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

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

Neuron 4393 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.6418. Min Act: -0.0001

Data Index: 892658 (C4 (Web Text))

Max Activating Token Index: 282

Text #1

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.3888. Min Act: -0.0000

Data Index: 751699 (C4 (Web Text))

Max Activating Token Index: 535

Text #2

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.3361. Min Act: -0.0001

Data Index: 591687 (C4 (Web Text))

Max Activating Token Index: 322

Text #3

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2673. Min Act: -0.0001

Data Index: 129956 (C4 (Web Text))

Max Activating Token Index: 721

Text #4

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2518. Min Act: -0.0000

Data Index: 1106324 (C4 (Web Text))

Max Activating Token Index: 474

Text #5

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2672. Min Act: -0.0001

Data Index: 212704 (C4 (Web Text))

Max Activating Token Index: 71

Text #6

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2476. Min Act: -0.0001

Data Index: 900980 (C4 (Web Text))

Max Activating Token Index: 528

Text #7

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2329. Min Act: -0.0000

Data Index: 917585 (C4 (Web Text))

Max Activating Token Index: 116

Text #8

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2183. Min Act: -0.0001

Data Index: 1337683 (C4 (Web Text))

Max Activating Token Index: 117

Text #9

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2252. Min Act: -0.0001

Data Index: 314908 (C4 (Web Text))

Max Activating Token Index: 51

Text #10

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2204. Min Act: -0.0000

Data Index: 1047207 (C4 (Web Text))

Max Activating Token Index: 243

Text #11

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2195. Min Act: -0.0001

Data Index: 290848 (C4 (Web Text))

Max Activating Token Index: 974

Text #12

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2067. Min Act: -0.0001

Data Index: 966934 (C4 (Web Text))

Max Activating Token Index: 566

Text #13

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.1907. Min Act: -0.0001

Data Index: 1141028 (C4 (Web Text))

Max Activating Token Index: 863

Text #14

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.1942. Min Act: -0.0000

Data Index: 1102853 (C4 (Web Text))

Max Activating Token Index: 109

Text #15

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.1901. Min Act: -0.0001

Data Index: 581097 (C4 (Web Text))

Max Activating Token Index: 885

Text #16

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.1823. Min Act: -0.0000

Data Index: 689686 (C4 (Web Text))

Max Activating Token Index: 116

Text #17

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.2006. Min Act: -0.0000

Data Index: 335394 (C4 (Web Text))

Max Activating Token Index: 570

Text #18

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.1813. Min Act: -0.0001

Data Index: 1270392 (C4 (Web Text))

Max Activating Token Index: 404

Text #19

Max Range: 0.6418. Min Range: -0.6418

Max Act: 0.1756. Min Act: -0.0000

Data Index: 934781 (C4 (Web Text))

Max Activating Token Index: 749