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

Dataset: The Pile

Neuron 287 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: 7.3740. Min Range: -7.3740

Max Act: 7.2572. Min Act: -0.0000

Data Index: 676960 (The Pile)

Max Activating Token Index: 389

Text #1

Max Range: 7.3740. Min Range: -7.3740

Max Act: 7.3740. Min Act: -0.0000

Data Index: 419696 (The Pile)

Max Activating Token Index: 133

Text #2

Max Range: 7.3740. Min Range: -7.3740

Max Act: 6.7038. Min Act: -0.0000

Data Index: 348258 (The Pile)

Max Activating Token Index: 86

Text #3

Max Range: 7.3740. Min Range: -7.3740

Max Act: 6.7490. Min Act: -0.0000

Data Index: 1131960 (The Pile)

Max Activating Token Index: 179

Text #4

Max Range: 7.3740. Min Range: -7.3740

Max Act: 6.6130. Min Act: -0.0000

Data Index: 1892012 (The Pile)

Max Activating Token Index: 497

Text #5

Max Range: 7.3740. Min Range: -7.3740

Max Act: 6.6877. Min Act: -0.0000

Data Index: 1411620 (The Pile)

Max Activating Token Index: 102

Text #6

Max Range: 7.3740. Min Range: -7.3740

Max Act: 6.0725. Min Act: -0.0000

Data Index: 679740 (The Pile)

Max Activating Token Index: 70

Text #7

Max Range: 7.3740. Min Range: -7.3740

Max Act: 5.8108. Min Act: -0.0000

Data Index: 656450 (The Pile)

Max Activating Token Index: 577

Text #8

Max Range: 7.3740. Min Range: -7.3740

Max Act: 5.7791. Min Act: -0.0000

Data Index: 1129263 (The Pile)

Max Activating Token Index: 398

Text #9

Max Range: 7.3740. Min Range: -7.3740

Max Act: 5.6648. Min Act: -0.0000

Data Index: 1023050 (The Pile)

Max Activating Token Index: 99

Text #10

Max Range: 7.3740. Min Range: -7.3740

Max Act: 5.2680. Min Act: -0.0000

Data Index: 558946 (The Pile)

Max Activating Token Index: 559

Text #11

Max Range: 7.3740. Min Range: -7.3740

Max Act: 5.4079. Min Act: -0.0000

Data Index: 1860645 (The Pile)

Max Activating Token Index: 669

Text #12

Max Range: 7.3740. Min Range: -7.3740

Max Act: 6.0007. Min Act: -0.0000

Data Index: 462442 (The Pile)

Max Activating Token Index: 30

Text #13

Max Range: 7.3740. Min Range: -7.3740

Max Act: 5.6086. Min Act: -0.0000

Data Index: 968768 (The Pile)

Max Activating Token Index: 476

Text #14

Max Range: 7.3740. Min Range: -7.3740

Max Act: 5.5518. Min Act: -0.0000

Data Index: 1640003 (The Pile)

Max Activating Token Index: 661

Text #15

Max Range: 7.3740. Min Range: -7.3740

Max Act: 4.9030. Min Act: -0.0000

Data Index: 1083419 (The Pile)

Max Activating Token Index: 316

Text #16

Max Range: 7.3740. Min Range: -7.3740

Max Act: 5.0886. Min Act: -0.0000

Data Index: 380202 (The Pile)

Max Activating Token Index: 477

Text #17

Max Range: 7.3740. Min Range: -7.3740

Max Act: 5.3116. Min Act: -0.0000

Data Index: 956238 (The Pile)

Max Activating Token Index: 596

Text #18

Max Range: 7.3740. Min Range: -7.3740

Max Act: 5.0799. Min Act: -0.0000

Data Index: 1533322 (The Pile)

Max Activating Token Index: 315

Text #19

Max Range: 7.3740. Min Range: -7.3740

Max Act: 4.6866. Min Act: -0.0000

Data Index: 542776 (The Pile)

Max Activating Token Index: 609