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

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 830 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.3725. Min Act: -0.0000

Data Index: 659090 (C4 (Web Text))

Max Activating Token Index: 506

Text #1

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.2865. Min Act: -0.0000

Data Index: 537412 (C4 (Web Text))

Max Activating Token Index: 332

Text #2

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.2199. Min Act: -0.0000

Data Index: 1133557 (C4 (Web Text))

Max Activating Token Index: 923

Text #3

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.1178. Min Act: -0.0000

Data Index: 275624 (C4 (Web Text))

Max Activating Token Index: 327

Text #4

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0989. Min Act: -0.0000

Data Index: 895137 (C4 (Web Text))

Max Activating Token Index: 245

Text #5

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.1333. Min Act: -0.0000

Data Index: 717800 (C4 (Web Text))

Max Activating Token Index: 667

Text #6

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.1372. Min Act: -0.0000

Data Index: 20935 (C4 (Web Text))

Max Activating Token Index: 481

Text #7

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.1021. Min Act: -0.0000

Data Index: 1642965 (Python Code)

Max Activating Token Index: 1000

Text #8

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0592. Min Act: -0.0000

Data Index: 793435 (C4 (Web Text))

Max Activating Token Index: 451

Text #9

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0558. Min Act: -0.0000

Data Index: 1281840 (C4 (Web Text))

Max Activating Token Index: 175

Text #10

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0404. Min Act: -0.0000

Data Index: 31223 (C4 (Web Text))

Max Activating Token Index: 75

Text #11

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0445. Min Act: -0.0000

Data Index: 722742 (C4 (Web Text))

Max Activating Token Index: 831

Text #12

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0019. Min Act: -0.0000

Data Index: 1514156 (Python Code)

Max Activating Token Index: 224

Text #13

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0335. Min Act: -0.0000

Data Index: 682342 (C4 (Web Text))

Max Activating Token Index: 246

Text #14

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0626. Min Act: -0.0000

Data Index: 744009 (C4 (Web Text))

Max Activating Token Index: 791

Text #15

Max Range: 1.3725. Min Range: -1.3725

Max Act: 0.9343. Min Act: -0.0000

Data Index: 482837 (C4 (Web Text))

Max Activating Token Index: 105

Text #16

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0539. Min Act: -0.0000

Data Index: 21216 (C4 (Web Text))

Max Activating Token Index: 123

Text #17

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0580. Min Act: -0.0000

Data Index: 210354 (C4 (Web Text))

Max Activating Token Index: 137

Text #18

Max Range: 1.3725. Min Range: -1.3725

Max Act: 0.9966. Min Act: -0.0000

Data Index: 665131 (C4 (Web Text))

Max Activating Token Index: 754

Text #19

Max Range: 1.3725. Min Range: -1.3725

Max Act: 1.0074. Min Act: -0.0000

Data Index: 131694 (C4 (Web Text))

Max Activating Token Index: 417