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

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

Neuron 130 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: 0.1637. Min Range: -0.1637

Max Act: 0.1637. Min Act: -0.0001

Data Index: 579029 (C4 (Web Text))

Max Activating Token Index: 225

Text #1

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1607. Min Act: -0.0001

Data Index: 1264626 (C4 (Web Text))

Max Activating Token Index: 577

Text #2

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1541. Min Act: -0.0001

Data Index: 1048715 (C4 (Web Text))

Max Activating Token Index: 133

Text #3

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1445. Min Act: -0.0001

Data Index: 1080970 (C4 (Web Text))

Max Activating Token Index: 721

Text #4

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1152. Min Act: -0.0001

Data Index: 1290425 (C4 (Web Text))

Max Activating Token Index: 130

Text #5

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.0980. Min Act: -0.0001

Data Index: 40085 (C4 (Web Text))

Max Activating Token Index: 170

Text #6

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.0999. Min Act: -0.0001

Data Index: 1485813 (Python Code)

Max Activating Token Index: 73

Text #7

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1006. Min Act: -0.0001

Data Index: 1048146 (C4 (Web Text))

Max Activating Token Index: 368

Text #8

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.0950. Min Act: -0.0001

Data Index: 1405490 (Python Code)

Max Activating Token Index: 95

Text #9

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1037. Min Act: -0.0001

Data Index: 1655247 (Python Code)

Max Activating Token Index: 81

Text #10

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.0995. Min Act: -0.0001

Data Index: 550369 (C4 (Web Text))

Max Activating Token Index: 706

Text #11

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.0937. Min Act: -0.0001

Data Index: 1493841 (Python Code)

Max Activating Token Index: 182

Text #12

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1013. Min Act: -0.0001

Data Index: 1153175 (C4 (Web Text))

Max Activating Token Index: 65

Text #13

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1098. Min Act: -0.0001

Data Index: 1334699 (C4 (Web Text))

Max Activating Token Index: 352

Text #14

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.0959. Min Act: -0.0001

Data Index: 1501552 (Python Code)

Max Activating Token Index: 75

Text #15

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.0910. Min Act: -0.0001

Data Index: 1520291 (Python Code)

Max Activating Token Index: 133

Text #16

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1002. Min Act: -0.0001

Data Index: 1530106 (Python Code)

Max Activating Token Index: 331

Text #17

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.0997. Min Act: -0.0001

Data Index: 1601310 (Python Code)

Max Activating Token Index: 110

Text #18

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1149. Min Act: -0.0001

Data Index: 22349 (C4 (Web Text))

Max Activating Token Index: 799

Text #19

Max Range: 0.1637. Min Range: -0.1637

Max Act: 0.1134. Min Act: -0.0001

Data Index: 1570251 (Python Code)

Max Activating Token Index: 228