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

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

Neuron 3630 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1589. Min Act: -0.0002

Data Index: 1130197 (C4 (Web Text))

Max Activating Token Index: 722

Text #1

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1824. Min Act: -0.0002

Data Index: 346149 (C4 (Web Text))

Max Activating Token Index: 219

Text #2

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1497. Min Act: -0.0002

Data Index: 1308396 (C4 (Web Text))

Max Activating Token Index: 791

Text #3

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1602. Min Act: -0.0002

Data Index: 1139930 (C4 (Web Text))

Max Activating Token Index: 78

Text #4

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1596. Min Act: -0.0002

Data Index: 385020 (C4 (Web Text))

Max Activating Token Index: 104

Text #5

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1539. Min Act: -0.0002

Data Index: 410272 (C4 (Web Text))

Max Activating Token Index: 648

Text #6

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1370. Min Act: -0.0002

Data Index: 401809 (C4 (Web Text))

Max Activating Token Index: 760

Text #7

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1542. Min Act: -0.0002

Data Index: 839860 (C4 (Web Text))

Max Activating Token Index: 912

Text #8

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1543. Min Act: -0.0002

Data Index: 1207129 (C4 (Web Text))

Max Activating Token Index: 308

Text #9

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1676. Min Act: -0.0002

Data Index: 956605 (C4 (Web Text))

Max Activating Token Index: 492

Text #10

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1700. Min Act: -0.0002

Data Index: 456195 (C4 (Web Text))

Max Activating Token Index: 695

Text #11

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1507. Min Act: -0.0002

Data Index: 466816 (C4 (Web Text))

Max Activating Token Index: 816

Text #12

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1583. Min Act: -0.0002

Data Index: 282071 (C4 (Web Text))

Max Activating Token Index: 838

Text #13

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.0974. Min Act: -0.0002

Data Index: 891020 (C4 (Web Text))

Max Activating Token Index: 679

Text #14

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1723. Min Act: -0.0002

Data Index: 983287 (C4 (Web Text))

Max Activating Token Index: 884

Text #15

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1498. Min Act: -0.0002

Data Index: 1059674 (C4 (Web Text))

Max Activating Token Index: 602

Text #16

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1400. Min Act: -0.0002

Data Index: 310614 (C4 (Web Text))

Max Activating Token Index: 924

Text #17

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1530. Min Act: -0.0002

Data Index: 1227684 (C4 (Web Text))

Max Activating Token Index: 248

Text #18

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1594. Min Act: -0.0002

Data Index: 421388 (C4 (Web Text))

Max Activating Token Index: 63

Text #19

Max Range: 0.1824. Min Range: -0.1824

Max Act: 0.1515. Min Act: -0.0002

Data Index: 84560 (C4 (Web Text))

Max Activating Token Index: 336