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

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

Neuron 967 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: 0.2008. Min Range: -0.2008

Max Act: 0.2008. Min Act: 0.0000

Data Index: 300485 (C4 (Web Text))

Max Activating Token Index: 791

Text #1

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.1304. Min Act: -0.0000

Data Index: 1156984 (C4 (Web Text))

Max Activating Token Index: 718

Text #2

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.1097. Min Act: -0.0000

Data Index: 1326047 (C4 (Web Text))

Max Activating Token Index: 419

Text #3

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.1074. Min Act: -0.0000

Data Index: 1384611 (Python Code)

Max Activating Token Index: 95

Text #4

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.1069. Min Act: -0.0000

Data Index: 1313833 (C4 (Web Text))

Max Activating Token Index: 72

Text #5

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0984. Min Act: -0.0000

Data Index: 122293 (C4 (Web Text))

Max Activating Token Index: 358

Text #6

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0973. Min Act: 0.0001

Data Index: 908276 (C4 (Web Text))

Max Activating Token Index: 13

Text #7

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0987. Min Act: -0.0000

Data Index: 1163426 (C4 (Web Text))

Max Activating Token Index: 37

Text #8

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0917. Min Act: -0.0000

Data Index: 521460 (C4 (Web Text))

Max Activating Token Index: 245

Text #9

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0862. Min Act: -0.0000

Data Index: 1169837 (C4 (Web Text))

Max Activating Token Index: 692

Text #10

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0859. Min Act: -0.0000

Data Index: 1520512 (Python Code)

Max Activating Token Index: 99

Text #11

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0855. Min Act: -0.0000

Data Index: 28777 (C4 (Web Text))

Max Activating Token Index: 435

Text #12

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0860. Min Act: -0.0000

Data Index: 1500555 (Python Code)

Max Activating Token Index: 83

Text #13

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0900. Min Act: 0.0000

Data Index: 774643 (C4 (Web Text))

Max Activating Token Index: 319

Text #14

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0832. Min Act: -0.0000

Data Index: 301531 (C4 (Web Text))

Max Activating Token Index: 253

Text #15

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0840. Min Act: 0.0000

Data Index: 1060053 (C4 (Web Text))

Max Activating Token Index: 957

Text #16

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0815. Min Act: 0.0000

Data Index: 735007 (C4 (Web Text))

Max Activating Token Index: 529

Text #17

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0868. Min Act: -0.0000

Data Index: 421820 (C4 (Web Text))

Max Activating Token Index: 951

Text #18

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0852. Min Act: -0.0000

Data Index: 1608815 (Python Code)

Max Activating Token Index: 556

Text #19

Max Range: 0.2008. Min Range: -0.2008

Max Act: 0.0829. Min Act: 0.0000

Data Index: 1062139 (C4 (Web Text))

Max Activating Token Index: 524