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

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

Neuron 3444 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.9143. Min Act: -0.0000

Data Index: 1500478 (Python Code)

Max Activating Token Index: 11

Text #1

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.8871. Min Act: -0.0000

Data Index: 1513109 (Python Code)

Max Activating Token Index: 10

Text #2

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.9063. Min Act: -0.0000

Data Index: 1507240 (Python Code)

Max Activating Token Index: 30

Text #3

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.8404. Min Act: -0.0000

Data Index: 1627952 (Python Code)

Max Activating Token Index: 24

Text #4

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.8031. Min Act: -0.0000

Data Index: 1386464 (Python Code)

Max Activating Token Index: 6

Text #5

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.7460. Min Act: -0.0000

Data Index: 1488209 (Python Code)

Max Activating Token Index: 15

Text #6

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.7009. Min Act: -0.0000

Data Index: 1427991 (Python Code)

Max Activating Token Index: 10

Text #7

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.6527. Min Act: -0.0000

Data Index: 885784 (C4 (Web Text))

Max Activating Token Index: 6

Text #8

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.6726. Min Act: -0.0000

Data Index: 1379500 (Python Code)

Max Activating Token Index: 8

Text #9

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.6724. Min Act: -0.0000

Data Index: 1622237 (Python Code)

Max Activating Token Index: 67

Text #10

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.6759. Min Act: -0.0000

Data Index: 1601561 (Python Code)

Max Activating Token Index: 34

Text #11

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.6377. Min Act: -0.0000

Data Index: 1384884 (Python Code)

Max Activating Token Index: 5

Text #12

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.6310. Min Act: -0.0000

Data Index: 1425924 (Python Code)

Max Activating Token Index: 54

Text #13

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.6114. Min Act: -0.0000

Data Index: 209049 (C4 (Web Text))

Max Activating Token Index: 777

Text #14

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.6220. Min Act: -0.0000

Data Index: 1463830 (Python Code)

Max Activating Token Index: 21

Text #15

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.6118. Min Act: -0.0000

Data Index: 1308153 (C4 (Web Text))

Max Activating Token Index: 1015

Text #16

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.5805. Min Act: -0.0000

Data Index: 521579 (C4 (Web Text))

Max Activating Token Index: 9

Text #17

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.5781. Min Act: -0.0000

Data Index: 1138567 (C4 (Web Text))

Max Activating Token Index: 888

Text #18

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.5422. Min Act: -0.0000

Data Index: 273769 (C4 (Web Text))

Max Activating Token Index: 16

Text #19

Max Range: 0.9143. Min Range: -0.9143

Max Act: 0.5749. Min Act: -0.0000

Data Index: 1032352 (C4 (Web Text))

Max Activating Token Index: 622