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

Dataset: The Pile

Neuron 3471 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.1988. Min Range: -8.1988

Max Act: 8.1988. Min Act: -0.0000

Data Index: 466788 (The Pile)

Max Activating Token Index: 735

Text #1

Max Range: 8.1988. Min Range: -8.1988

Max Act: 7.4275. Min Act: -0.0000

Data Index: 545072 (The Pile)

Max Activating Token Index: 71

Text #2

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.9904. Min Act: -0.0000

Data Index: 913031 (The Pile)

Max Activating Token Index: 953

Text #3

Max Range: 8.1988. Min Range: -8.1988

Max Act: 7.5342. Min Act: -0.0000

Data Index: 896464 (The Pile)

Max Activating Token Index: 690

Text #4

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.9907. Min Act: -0.0000

Data Index: 85065 (The Pile)

Max Activating Token Index: 669

Text #5

Max Range: 8.1988. Min Range: -8.1988

Max Act: 7.0118. Min Act: -0.0000

Data Index: 1585664 (The Pile)

Max Activating Token Index: 644

Text #6

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.8910. Min Act: -0.0000

Data Index: 160497 (The Pile)

Max Activating Token Index: 429

Text #7

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.8954. Min Act: -0.0000

Data Index: 963348 (The Pile)

Max Activating Token Index: 465

Text #8

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.6342. Min Act: -0.0000

Data Index: 1173090 (The Pile)

Max Activating Token Index: 892

Text #9

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.5800. Min Act: -0.0000

Data Index: 636499 (The Pile)

Max Activating Token Index: 705

Text #10

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.5421. Min Act: -0.0000

Data Index: 330771 (The Pile)

Max Activating Token Index: 622

Text #11

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.7293. Min Act: -0.0000

Data Index: 1917060 (The Pile)

Max Activating Token Index: 907

Text #12

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.2164. Min Act: -0.0000

Data Index: 337759 (The Pile)

Max Activating Token Index: 961

Text #13

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.6034. Min Act: -0.0000

Data Index: 1909642 (The Pile)

Max Activating Token Index: 369

Text #14

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.1116. Min Act: -0.0000

Data Index: 1346736 (The Pile)

Max Activating Token Index: 879

Text #15

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.4004. Min Act: -0.0000

Data Index: 1702351 (The Pile)

Max Activating Token Index: 281

Text #16

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.4334. Min Act: -0.0000

Data Index: 1914039 (The Pile)

Max Activating Token Index: 214

Text #17

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.1806. Min Act: -0.0000

Data Index: 843932 (The Pile)

Max Activating Token Index: 254

Text #18

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.3604. Min Act: -0.0000

Data Index: 1643265 (The Pile)

Max Activating Token Index: 807

Text #19

Max Range: 8.1988. Min Range: -8.1988

Max Act: 6.7771. Min Act: -0.0000

Data Index: 1824295 (The Pile)

Max Activating Token Index: 919