Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3681 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.2856. Min Act: -0.0001

Data Index: 195599 (C4 (Web Text))

Max Activating Token Index: 144

Text #1

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.2998. Min Act: -0.0001

Data Index: 687598 (C4 (Web Text))

Max Activating Token Index: 259

Text #2

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.2148. Min Act: -0.0001

Data Index: 830196 (C4 (Web Text))

Max Activating Token Index: 722

Text #3

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1952. Min Act: -0.0001

Data Index: 1366303 (Python Code)

Max Activating Token Index: 630

Text #4

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.2083. Min Act: -0.0001

Data Index: 1499180 (Python Code)

Max Activating Token Index: 1001

Text #5

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.2017. Min Act: -0.0001

Data Index: 1111609 (C4 (Web Text))

Max Activating Token Index: 159

Text #6

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.2016. Min Act: -0.0001

Data Index: 568417 (C4 (Web Text))

Max Activating Token Index: 978

Text #7

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1972. Min Act: -0.0001

Data Index: 412242 (C4 (Web Text))

Max Activating Token Index: 907

Text #8

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1804. Min Act: -0.0001

Data Index: 1583749 (Python Code)

Max Activating Token Index: 222

Text #9

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1835. Min Act: -0.0001

Data Index: 919722 (C4 (Web Text))

Max Activating Token Index: 25

Text #10

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1766. Min Act: -0.0001

Data Index: 1524217 (Python Code)

Max Activating Token Index: 872

Text #11

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1705. Min Act: -0.0001

Data Index: 1420876 (Python Code)

Max Activating Token Index: 594

Text #12

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1608. Min Act: -0.0001

Data Index: 594826 (C4 (Web Text))

Max Activating Token Index: 104

Text #13

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1717. Min Act: -0.0001

Data Index: 1541752 (Python Code)

Max Activating Token Index: 258

Text #14

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1781. Min Act: -0.0001

Data Index: 1407137 (Python Code)

Max Activating Token Index: 462

Text #15

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1703. Min Act: -0.0001

Data Index: 1547899 (Python Code)

Max Activating Token Index: 201

Text #16

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1738. Min Act: -0.0001

Data Index: 1518752 (Python Code)

Max Activating Token Index: 53

Text #17

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1619. Min Act: -0.0001

Data Index: 1625280 (Python Code)

Max Activating Token Index: 39

Text #18

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1715. Min Act: -0.0001

Data Index: 210349 (C4 (Web Text))

Max Activating Token Index: 642

Text #19

Max Range: 0.2998. Min Range: -0.2998

Max Act: 0.1634. Min Act: -0.0001

Data Index: 492504 (C4 (Web Text))

Max Activating Token Index: 374