Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

Dataset: The Pile

Neuron 672 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l-pile')

Text #0

Max Range: 1.1992. Min Range: -1.1992

Max Act: 1.1365. Min Act: -0.0000

Data Index: 859584 (The Pile)

Max Activating Token Index: 697

Text #1

Max Range: 1.1992. Min Range: -1.1992

Max Act: 1.1992. Min Act: -0.0000

Data Index: 1427768 (The Pile)

Max Activating Token Index: 892

Text #2

Max Range: 1.1992. Min Range: -1.1992

Max Act: 1.1244. Min Act: -0.0000

Data Index: 857727 (The Pile)

Max Activating Token Index: 356

Text #3

Max Range: 1.1992. Min Range: -1.1992

Max Act: 1.1106. Min Act: -0.0000

Data Index: 182258 (The Pile)

Max Activating Token Index: 394

Text #4

Max Range: 1.1992. Min Range: -1.1992

Max Act: 1.1320. Min Act: -0.0000

Data Index: 1498859 (The Pile)

Max Activating Token Index: 616

Text #5

Max Range: 1.1992. Min Range: -1.1992

Max Act: 1.0909. Min Act: -0.0000

Data Index: 839102 (The Pile)

Max Activating Token Index: 672

Text #6

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9947. Min Act: -0.0000

Data Index: 459201 (The Pile)

Max Activating Token Index: 28

Text #7

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9800. Min Act: -0.0000

Data Index: 1719123 (The Pile)

Max Activating Token Index: 665

Text #8

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.8618. Min Act: -0.0000

Data Index: 471528 (The Pile)

Max Activating Token Index: 764

Text #9

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9277. Min Act: -0.0000

Data Index: 1503611 (The Pile)

Max Activating Token Index: 415

Text #10

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9278. Min Act: -0.0000

Data Index: 262424 (The Pile)

Max Activating Token Index: 675

Text #11

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9499. Min Act: -0.0000

Data Index: 488095 (The Pile)

Max Activating Token Index: 840

Text #12

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9175. Min Act: -0.0000

Data Index: 230232 (The Pile)

Max Activating Token Index: 422

Text #13

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9077. Min Act: -0.0000

Data Index: 1904428 (The Pile)

Max Activating Token Index: 518

Text #14

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9264. Min Act: -0.0000

Data Index: 603215 (The Pile)

Max Activating Token Index: 845

Text #15

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9264. Min Act: -0.0000

Data Index: 1768206 (The Pile)

Max Activating Token Index: 588

Text #16

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9505. Min Act: -0.0000

Data Index: 1031849 (The Pile)

Max Activating Token Index: 174

Text #17

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.7792. Min Act: -0.0000

Data Index: 780900 (The Pile)

Max Activating Token Index: 858

Text #18

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.9601. Min Act: -0.0000

Data Index: 223613 (The Pile)

Max Activating Token Index: 499

Text #19

Max Range: 1.1992. Min Range: -1.1992

Max Act: 0.8717. Min Act: -0.0001

Data Index: 1232584 (The Pile)

Max Activating Token Index: 1009