Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 4032 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.2830. Min Act: -0.0001

Data Index: 1343649 (C4 (Web Text))

Max Activating Token Index: 224

Text #1

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.2174. Min Act: -0.0001

Data Index: 1519125 (Python Code)

Max Activating Token Index: 42

Text #2

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.2115. Min Act: -0.0001

Data Index: 1251849 (C4 (Web Text))

Max Activating Token Index: 42

Text #3

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1960. Min Act: -0.0001

Data Index: 24081 (C4 (Web Text))

Max Activating Token Index: 372

Text #4

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1754. Min Act: -0.0001

Data Index: 1183894 (C4 (Web Text))

Max Activating Token Index: 529

Text #5

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1837. Min Act: -0.0001

Data Index: 1390297 (Python Code)

Max Activating Token Index: 480

Text #6

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1826. Min Act: -0.0001

Data Index: 193515 (C4 (Web Text))

Max Activating Token Index: 479

Text #7

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1788. Min Act: -0.0001

Data Index: 1442370 (Python Code)

Max Activating Token Index: 946

Text #8

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1777. Min Act: -0.0001

Data Index: 844930 (C4 (Web Text))

Max Activating Token Index: 401

Text #9

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1650. Min Act: -0.0001

Data Index: 841469 (C4 (Web Text))

Max Activating Token Index: 509

Text #10

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1757. Min Act: -0.0001

Data Index: 1508978 (Python Code)

Max Activating Token Index: 35

Text #11

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1618. Min Act: -0.0001

Data Index: 359279 (C4 (Web Text))

Max Activating Token Index: 502

Text #12

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1730. Min Act: -0.0001

Data Index: 874979 (C4 (Web Text))

Max Activating Token Index: 496

Text #13

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1715. Min Act: -0.0001

Data Index: 1004046 (C4 (Web Text))

Max Activating Token Index: 121

Text #14

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1914. Min Act: -0.0001

Data Index: 1352361 (C4 (Web Text))

Max Activating Token Index: 591

Text #15

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1611. Min Act: -0.0001

Data Index: 1529386 (Python Code)

Max Activating Token Index: 855

Text #16

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1489. Min Act: -0.0001

Data Index: 1014956 (C4 (Web Text))

Max Activating Token Index: 150

Text #17

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1416. Min Act: -0.0001

Data Index: 905022 (C4 (Web Text))

Max Activating Token Index: 375

Text #18

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1599. Min Act: -0.0001

Data Index: 1547230 (Python Code)

Max Activating Token Index: 997

Text #19

Max Range: 0.2830. Min Range: -0.2830

Max Act: 0.1538. Min Act: -0.0001

Data Index: 1565103 (Python Code)

Max Activating Token Index: 246