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

Dataset: The Pile

Neuron 4652 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: 8.1221. Min Range: -8.1221

Max Act: 8.1221. Min Act: -0.0000

Data Index: 1681688 (The Pile)

Max Activating Token Index: 628

Text #1

Max Range: 8.1221. Min Range: -8.1221

Max Act: 7.3789. Min Act: -0.0000

Data Index: 792323 (The Pile)

Max Activating Token Index: 389

Text #2

Max Range: 8.1221. Min Range: -8.1221

Max Act: 7.0807. Min Act: -0.0000

Data Index: 1706131 (The Pile)

Max Activating Token Index: 803

Text #3

Max Range: 8.1221. Min Range: -8.1221

Max Act: 6.6253. Min Act: -0.0000

Data Index: 928629 (The Pile)

Max Activating Token Index: 314

Text #4

Max Range: 8.1221. Min Range: -8.1221

Max Act: 6.2095. Min Act: -0.0000

Data Index: 1516630 (The Pile)

Max Activating Token Index: 731

Text #5

Max Range: 8.1221. Min Range: -8.1221

Max Act: 6.1899. Min Act: -0.0000

Data Index: 1566492 (The Pile)

Max Activating Token Index: 761

Text #6

Max Range: 8.1221. Min Range: -8.1221

Max Act: 6.1283. Min Act: -0.0000

Data Index: 1428095 (The Pile)

Max Activating Token Index: 691

Text #7

Max Range: 8.1221. Min Range: -8.1221

Max Act: 5.8509. Min Act: -0.0000

Data Index: 1512049 (The Pile)

Max Activating Token Index: 647

Text #8

Max Range: 8.1221. Min Range: -8.1221

Max Act: 5.9672. Min Act: -0.0000

Data Index: 1752042 (The Pile)

Max Activating Token Index: 84

Text #9

Max Range: 8.1221. Min Range: -8.1221

Max Act: 5.8964. Min Act: -0.0000

Data Index: 94931 (The Pile)

Max Activating Token Index: 451

Text #10

Max Range: 8.1221. Min Range: -8.1221

Max Act: 6.0492. Min Act: -0.0000

Data Index: 1491452 (The Pile)

Max Activating Token Index: 766

Text #11

Max Range: 8.1221. Min Range: -8.1221

Max Act: 5.7979. Min Act: -0.0000

Data Index: 1229679 (The Pile)

Max Activating Token Index: 829

Text #12

Max Range: 8.1221. Min Range: -8.1221

Max Act: 5.4990. Min Act: -0.0000

Data Index: 1922626 (The Pile)

Max Activating Token Index: 935

Text #13

Max Range: 8.1221. Min Range: -8.1221

Max Act: 5.4111. Min Act: -0.0000

Data Index: 1113990 (The Pile)

Max Activating Token Index: 447

Text #14

Max Range: 8.1221. Min Range: -8.1221

Max Act: 5.0710. Min Act: -0.0000

Data Index: 641216 (The Pile)

Max Activating Token Index: 794

Text #15

Max Range: 8.1221. Min Range: -8.1221

Max Act: 4.8635. Min Act: -0.0000

Data Index: 912531 (The Pile)

Max Activating Token Index: 343

Text #16

Max Range: 8.1221. Min Range: -8.1221

Max Act: 4.5431. Min Act: -0.0000

Data Index: 876509 (The Pile)

Max Activating Token Index: 888

Text #17

Max Range: 8.1221. Min Range: -8.1221

Max Act: 4.7596. Min Act: -0.0000

Data Index: 770547 (The Pile)

Max Activating Token Index: 878

Text #18

Max Range: 8.1221. Min Range: -8.1221

Max Act: 5.2004. Min Act: -0.0000

Data Index: 702503 (The Pile)

Max Activating Token Index: 870

Text #19

Max Range: 8.1221. Min Range: -8.1221

Max Act: 4.7047. Min Act: -0.0000

Data Index: 1982126 (The Pile)

Max Activating Token Index: 677