Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 2062 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.2662. Min Act: -0.0001

Data Index: 618359 (C4 (Web Text))

Max Activating Token Index: 354

Text #1

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1989. Min Act: -0.0001

Data Index: 632376 (C4 (Web Text))

Max Activating Token Index: 590

Text #2

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.2003. Min Act: -0.0001

Data Index: 652494 (C4 (Web Text))

Max Activating Token Index: 246

Text #3

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.2200. Min Act: -0.0001

Data Index: 941350 (C4 (Web Text))

Max Activating Token Index: 269

Text #4

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1986. Min Act: -0.0001

Data Index: 1396354 (Python Code)

Max Activating Token Index: 568

Text #5

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1899. Min Act: -0.0001

Data Index: 1017061 (C4 (Web Text))

Max Activating Token Index: 582

Text #6

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1782. Min Act: -0.0001

Data Index: 596769 (C4 (Web Text))

Max Activating Token Index: 533

Text #7

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1943. Min Act: -0.0001

Data Index: 1311300 (C4 (Web Text))

Max Activating Token Index: 787

Text #8

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1905. Min Act: -0.0001

Data Index: 214192 (C4 (Web Text))

Max Activating Token Index: 275

Text #9

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1739. Min Act: -0.0001

Data Index: 59809 (C4 (Web Text))

Max Activating Token Index: 118

Text #10

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1728. Min Act: -0.0001

Data Index: 169794 (C4 (Web Text))

Max Activating Token Index: 514

Text #11

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1654. Min Act: -0.0001

Data Index: 973448 (C4 (Web Text))

Max Activating Token Index: 43

Text #12

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1804. Min Act: -0.0001

Data Index: 55187 (C4 (Web Text))

Max Activating Token Index: 362

Text #13

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1825. Min Act: -0.0001

Data Index: 1153215 (C4 (Web Text))

Max Activating Token Index: 809

Text #14

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1642. Min Act: -0.0001

Data Index: 921533 (C4 (Web Text))

Max Activating Token Index: 873

Text #15

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1617. Min Act: -0.0001

Data Index: 1200105 (C4 (Web Text))

Max Activating Token Index: 670

Text #16

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1670. Min Act: -0.0001

Data Index: 1228312 (C4 (Web Text))

Max Activating Token Index: 98

Text #17

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1653. Min Act: -0.0001

Data Index: 1633679 (Python Code)

Max Activating Token Index: 115

Text #18

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1649. Min Act: -0.0001

Data Index: 52790 (C4 (Web Text))

Max Activating Token Index: 269

Text #19

Max Range: 0.2662. Min Range: -0.2662

Max Act: 0.1667. Min Act: -0.0001

Data Index: 193286 (C4 (Web Text))

Max Activating Token Index: 817