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

Dataset: The Pile

Neuron 2613 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: 8.3406. Min Range: -8.3406

Max Act: 8.1692. Min Act: -0.0000

Data Index: 560048 (The Pile)

Max Activating Token Index: 897

Text #1

Max Range: 8.3406. Min Range: -8.3406

Max Act: 8.3406. Min Act: -0.0000

Data Index: 1730271 (The Pile)

Max Activating Token Index: 971

Text #2

Max Range: 8.3406. Min Range: -8.3406

Max Act: 7.3167. Min Act: -0.0000

Data Index: 437181 (The Pile)

Max Activating Token Index: 249

Text #3

Max Range: 8.3406. Min Range: -8.3406

Max Act: 7.4506. Min Act: -0.0000

Data Index: 751497 (The Pile)

Max Activating Token Index: 720

Text #4

Max Range: 8.3406. Min Range: -8.3406

Max Act: 7.1431. Min Act: -0.0000

Data Index: 669750 (The Pile)

Max Activating Token Index: 706

Text #5

Max Range: 8.3406. Min Range: -8.3406

Max Act: 7.5961. Min Act: -0.0000

Data Index: 319238 (The Pile)

Max Activating Token Index: 604

Text #6

Max Range: 8.3406. Min Range: -8.3406

Max Act: 7.4816. Min Act: -0.0000

Data Index: 1922304 (The Pile)

Max Activating Token Index: 1014

Text #7

Max Range: 8.3406. Min Range: -8.3406

Max Act: 6.4153. Min Act: -0.0000

Data Index: 1345142 (The Pile)

Max Activating Token Index: 984

Text #8

Max Range: 8.3406. Min Range: -8.3406

Max Act: 6.9565. Min Act: -0.0000

Data Index: 1266705 (The Pile)

Max Activating Token Index: 205

Text #9

Max Range: 8.3406. Min Range: -8.3406

Max Act: 7.3013. Min Act: -0.0000

Data Index: 225607 (The Pile)

Max Activating Token Index: 730

Text #10

Max Range: 8.3406. Min Range: -8.3406

Max Act: 6.8549. Min Act: -0.0000

Data Index: 770865 (The Pile)

Max Activating Token Index: 755

Text #11

Max Range: 8.3406. Min Range: -8.3406

Max Act: 6.2218. Min Act: -0.0000

Data Index: 608648 (The Pile)

Max Activating Token Index: 964

Text #12

Max Range: 8.3406. Min Range: -8.3406

Max Act: 6.6173. Min Act: -0.0000

Data Index: 1438099 (The Pile)

Max Activating Token Index: 653

Text #13

Max Range: 8.3406. Min Range: -8.3406

Max Act: 6.5838. Min Act: -0.0000

Data Index: 910731 (The Pile)

Max Activating Token Index: 741

Text #14

Max Range: 8.3406. Min Range: -8.3406

Max Act: 6.5293. Min Act: -0.0000

Data Index: 1333119 (The Pile)

Max Activating Token Index: 243

Text #15

Max Range: 8.3406. Min Range: -8.3406

Max Act: 6.2519. Min Act: -0.0000

Data Index: 1076969 (The Pile)

Max Activating Token Index: 192

Text #16

Max Range: 8.3406. Min Range: -8.3406

Max Act: 5.7140. Min Act: -0.0000

Data Index: 464619 (The Pile)

Max Activating Token Index: 527

Text #17

Max Range: 8.3406. Min Range: -8.3406

Max Act: 6.0539. Min Act: -0.0000

Data Index: 397058 (The Pile)

Max Activating Token Index: 27

Text #18

Max Range: 8.3406. Min Range: -8.3406

Max Act: 5.8634. Min Act: -0.0000

Data Index: 803804 (The Pile)

Max Activating Token Index: 662

Text #19

Max Range: 8.3406. Min Range: -8.3406

Max Act: 4.5633. Min Act: -0.0000

Data Index: 150815 (The Pile)

Max Activating Token Index: 849