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

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

Neuron 3545 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.2737. Min Act: -0.0000

Data Index: 1302465 (C4 (Web Text))

Max Activating Token Index: 755

Text #1

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.2590. Min Act: -0.0000

Data Index: 1182688 (C4 (Web Text))

Max Activating Token Index: 951

Text #2

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.2507. Min Act: -0.0000

Data Index: 1096001 (C4 (Web Text))

Max Activating Token Index: 208

Text #3

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.2026. Min Act: -0.0000

Data Index: 1161640 (C4 (Web Text))

Max Activating Token Index: 455

Text #4

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.2005. Min Act: -0.0000

Data Index: 1555852 (Python Code)

Max Activating Token Index: 983

Text #5

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.2099. Min Act: -0.0000

Data Index: 173058 (C4 (Web Text))

Max Activating Token Index: 798

Text #6

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.2202. Min Act: -0.0000

Data Index: 1243011 (C4 (Web Text))

Max Activating Token Index: 891

Text #7

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.1991. Min Act: -0.0000

Data Index: 1329518 (C4 (Web Text))

Max Activating Token Index: 910

Text #8

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.2062. Min Act: -0.0000

Data Index: 1627868 (Python Code)

Max Activating Token Index: 218

Text #9

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.2217. Min Act: -0.0000

Data Index: 1632724 (Python Code)

Max Activating Token Index: 546

Text #10

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.1975. Min Act: -0.0000

Data Index: 680713 (C4 (Web Text))

Max Activating Token Index: 484

Text #11

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.1970. Min Act: -0.0000

Data Index: 1511065 (Python Code)

Max Activating Token Index: 541

Text #12

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.2119. Min Act: 0.0000

Data Index: 261064 (C4 (Web Text))

Max Activating Token Index: 675

Text #13

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.1997. Min Act: -0.0000

Data Index: 1383455 (Python Code)

Max Activating Token Index: 816

Text #14

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.1907. Min Act: -0.0000

Data Index: 393696 (C4 (Web Text))

Max Activating Token Index: 542

Text #15

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.1883. Min Act: -0.0000

Data Index: 1464332 (Python Code)

Max Activating Token Index: 996

Text #16

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.1937. Min Act: -0.0000

Data Index: 1438333 (Python Code)

Max Activating Token Index: 649

Text #17

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.1805. Min Act: -0.0000

Data Index: 1432283 (Python Code)

Max Activating Token Index: 975

Text #18

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.1868. Min Act: -0.0000

Data Index: 43682 (C4 (Web Text))

Max Activating Token Index: 644

Text #19

Max Range: 0.2737. Min Range: -0.2737

Max Act: 0.1840. Min Act: -0.0000

Data Index: 1013277 (C4 (Web Text))

Max Activating Token Index: 508