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

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

Neuron 4032 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: 0.6432. Min Range: -0.6432

Max Act: 0.6432. Min Act: -0.0000

Data Index: 1655167 (Python Code)

Max Activating Token Index: 903

Text #1

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.6057. Min Act: -0.0000

Data Index: 19137 (C4 (Web Text))

Max Activating Token Index: 789

Text #2

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.6092. Min Act: -0.0000

Data Index: 1425451 (Python Code)

Max Activating Token Index: 877

Text #3

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5779. Min Act: -0.0000

Data Index: 1519308 (Python Code)

Max Activating Token Index: 1015

Text #4

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5848. Min Act: -0.0000

Data Index: 1605806 (Python Code)

Max Activating Token Index: 931

Text #5

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5908. Min Act: -0.0000

Data Index: 1640244 (Python Code)

Max Activating Token Index: 578

Text #6

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5671. Min Act: -0.0000

Data Index: 1495329 (Python Code)

Max Activating Token Index: 874

Text #7

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5349. Min Act: -0.0000

Data Index: 1522995 (Python Code)

Max Activating Token Index: 572

Text #8

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5077. Min Act: -0.0000

Data Index: 1386815 (Python Code)

Max Activating Token Index: 903

Text #9

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5381. Min Act: -0.0000

Data Index: 1493452 (Python Code)

Max Activating Token Index: 796

Text #10

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5155. Min Act: -0.0000

Data Index: 1618833 (Python Code)

Max Activating Token Index: 131

Text #11

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5131. Min Act: -0.0000

Data Index: 1560473 (Python Code)

Max Activating Token Index: 428

Text #12

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5240. Min Act: -0.0000

Data Index: 1614752 (Python Code)

Max Activating Token Index: 253

Text #13

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5052. Min Act: -0.0000

Data Index: 634687 (C4 (Web Text))

Max Activating Token Index: 536

Text #14

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.4999. Min Act: -0.0000

Data Index: 622483 (C4 (Web Text))

Max Activating Token Index: 678

Text #15

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.4924. Min Act: -0.0000

Data Index: 1600269 (Python Code)

Max Activating Token Index: 195

Text #16

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.5083. Min Act: -0.0000

Data Index: 1599585 (Python Code)

Max Activating Token Index: 184

Text #17

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.4856. Min Act: -0.0000

Data Index: 1404449 (Python Code)

Max Activating Token Index: 73

Text #18

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.4872. Min Act: -0.0000

Data Index: 1533199 (Python Code)

Max Activating Token Index: 281

Text #19

Max Range: 0.6432. Min Range: -0.6432

Max Act: 0.4863. Min Act: -0.0000

Data Index: 1433181 (Python Code)

Max Activating Token Index: 690