Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 382 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0697. Min Act: -0.0006

Data Index: 774381 (C4 (Web Text))

Max Activating Token Index: 157

Text #1

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0632. Min Act: -0.0005

Data Index: 759184 (C4 (Web Text))

Max Activating Token Index: 369

Text #2

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0639. Min Act: -0.0005

Data Index: 19694 (C4 (Web Text))

Max Activating Token Index: 80

Text #3

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0604. Min Act: -0.0006

Data Index: 446538 (C4 (Web Text))

Max Activating Token Index: 319

Text #4

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0625. Min Act: -0.0006

Data Index: 705724 (C4 (Web Text))

Max Activating Token Index: 568

Text #5

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0571. Min Act: -0.0006

Data Index: 734336 (C4 (Web Text))

Max Activating Token Index: 263

Text #6

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0590. Min Act: -0.0006

Data Index: 1228343 (C4 (Web Text))

Max Activating Token Index: 697

Text #7

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0575. Min Act: -0.0006

Data Index: 1007243 (C4 (Web Text))

Max Activating Token Index: 586

Text #8

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0548. Min Act: -0.0006

Data Index: 776706 (C4 (Web Text))

Max Activating Token Index: 202

Text #9

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0581. Min Act: -0.0005

Data Index: 728410 (C4 (Web Text))

Max Activating Token Index: 779

Text #10

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0553. Min Act: -0.0006

Data Index: 577894 (C4 (Web Text))

Max Activating Token Index: 37

Text #11

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0532. Min Act: -0.0006

Data Index: 224544 (C4 (Web Text))

Max Activating Token Index: 822

Text #12

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0537. Min Act: -0.0006

Data Index: 843799 (C4 (Web Text))

Max Activating Token Index: 223

Text #13

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0548. Min Act: -0.0006

Data Index: 284834 (C4 (Web Text))

Max Activating Token Index: 219

Text #14

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0515. Min Act: -0.0005

Data Index: 1117388 (C4 (Web Text))

Max Activating Token Index: 27

Text #15

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0548. Min Act: -0.0006

Data Index: 794686 (C4 (Web Text))

Max Activating Token Index: 226

Text #16

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0535. Min Act: -0.0006

Data Index: 769777 (C4 (Web Text))

Max Activating Token Index: 917

Text #17

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0555. Min Act: -0.0006

Data Index: 256286 (C4 (Web Text))

Max Activating Token Index: 531

Text #18

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0513. Min Act: -0.0006

Data Index: 1141686 (C4 (Web Text))

Max Activating Token Index: 755

Text #19

Max Range: 0.0697. Min Range: -0.0697

Max Act: 0.0561. Min Act: -0.0006

Data Index: 335641 (C4 (Web Text))

Max Activating Token Index: 923