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

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

Neuron 9 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1988. Min Act: -0.0003

Data Index: 1034939 (C4 (Web Text))

Max Activating Token Index: 201

Text #1

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1846. Min Act: -0.0003

Data Index: 652656 (C4 (Web Text))

Max Activating Token Index: 587

Text #2

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1736. Min Act: -0.0003

Data Index: 1277610 (C4 (Web Text))

Max Activating Token Index: 1001

Text #3

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1668. Min Act: -0.0003

Data Index: 1078569 (C4 (Web Text))

Max Activating Token Index: 180

Text #4

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1629. Min Act: -0.0003

Data Index: 236979 (C4 (Web Text))

Max Activating Token Index: 282

Text #5

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1592. Min Act: -0.0002

Data Index: 770706 (C4 (Web Text))

Max Activating Token Index: 522

Text #6

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1577. Min Act: -0.0002

Data Index: 99951 (C4 (Web Text))

Max Activating Token Index: 24

Text #7

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1620. Min Act: -0.0003

Data Index: 1113815 (C4 (Web Text))

Max Activating Token Index: 78

Text #8

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1567. Min Act: -0.0004

Data Index: 557853 (C4 (Web Text))

Max Activating Token Index: 604

Text #9

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1534. Min Act: -0.0002

Data Index: 280957 (C4 (Web Text))

Max Activating Token Index: 29

Text #10

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1619. Min Act: -0.0003

Data Index: 907091 (C4 (Web Text))

Max Activating Token Index: 927

Text #11

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1573. Min Act: -0.0003

Data Index: 611992 (C4 (Web Text))

Max Activating Token Index: 152

Text #12

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1509. Min Act: -0.0003

Data Index: 1069853 (C4 (Web Text))

Max Activating Token Index: 580

Text #13

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1521. Min Act: -0.0003

Data Index: 1134016 (C4 (Web Text))

Max Activating Token Index: 669

Text #14

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1520. Min Act: -0.0003

Data Index: 684686 (C4 (Web Text))

Max Activating Token Index: 832

Text #15

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1508. Min Act: -0.0003

Data Index: 107518 (C4 (Web Text))

Max Activating Token Index: 995

Text #16

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1444. Min Act: -0.0003

Data Index: 173364 (C4 (Web Text))

Max Activating Token Index: 851

Text #17

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1357. Min Act: -0.0003

Data Index: 1284148 (C4 (Web Text))

Max Activating Token Index: 324

Text #18

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1401. Min Act: -0.0003

Data Index: 677951 (C4 (Web Text))

Max Activating Token Index: 617

Text #19

Max Range: 0.1988. Min Range: -0.1988

Max Act: 0.1398. Min Act: -0.0004

Data Index: 350954 (C4 (Web Text))

Max Activating Token Index: 688