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

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

Neuron 2545 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.2589. Min Act: -0.0001

Data Index: 1321181 (C4 (Web Text))

Max Activating Token Index: 868

Text #1

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.2648. Min Act: -0.0000

Data Index: 1459776 (Python Code)

Max Activating Token Index: 143

Text #2

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.2767. Min Act: -0.0001

Data Index: 1456079 (Python Code)

Max Activating Token Index: 639

Text #3

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.2585. Min Act: -0.0000

Data Index: 1368777 (Python Code)

Max Activating Token Index: 701

Text #4

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.2421. Min Act: -0.0001

Data Index: 879979 (C4 (Web Text))

Max Activating Token Index: 523

Text #5

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.2299. Min Act: -0.0001

Data Index: 629818 (C4 (Web Text))

Max Activating Token Index: 582

Text #6

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.2398. Min Act: -0.0001

Data Index: 1318235 (C4 (Web Text))

Max Activating Token Index: 795

Text #7

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.1912. Min Act: -0.0001

Data Index: 1057056 (C4 (Web Text))

Max Activating Token Index: 121

Text #8

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.2028. Min Act: -0.0001

Data Index: 1529587 (Python Code)

Max Activating Token Index: 577

Text #9

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.1995. Min Act: -0.0001

Data Index: 743157 (C4 (Web Text))

Max Activating Token Index: 971

Text #10

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.2112. Min Act: -0.0001

Data Index: 409901 (C4 (Web Text))

Max Activating Token Index: 333

Text #11

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.1825. Min Act: -0.0001

Data Index: 855108 (C4 (Web Text))

Max Activating Token Index: 169

Text #12

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.2002. Min Act: -0.0001

Data Index: 1376132 (Python Code)

Max Activating Token Index: 649

Text #13

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.1942. Min Act: -0.0000

Data Index: 1424185 (Python Code)

Max Activating Token Index: 619

Text #14

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.1683. Min Act: -0.0001

Data Index: 1549344 (Python Code)

Max Activating Token Index: 306

Text #15

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.1857. Min Act: -0.0000

Data Index: 1554834 (Python Code)

Max Activating Token Index: 508

Text #16

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.1900. Min Act: -0.0000

Data Index: 1211713 (C4 (Web Text))

Max Activating Token Index: 358

Text #17

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.1824. Min Act: -0.0000

Data Index: 1489565 (Python Code)

Max Activating Token Index: 603

Text #18

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.1868. Min Act: -0.0000

Data Index: 1093322 (C4 (Web Text))

Max Activating Token Index: 830

Text #19

Max Range: 0.2767. Min Range: -0.2767

Max Act: 0.1839. Min Act: -0.0001

Data Index: 281599 (C4 (Web Text))

Max Activating Token Index: 976