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

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

Neuron 2121 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.9025. Min Range: -1.9025

Max Act: 1.9025. Min Act: -0.0000

Data Index: 1273629 (C4 (Web Text))

Max Activating Token Index: 686

Text #1

Max Range: 1.9025. Min Range: -1.9025

Max Act: 1.5900. Min Act: -0.0000

Data Index: 108434 (C4 (Web Text))

Max Activating Token Index: 894

Text #2

Max Range: 1.9025. Min Range: -1.9025

Max Act: 1.4557. Min Act: -0.0000

Data Index: 970744 (C4 (Web Text))

Max Activating Token Index: 594

Text #3

Max Range: 1.9025. Min Range: -1.9025

Max Act: 1.3238. Min Act: -0.0000

Data Index: 151617 (C4 (Web Text))

Max Activating Token Index: 868

Text #4

Max Range: 1.9025. Min Range: -1.9025

Max Act: 1.3459. Min Act: -0.0000

Data Index: 759634 (C4 (Web Text))

Max Activating Token Index: 605

Text #5

Max Range: 1.9025. Min Range: -1.9025

Max Act: 1.0562. Min Act: -0.0000

Data Index: 1342224 (C4 (Web Text))

Max Activating Token Index: 328

Text #6

Max Range: 1.9025. Min Range: -1.9025

Max Act: 1.1175. Min Act: -0.0000

Data Index: 768228 (C4 (Web Text))

Max Activating Token Index: 651

Text #7

Max Range: 1.9025. Min Range: -1.9025

Max Act: 1.0328. Min Act: -0.0000

Data Index: 1158764 (C4 (Web Text))

Max Activating Token Index: 643

Text #8

Max Range: 1.9025. Min Range: -1.9025

Max Act: 0.9891. Min Act: 0.0000

Data Index: 277092 (C4 (Web Text))

Max Activating Token Index: 211

Text #9

Max Range: 1.9025. Min Range: -1.9025

Max Act: 0.9490. Min Act: -0.0000

Data Index: 1364416 (Python Code)

Max Activating Token Index: 51

Text #10

Max Range: 1.9025. Min Range: -1.9025

Max Act: 1.1411. Min Act: -0.0000

Data Index: 265942 (C4 (Web Text))

Max Activating Token Index: 874

Text #11

Max Range: 1.9025. Min Range: -1.9025

Max Act: 0.9782. Min Act: -0.0000

Data Index: 364535 (C4 (Web Text))

Max Activating Token Index: 362

Text #12

Max Range: 1.9025. Min Range: -1.9025

Max Act: 0.9711. Min Act: -0.0000

Data Index: 53103 (C4 (Web Text))

Max Activating Token Index: 885

Text #13

Max Range: 1.9025. Min Range: -1.9025

Max Act: 0.9566. Min Act: 0.0000

Data Index: 817863 (C4 (Web Text))

Max Activating Token Index: 246

Text #14

Max Range: 1.9025. Min Range: -1.9025

Max Act: 1.0387. Min Act: -0.0000

Data Index: 1307802 (C4 (Web Text))

Max Activating Token Index: 616

Text #15

Max Range: 1.9025. Min Range: -1.9025

Max Act: 0.9602. Min Act: -0.0000

Data Index: 1486830 (Python Code)

Max Activating Token Index: 953

Text #16

Max Range: 1.9025. Min Range: -1.9025

Max Act: 0.9541. Min Act: -0.0000

Data Index: 666582 (C4 (Web Text))

Max Activating Token Index: 790

Text #17

Max Range: 1.9025. Min Range: -1.9025

Max Act: 0.8808. Min Act: -0.0000

Data Index: 248121 (C4 (Web Text))

Max Activating Token Index: 122

Text #18

Max Range: 1.9025. Min Range: -1.9025

Max Act: 0.8798. Min Act: -0.0000

Data Index: 726180 (C4 (Web Text))

Max Activating Token Index: 297

Text #19

Max Range: 1.9025. Min Range: -1.9025

Max Act: 0.7633. Min Act: -0.0000

Data Index: 1249780 (C4 (Web Text))

Max Activating Token Index: 594