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

Dataset: The Pile

Neuron 2164 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.7845. Min Range: -8.7845

Max Act: 8.7845. Min Act: -0.0000

Data Index: 468599 (The Pile)

Max Activating Token Index: 767

Text #1

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.7706. Min Act: -0.0000

Data Index: 762531 (The Pile)

Max Activating Token Index: 392

Text #2

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.9674. Min Act: -0.0000

Data Index: 1016751 (The Pile)

Max Activating Token Index: 509

Text #3

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.9857. Min Act: -0.0000

Data Index: 729338 (The Pile)

Max Activating Token Index: 571

Text #4

Max Range: 8.7845. Min Range: -8.7845

Max Act: 8.0141. Min Act: -0.0000

Data Index: 867190 (The Pile)

Max Activating Token Index: 266

Text #5

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.8407. Min Act: -0.0000

Data Index: 608850 (The Pile)

Max Activating Token Index: 584

Text #6

Max Range: 8.7845. Min Range: -8.7845

Max Act: 4.8355. Min Act: -0.0000

Data Index: 595415 (The Pile)

Max Activating Token Index: 696

Text #7

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.4607. Min Act: -0.0000

Data Index: 143344 (The Pile)

Max Activating Token Index: 813

Text #8

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.4112. Min Act: -0.0000

Data Index: 1529505 (The Pile)

Max Activating Token Index: 18

Text #9

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.3854. Min Act: -0.0000

Data Index: 1823861 (The Pile)

Max Activating Token Index: 460

Text #10

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.2687. Min Act: -0.0000

Data Index: 688637 (The Pile)

Max Activating Token Index: 525

Text #11

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.2099. Min Act: -0.0000

Data Index: 614212 (The Pile)

Max Activating Token Index: 20

Text #12

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.3587. Min Act: -0.0000

Data Index: 570138 (The Pile)

Max Activating Token Index: 398

Text #13

Max Range: 8.7845. Min Range: -8.7845

Max Act: 6.9454. Min Act: -0.0000

Data Index: 1601445 (The Pile)

Max Activating Token Index: 721

Text #14

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.2137. Min Act: -0.0000

Data Index: 1516362 (The Pile)

Max Activating Token Index: 353

Text #15

Max Range: 8.7845. Min Range: -8.7845

Max Act: 6.8664. Min Act: -0.0000

Data Index: 1796265 (The Pile)

Max Activating Token Index: 714

Text #16

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.1485. Min Act: -0.0000

Data Index: 24966 (The Pile)

Max Activating Token Index: 339

Text #17

Max Range: 8.7845. Min Range: -8.7845

Max Act: 6.5730. Min Act: -0.0000

Data Index: 1854416 (The Pile)

Max Activating Token Index: 475

Text #18

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.0573. Min Act: -0.0000

Data Index: 199399 (The Pile)

Max Activating Token Index: 813

Text #19

Max Range: 8.7845. Min Range: -8.7845

Max Act: 7.4182. Min Act: -0.0000

Data Index: 383898 (The Pile)

Max Activating Token Index: 798