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

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

Neuron 4852 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: 0.2250. Min Range: -0.2250

Max Act: 0.2250. Min Act: -0.0001

Data Index: 748733 (C4 (Web Text))

Max Activating Token Index: 727

Text #1

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.2058. Min Act: -0.0001

Data Index: 554030 (C4 (Web Text))

Max Activating Token Index: 71

Text #2

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.2113. Min Act: -0.0001

Data Index: 1215772 (C4 (Web Text))

Max Activating Token Index: 86

Text #3

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1937. Min Act: -0.0001

Data Index: 724456 (C4 (Web Text))

Max Activating Token Index: 670

Text #4

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.2062. Min Act: -0.0001

Data Index: 158179 (C4 (Web Text))

Max Activating Token Index: 28

Text #5

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1898. Min Act: -0.0001

Data Index: 228923 (C4 (Web Text))

Max Activating Token Index: 867

Text #6

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.2042. Min Act: -0.0001

Data Index: 899058 (C4 (Web Text))

Max Activating Token Index: 128

Text #7

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.2100. Min Act: -0.0001

Data Index: 885904 (C4 (Web Text))

Max Activating Token Index: 40

Text #8

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1914. Min Act: -0.0001

Data Index: 1059445 (C4 (Web Text))

Max Activating Token Index: 51

Text #9

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1929. Min Act: -0.0001

Data Index: 132944 (C4 (Web Text))

Max Activating Token Index: 390

Text #10

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1993. Min Act: -0.0001

Data Index: 648555 (C4 (Web Text))

Max Activating Token Index: 140

Text #11

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1989. Min Act: -0.0001

Data Index: 211283 (C4 (Web Text))

Max Activating Token Index: 42

Text #12

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.2010. Min Act: -0.0001

Data Index: 1079925 (C4 (Web Text))

Max Activating Token Index: 55

Text #13

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1927. Min Act: -0.0001

Data Index: 847880 (C4 (Web Text))

Max Activating Token Index: 33

Text #14

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1915. Min Act: -0.0001

Data Index: 512301 (C4 (Web Text))

Max Activating Token Index: 65

Text #15

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1906. Min Act: -0.0001

Data Index: 555019 (C4 (Web Text))

Max Activating Token Index: 84

Text #16

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1829. Min Act: -0.0001

Data Index: 1021238 (C4 (Web Text))

Max Activating Token Index: 349

Text #17

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1950. Min Act: -0.0001

Data Index: 524882 (C4 (Web Text))

Max Activating Token Index: 291

Text #18

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1896. Min Act: -0.0001

Data Index: 707993 (C4 (Web Text))

Max Activating Token Index: 689

Text #19

Max Range: 0.2250. Min Range: -0.2250

Max Act: 0.1892. Min Act: -0.0001

Data Index: 1274920 (C4 (Web Text))

Max Activating Token Index: 861