Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 850 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.2260. Min Act: -0.0006

Data Index: 1123012 (C4 (Web Text))

Max Activating Token Index: 903

Text #1

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.2004. Min Act: -0.0004

Data Index: 38442 (C4 (Web Text))

Max Activating Token Index: 848

Text #2

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1924. Min Act: -0.0004

Data Index: 1153470 (C4 (Web Text))

Max Activating Token Index: 677

Text #3

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1905. Min Act: -0.0006

Data Index: 978418 (C4 (Web Text))

Max Activating Token Index: 577

Text #4

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1818. Min Act: -0.0004

Data Index: 265300 (C4 (Web Text))

Max Activating Token Index: 916

Text #5

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1873. Min Act: -0.0004

Data Index: 580956 (C4 (Web Text))

Max Activating Token Index: 964

Text #6

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1866. Min Act: -0.0006

Data Index: 1299192 (C4 (Web Text))

Max Activating Token Index: 919

Text #7

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1865. Min Act: -0.0006

Data Index: 364637 (C4 (Web Text))

Max Activating Token Index: 172

Text #8

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1840. Min Act: -0.0004

Data Index: 887049 (C4 (Web Text))

Max Activating Token Index: 328

Text #9

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1827. Min Act: -0.0004

Data Index: 554926 (C4 (Web Text))

Max Activating Token Index: 762

Text #10

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1754. Min Act: -0.0004

Data Index: 353483 (C4 (Web Text))

Max Activating Token Index: 865

Text #11

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1711. Min Act: -0.0004

Data Index: 1029301 (C4 (Web Text))

Max Activating Token Index: 548

Text #12

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1744. Min Act: -0.0004

Data Index: 899750 (C4 (Web Text))

Max Activating Token Index: 863

Text #13

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1723. Min Act: -0.0006

Data Index: 729483 (C4 (Web Text))

Max Activating Token Index: 270

Text #14

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1709. Min Act: -0.0004

Data Index: 343082 (C4 (Web Text))

Max Activating Token Index: 771

Text #15

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1691. Min Act: -0.0006

Data Index: 1303361 (C4 (Web Text))

Max Activating Token Index: 951

Text #16

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1709. Min Act: -0.0006

Data Index: 849287 (C4 (Web Text))

Max Activating Token Index: 485

Text #17

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1742. Min Act: -0.0004

Data Index: 210776 (C4 (Web Text))

Max Activating Token Index: 370

Text #18

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1668. Min Act: -0.0003

Data Index: 315826 (C4 (Web Text))

Max Activating Token Index: 195

Text #19

Max Range: 0.2260. Min Range: -0.2260

Max Act: 0.1662. Min Act: -0.0004

Data Index: 164678 (C4 (Web Text))

Max Activating Token Index: 536