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

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

Neuron 4737 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1999. Min Range: -1.1999

Max Act: 1.1999. Min Act: -0.0001

Data Index: 506041 (C4 (Web Text))

Max Activating Token Index: 121

Text #1

Max Range: 1.1999. Min Range: -1.1999

Max Act: 1.0796. Min Act: -0.0001

Data Index: 594771 (C4 (Web Text))

Max Activating Token Index: 156

Text #2

Max Range: 1.1999. Min Range: -1.1999

Max Act: 1.0823. Min Act: -0.0001

Data Index: 31208 (C4 (Web Text))

Max Activating Token Index: 399

Text #3

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.9542. Min Act: -0.0001

Data Index: 270220 (C4 (Web Text))

Max Activating Token Index: 652

Text #4

Max Range: 1.1999. Min Range: -1.1999

Max Act: 1.0085. Min Act: -0.0001

Data Index: 629542 (C4 (Web Text))

Max Activating Token Index: 287

Text #5

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.9640. Min Act: -0.0001

Data Index: 721577 (C4 (Web Text))

Max Activating Token Index: 1019

Text #6

Max Range: 1.1999. Min Range: -1.1999

Max Act: 1.0054. Min Act: -0.0001

Data Index: 580800 (C4 (Web Text))

Max Activating Token Index: 999

Text #7

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.9763. Min Act: -0.0001

Data Index: 1653055 (Python Code)

Max Activating Token Index: 41

Text #8

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.8541. Min Act: -0.0001

Data Index: 1601225 (Python Code)

Max Activating Token Index: 128

Text #9

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.9760. Min Act: -0.0001

Data Index: 437766 (C4 (Web Text))

Max Activating Token Index: 963

Text #10

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.9069. Min Act: -0.0001

Data Index: 1621756 (Python Code)

Max Activating Token Index: 47

Text #11

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.9008. Min Act: -0.0001

Data Index: 1392897 (Python Code)

Max Activating Token Index: 972

Text #12

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.9235. Min Act: -0.0001

Data Index: 176742 (C4 (Web Text))

Max Activating Token Index: 621

Text #13

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.8400. Min Act: -0.0001

Data Index: 1281370 (C4 (Web Text))

Max Activating Token Index: 731

Text #14

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.9506. Min Act: -0.0001

Data Index: 1641204 (Python Code)

Max Activating Token Index: 68

Text #15

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.8833. Min Act: -0.0001

Data Index: 1268945 (C4 (Web Text))

Max Activating Token Index: 437

Text #16

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.9274. Min Act: -0.0001

Data Index: 521352 (C4 (Web Text))

Max Activating Token Index: 521

Text #17

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.9061. Min Act: -0.0001

Data Index: 815737 (C4 (Web Text))

Max Activating Token Index: 854

Text #18

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.7761. Min Act: -0.0001

Data Index: 1386589 (Python Code)

Max Activating Token Index: 844

Text #19

Max Range: 1.1999. Min Range: -1.1999

Max Act: 0.8706. Min Act: -0.0001

Data Index: 1471943 (Python Code)

Max Activating Token Index: 350