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

Dataset: The Pile

Neuron 878 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: 7.2617. Min Range: -7.2617

Max Act: 6.8608. Min Act: -0.0000

Data Index: 1546178 (The Pile)

Max Activating Token Index: 672

Text #1

Max Range: 7.2617. Min Range: -7.2617

Max Act: 7.0292. Min Act: -0.0000

Data Index: 754600 (The Pile)

Max Activating Token Index: 445

Text #2

Max Range: 7.2617. Min Range: -7.2617

Max Act: 7.2617. Min Act: -0.0000

Data Index: 114194 (The Pile)

Max Activating Token Index: 635

Text #3

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.9893. Min Act: -0.0000

Data Index: 844422 (The Pile)

Max Activating Token Index: 622

Text #4

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.6065. Min Act: -0.0000

Data Index: 1645 (The Pile)

Max Activating Token Index: 436

Text #5

Max Range: 7.2617. Min Range: -7.2617

Max Act: 7.0196. Min Act: -0.0000

Data Index: 886128 (The Pile)

Max Activating Token Index: 755

Text #6

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.3760. Min Act: -0.0000

Data Index: 663063 (The Pile)

Max Activating Token Index: 819

Text #7

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.5873. Min Act: -0.0000

Data Index: 1220961 (The Pile)

Max Activating Token Index: 676

Text #8

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.7917. Min Act: -0.0000

Data Index: 1796174 (The Pile)

Max Activating Token Index: 504

Text #9

Max Range: 7.2617. Min Range: -7.2617

Max Act: 5.9282. Min Act: -0.0000

Data Index: 978844 (The Pile)

Max Activating Token Index: 305

Text #10

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.5251. Min Act: -0.0000

Data Index: 77593 (The Pile)

Max Activating Token Index: 716

Text #11

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.3076. Min Act: -0.0000

Data Index: 1642403 (The Pile)

Max Activating Token Index: 516

Text #12

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.2565. Min Act: -0.0000

Data Index: 742067 (The Pile)

Max Activating Token Index: 603

Text #13

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.2173. Min Act: -0.0000

Data Index: 551496 (The Pile)

Max Activating Token Index: 710

Text #14

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.2285. Min Act: -0.0000

Data Index: 1805462 (The Pile)

Max Activating Token Index: 214

Text #15

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.1565. Min Act: -0.0000

Data Index: 788660 (The Pile)

Max Activating Token Index: 554

Text #16

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.4346. Min Act: -0.0000

Data Index: 1423853 (The Pile)

Max Activating Token Index: 546

Text #17

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.8134. Min Act: -0.0000

Data Index: 1566743 (The Pile)

Max Activating Token Index: 614

Text #18

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.5533. Min Act: -0.0000

Data Index: 1557500 (The Pile)

Max Activating Token Index: 692

Text #19

Max Range: 7.2617. Min Range: -7.2617

Max Act: 6.1251. Min Act: -0.0000

Data Index: 490112 (The Pile)

Max Activating Token Index: 642