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

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

Neuron 850 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: 1.1715. Min Range: -1.1715

Max Act: 1.1715. Min Act: -0.0000

Data Index: 912088 (C4 (Web Text))

Max Activating Token Index: 743

Text #1

Max Range: 1.1715. Min Range: -1.1715

Max Act: 1.1068. Min Act: -0.0000

Data Index: 466326 (C4 (Web Text))

Max Activating Token Index: 880

Text #2

Max Range: 1.1715. Min Range: -1.1715

Max Act: 1.0358. Min Act: -0.0000

Data Index: 1104029 (C4 (Web Text))

Max Activating Token Index: 996

Text #3

Max Range: 1.1715. Min Range: -1.1715

Max Act: 1.0067. Min Act: -0.0000

Data Index: 1375933 (Python Code)

Max Activating Token Index: 209

Text #4

Max Range: 1.1715. Min Range: -1.1715

Max Act: 1.0237. Min Act: -0.0000

Data Index: 35588 (C4 (Web Text))

Max Activating Token Index: 642

Text #5

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.9790. Min Act: -0.0000

Data Index: 559433 (C4 (Web Text))

Max Activating Token Index: 862

Text #6

Max Range: 1.1715. Min Range: -1.1715

Max Act: 1.0144. Min Act: -0.0000

Data Index: 686541 (C4 (Web Text))

Max Activating Token Index: 461

Text #7

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.9380. Min Act: -0.0000

Data Index: 1468172 (Python Code)

Max Activating Token Index: 417

Text #8

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.9623. Min Act: -0.0000

Data Index: 1391195 (Python Code)

Max Activating Token Index: 436

Text #9

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.9538. Min Act: -0.0000

Data Index: 1473670 (Python Code)

Max Activating Token Index: 522

Text #10

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.8579. Min Act: -0.0000

Data Index: 1225297 (C4 (Web Text))

Max Activating Token Index: 936

Text #11

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.8219. Min Act: -0.0000

Data Index: 1649848 (Python Code)

Max Activating Token Index: 597

Text #12

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.9108. Min Act: -0.0000

Data Index: 464491 (C4 (Web Text))

Max Activating Token Index: 468

Text #13

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.8679. Min Act: -0.0000

Data Index: 1371432 (Python Code)

Max Activating Token Index: 469

Text #14

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.8607. Min Act: -0.0000

Data Index: 47951 (C4 (Web Text))

Max Activating Token Index: 162

Text #15

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.8346. Min Act: -0.0000

Data Index: 1134487 (C4 (Web Text))

Max Activating Token Index: 915

Text #16

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.8277. Min Act: -0.0000

Data Index: 309753 (C4 (Web Text))

Max Activating Token Index: 605

Text #17

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.8042. Min Act: -0.0000

Data Index: 1076987 (C4 (Web Text))

Max Activating Token Index: 690

Text #18

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.8031. Min Act: -0.0000

Data Index: 1569685 (Python Code)

Max Activating Token Index: 105

Text #19

Max Range: 1.1715. Min Range: -1.1715

Max Act: 0.8019. Min Act: -0.0000

Data Index: 1431000 (Python Code)

Max Activating Token Index: 558