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

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

Neuron 1765 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3792. Min Range: -3.3792

Max Act: 3.3792. Min Act: -0.0000

Data Index: 921499 (C4 (Web Text))

Max Activating Token Index: 600

Text #1

Max Range: 3.3792. Min Range: -3.3792

Max Act: 2.8183. Min Act: -0.0000

Data Index: 725556 (C4 (Web Text))

Max Activating Token Index: 945

Text #2

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.5848. Min Act: -0.0000

Data Index: 574160 (C4 (Web Text))

Max Activating Token Index: 242

Text #3

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.7153. Min Act: -0.0000

Data Index: 684298 (C4 (Web Text))

Max Activating Token Index: 661

Text #4

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.4067. Min Act: -0.0000

Data Index: 632045 (C4 (Web Text))

Max Activating Token Index: 766

Text #5

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.3651. Min Act: -0.0000

Data Index: 220893 (C4 (Web Text))

Max Activating Token Index: 488

Text #6

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.3095. Min Act: -0.0000

Data Index: 568608 (C4 (Web Text))

Max Activating Token Index: 406

Text #7

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.2510. Min Act: -0.0000

Data Index: 418498 (C4 (Web Text))

Max Activating Token Index: 703

Text #8

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.2846. Min Act: -0.0000

Data Index: 13352 (C4 (Web Text))

Max Activating Token Index: 380

Text #9

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.3187. Min Act: -0.0000

Data Index: 1220497 (C4 (Web Text))

Max Activating Token Index: 152

Text #10

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.1755. Min Act: -0.0000

Data Index: 1364232 (Python Code)

Max Activating Token Index: 873

Text #11

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.2236. Min Act: -0.0000

Data Index: 1389039 (Python Code)

Max Activating Token Index: 471

Text #12

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.1659. Min Act: -0.0000

Data Index: 370742 (C4 (Web Text))

Max Activating Token Index: 493

Text #13

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.2972. Min Act: -0.0000

Data Index: 833629 (C4 (Web Text))

Max Activating Token Index: 549

Text #14

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.1275. Min Act: -0.0000

Data Index: 1650802 (Python Code)

Max Activating Token Index: 837

Text #15

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.0182. Min Act: -0.0000

Data Index: 63974 (C4 (Web Text))

Max Activating Token Index: 294

Text #16

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.0550. Min Act: -0.0000

Data Index: 800590 (C4 (Web Text))

Max Activating Token Index: 560

Text #17

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.0820. Min Act: -0.0000

Data Index: 1007047 (C4 (Web Text))

Max Activating Token Index: 722

Text #18

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.0356. Min Act: -0.0000

Data Index: 483995 (C4 (Web Text))

Max Activating Token Index: 335

Text #19

Max Range: 3.3792. Min Range: -3.3792

Max Act: 1.0792. Min Act: -0.0000

Data Index: 1229770 (C4 (Web Text))

Max Activating Token Index: 633