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

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

Neuron 3144 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: 0.6239. Min Range: -0.6239

Max Act: 0.6239. Min Act: -0.0000

Data Index: 1545047 (Python Code)

Max Activating Token Index: 369

Text #1

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.5114. Min Act: -0.0000

Data Index: 972046 (C4 (Web Text))

Max Activating Token Index: 603

Text #2

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.4934. Min Act: -0.0000

Data Index: 765795 (C4 (Web Text))

Max Activating Token Index: 50

Text #3

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.4358. Min Act: -0.0000

Data Index: 1098944 (C4 (Web Text))

Max Activating Token Index: 256

Text #4

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.4415. Min Act: -0.0000

Data Index: 1324093 (C4 (Web Text))

Max Activating Token Index: 47

Text #5

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.4188. Min Act: -0.0000

Data Index: 1498658 (Python Code)

Max Activating Token Index: 425

Text #6

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.3825. Min Act: -0.0000

Data Index: 372286 (C4 (Web Text))

Max Activating Token Index: 867

Text #7

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.4200. Min Act: -0.0000

Data Index: 673648 (C4 (Web Text))

Max Activating Token Index: 414

Text #8

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.3966. Min Act: -0.0000

Data Index: 1191334 (C4 (Web Text))

Max Activating Token Index: 91

Text #9

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.3648. Min Act: -0.0000

Data Index: 1236876 (C4 (Web Text))

Max Activating Token Index: 914

Text #10

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.3815. Min Act: -0.0000

Data Index: 100691 (C4 (Web Text))

Max Activating Token Index: 523

Text #11

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.4093. Min Act: -0.0000

Data Index: 246237 (C4 (Web Text))

Max Activating Token Index: 680

Text #12

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.4084. Min Act: -0.0000

Data Index: 777918 (C4 (Web Text))

Max Activating Token Index: 593

Text #13

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.4037. Min Act: -0.0000

Data Index: 685802 (C4 (Web Text))

Max Activating Token Index: 755

Text #14

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.4030. Min Act: -0.0000

Data Index: 854727 (C4 (Web Text))

Max Activating Token Index: 697

Text #15

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.3762. Min Act: -0.0000

Data Index: 1457639 (Python Code)

Max Activating Token Index: 45

Text #16

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.3970. Min Act: -0.0000

Data Index: 670018 (C4 (Web Text))

Max Activating Token Index: 833

Text #17

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.3678. Min Act: -0.0000

Data Index: 754494 (C4 (Web Text))

Max Activating Token Index: 245

Text #18

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.4035. Min Act: -0.0000

Data Index: 735617 (C4 (Web Text))

Max Activating Token Index: 209

Text #19

Max Range: 0.6239. Min Range: -0.6239

Max Act: 0.3614. Min Act: -0.0000

Data Index: 1077233 (C4 (Web Text))

Max Activating Token Index: 762