Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: The Pile

Neuron 4874 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l-pile')

Text #0

Max Range: 5.8725. Min Range: -5.8725

Max Act: 5.5600. Min Act: -0.0000

Data Index: 109598 (The Pile)

Max Activating Token Index: 897

Text #1

Max Range: 5.8725. Min Range: -5.8725

Max Act: 5.8725. Min Act: -0.0000

Data Index: 950711 (The Pile)

Max Activating Token Index: 836

Text #2

Max Range: 5.8725. Min Range: -5.8725

Max Act: 5.8643. Min Act: -0.0000

Data Index: 1179705 (The Pile)

Max Activating Token Index: 865

Text #3

Max Range: 5.8725. Min Range: -5.8725

Max Act: 5.8567. Min Act: -0.0000

Data Index: 1815766 (The Pile)

Max Activating Token Index: 871

Text #4

Max Range: 5.8725. Min Range: -5.8725

Max Act: 5.3482. Min Act: -0.0000

Data Index: 1949177 (The Pile)

Max Activating Token Index: 994

Text #5

Max Range: 5.8725. Min Range: -5.8725

Max Act: 5.5439. Min Act: -0.0000

Data Index: 550578 (The Pile)

Max Activating Token Index: 579

Text #6

Max Range: 5.8725. Min Range: -5.8725

Max Act: 5.2170. Min Act: -0.0000

Data Index: 536290 (The Pile)

Max Activating Token Index: 541

Text #7

Max Range: 5.8725. Min Range: -5.8725

Max Act: 4.7867. Min Act: -0.0000

Data Index: 673218 (The Pile)

Max Activating Token Index: 465

Text #8

Max Range: 5.8725. Min Range: -5.8725

Max Act: 4.2557. Min Act: -0.0000

Data Index: 236511 (The Pile)

Max Activating Token Index: 443

Text #9

Max Range: 5.8725. Min Range: -5.8725

Max Act: 4.5811. Min Act: -0.0000

Data Index: 1514823 (The Pile)

Max Activating Token Index: 402

Text #10

Max Range: 5.8725. Min Range: -5.8725

Max Act: 4.1601. Min Act: -0.0000

Data Index: 782900 (The Pile)

Max Activating Token Index: 58

Text #11

Max Range: 5.8725. Min Range: -5.8725

Max Act: 4.1644. Min Act: -0.0000

Data Index: 387362 (The Pile)

Max Activating Token Index: 80

Text #12

Max Range: 5.8725. Min Range: -5.8725

Max Act: 4.0211. Min Act: -0.0000

Data Index: 895292 (The Pile)

Max Activating Token Index: 119

Text #13

Max Range: 5.8725. Min Range: -5.8725

Max Act: 4.0631. Min Act: -0.0000

Data Index: 1796812 (The Pile)

Max Activating Token Index: 349

Text #14

Max Range: 5.8725. Min Range: -5.8725

Max Act: 3.4499. Min Act: -0.0000

Data Index: 1141122 (The Pile)

Max Activating Token Index: 355

Text #15

Max Range: 5.8725. Min Range: -5.8725

Max Act: 3.1407. Min Act: 0.0000

Data Index: 1036808 (The Pile)

Max Activating Token Index: 987

Text #16

Max Range: 5.8725. Min Range: -5.8725

Max Act: 3.3744. Min Act: -0.0000

Data Index: 1674625 (The Pile)

Max Activating Token Index: 26

Text #17

Max Range: 5.8725. Min Range: -5.8725

Max Act: 3.7824. Min Act: -0.0000

Data Index: 1584158 (The Pile)

Max Activating Token Index: 862

Text #18

Max Range: 5.8725. Min Range: -5.8725

Max Act: 3.0877. Min Act: -0.0000

Data Index: 1992934 (The Pile)

Max Activating Token Index: 810

Text #19

Max Range: 5.8725. Min Range: -5.8725

Max Act: 2.8764. Min Act: -0.0000

Data Index: 1187455 (The Pile)

Max Activating Token Index: 395