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

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

Neuron 4185 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: 1.5039. Min Range: -1.5039

Max Act: 1.4987. Min Act: -0.0000

Data Index: 597226 (C4 (Web Text))

Max Activating Token Index: 243

Text #1

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.3594. Min Act: -0.0000

Data Index: 84308 (C4 (Web Text))

Max Activating Token Index: 740

Text #2

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.3379. Min Act: -0.0000

Data Index: 918295 (C4 (Web Text))

Max Activating Token Index: 172

Text #3

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.3391. Min Act: -0.0000

Data Index: 387158 (C4 (Web Text))

Max Activating Token Index: 213

Text #4

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.5039. Min Act: -0.0000

Data Index: 1293109 (C4 (Web Text))

Max Activating Token Index: 698

Text #5

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.2777. Min Act: -0.0000

Data Index: 1378462 (Python Code)

Max Activating Token Index: 388

Text #6

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.2784. Min Act: -0.0000

Data Index: 1173532 (C4 (Web Text))

Max Activating Token Index: 413

Text #7

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.2052. Min Act: -0.0000

Data Index: 1052729 (C4 (Web Text))

Max Activating Token Index: 561

Text #8

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.1886. Min Act: -0.0000

Data Index: 1376897 (Python Code)

Max Activating Token Index: 959

Text #9

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.1518. Min Act: -0.0000

Data Index: 1386893 (Python Code)

Max Activating Token Index: 668

Text #10

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.1825. Min Act: -0.0000

Data Index: 1085783 (C4 (Web Text))

Max Activating Token Index: 108

Text #11

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.1423. Min Act: -0.0000

Data Index: 1033122 (C4 (Web Text))

Max Activating Token Index: 106

Text #12

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.1820. Min Act: -0.0000

Data Index: 1438060 (Python Code)

Max Activating Token Index: 33

Text #13

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.1899. Min Act: -0.0000

Data Index: 1018197 (C4 (Web Text))

Max Activating Token Index: 674

Text #14

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.2080. Min Act: -0.0000

Data Index: 694131 (C4 (Web Text))

Max Activating Token Index: 180

Text #15

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.2007. Min Act: -0.0000

Data Index: 1243489 (C4 (Web Text))

Max Activating Token Index: 217

Text #16

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.1360. Min Act: -0.0000

Data Index: 1492997 (Python Code)

Max Activating Token Index: 90

Text #17

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.0694. Min Act: -0.0000

Data Index: 1522222 (Python Code)

Max Activating Token Index: 864

Text #18

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.0879. Min Act: -0.0000

Data Index: 1014577 (C4 (Web Text))

Max Activating Token Index: 1000

Text #19

Max Range: 1.5039. Min Range: -1.5039

Max Act: 1.1528. Min Act: -0.0000

Data Index: 516948 (C4 (Web Text))

Max Activating Token Index: 368