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

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

Neuron 3866 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.2680. Min Act: -0.0003

Data Index: 1184195 (C4 (Web Text))

Max Activating Token Index: 767

Text #1

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.2519. Min Act: -0.0003

Data Index: 736588 (C4 (Web Text))

Max Activating Token Index: 486

Text #2

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.2401. Min Act: -0.0003

Data Index: 656671 (C4 (Web Text))

Max Activating Token Index: 769

Text #3

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.2349. Min Act: -0.0003

Data Index: 1626896 (Python Code)

Max Activating Token Index: 218

Text #4

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.2161. Min Act: -0.0003

Data Index: 11605 (C4 (Web Text))

Max Activating Token Index: 264

Text #5

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.2009. Min Act: -0.0003

Data Index: 606585 (C4 (Web Text))

Max Activating Token Index: 354

Text #6

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.2027. Min Act: -0.0003

Data Index: 1073246 (C4 (Web Text))

Max Activating Token Index: 1017

Text #7

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.2017. Min Act: -0.0003

Data Index: 467284 (C4 (Web Text))

Max Activating Token Index: 710

Text #8

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1941. Min Act: -0.0003

Data Index: 651989 (C4 (Web Text))

Max Activating Token Index: 153

Text #9

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1941. Min Act: -0.0003

Data Index: 754236 (C4 (Web Text))

Max Activating Token Index: 788

Text #10

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1797. Min Act: -0.0003

Data Index: 1422784 (Python Code)

Max Activating Token Index: 660

Text #11

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1890. Min Act: -0.0002

Data Index: 903593 (C4 (Web Text))

Max Activating Token Index: 142

Text #12

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1881. Min Act: -0.0002

Data Index: 1052341 (C4 (Web Text))

Max Activating Token Index: 189

Text #13

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1869. Min Act: -0.0003

Data Index: 1505355 (Python Code)

Max Activating Token Index: 47

Text #14

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1847. Min Act: -0.0002

Data Index: 1570349 (Python Code)

Max Activating Token Index: 481

Text #15

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1832. Min Act: -0.0003

Data Index: 283174 (C4 (Web Text))

Max Activating Token Index: 168

Text #16

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1824. Min Act: -0.0003

Data Index: 1448118 (Python Code)

Max Activating Token Index: 81

Text #17

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1875. Min Act: -0.0002

Data Index: 1065360 (C4 (Web Text))

Max Activating Token Index: 687

Text #18

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1802. Min Act: -0.0003

Data Index: 1079830 (C4 (Web Text))

Max Activating Token Index: 191

Text #19

Max Range: 0.2680. Min Range: -0.2680

Max Act: 0.1784. Min Act: -0.0003

Data Index: 1409318 (Python Code)

Max Activating Token Index: 7