Act: -0.0000

Data Index: 1440209 (The Pile)

Max Activating Token Index: 945

Text #4

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8331. Min Act: -0.0000

Data Index: 960475 (The Pile)

Max Activating Token Index: 411

Text #5

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8442. Min Act: 0.0000

Data Index: 1793937 (The Pile)

Max Activating Token Index: 317

Text #6

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8162. Min Act: -0.0000

Data Index: 905806 (The Pile)

Max Activating Token Index: 223

Text #7

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8694. Min Act: -0.0000

Data Index: 503947 (The Pile)

Max Activating Token Index: 848

Text #8

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8456. Min Act: -0.0000

Data Index: 139943 (The Pile)

Max Activating Token Index: 685

Text #9

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.7938. Min Act: -0.0000

Data Index: 1494925 (The Pile)

Max Activating Token Index: 423

Text #10

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.7893. Min Act: -0.0000

Data Index: 784840 (The Pile)

Max Activating Token Index: 874

Text #11

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8416. Min Act: 0.0000

Data Index: 305549 (The Pile)

Max Activating Token Index: 969

Text #12

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8430. Min Act: -0.0000

Data Index: 1668889 (The Pile)

Max Activating Token Index: 902

Text #13

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.7891. Min Act: -0.0000

Data Index: 1991795 (The Pile)

Max Activating Token Index: 330

Text #14

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8118. Min Act: -0.0000

Data Index: 103260 (The Pile)

Max Activating Token Index: 989

Text #15

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.7877. Min Act: -0.0000

Data Index: 36860 (The Pile)

Max Activating Token Index: 241

Text #16

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.7839. Min Act: 0.0000

Data Index: 437126 (The Pile)

Max Activating Token Index: 618

Text #17

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8388. Min Act: -0.0000

Data Index: 1399542 (The Pile)

Max Activating Token Index: 572

Text #18

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.8634. Min Act: -0.0000

Data Index: 1545858 (The Pile)

Max Activating Token Index: 492

Text #19

Max Range: 0.9732. Min Range: -0.9732

Max Act: 0.7824. Min Act: -0.0000

Data Index: 361860 (The Pile)

Max Activating Token Index: 942