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

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

Neuron 4450 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.3575. Min Act: -0.0000

Data Index: 246449 (C4 (Web Text))

Max Activating Token Index: 700

Text #1

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.3995. Min Act: -0.0000

Data Index: 1207894 (C4 (Web Text))

Max Activating Token Index: 608

Text #2

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.2964. Min Act: -0.0000

Data Index: 569748 (C4 (Web Text))

Max Activating Token Index: 505

Text #3

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.1342. Min Act: -0.0000

Data Index: 222782 (C4 (Web Text))

Max Activating Token Index: 74

Text #4

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.0960. Min Act: -0.0000

Data Index: 1263192 (C4 (Web Text))

Max Activating Token Index: 481

Text #5

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.0887. Min Act: -0.0000

Data Index: 1315950 (C4 (Web Text))

Max Activating Token Index: 484

Text #6

Max Range: 1.3995. Min Range: -1.3995

Max Act: 0.9893. Min Act: -0.0000

Data Index: 1338945 (C4 (Web Text))

Max Activating Token Index: 748

Text #7

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.0261. Min Act: -0.0000

Data Index: 15873 (C4 (Web Text))

Max Activating Token Index: 369

Text #8

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.0171. Min Act: -0.0000

Data Index: 890484 (C4 (Web Text))

Max Activating Token Index: 810

Text #9

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.0142. Min Act: -0.0000

Data Index: 1355532 (C4 (Web Text))

Max Activating Token Index: 129

Text #10

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.0167. Min Act: -0.0000

Data Index: 277094 (C4 (Web Text))

Max Activating Token Index: 988

Text #11

Max Range: 1.3995. Min Range: -1.3995

Max Act: 1.0190. Min Act: -0.0000

Data Index: 1087081 (C4 (Web Text))

Max Activating Token Index: 705

Text #12

Max Range: 1.3995. Min Range: -1.3995

Max Act: 0.9752. Min Act: -0.0000

Data Index: 87760 (C4 (Web Text))

Max Activating Token Index: 799

Text #13

Max Range: 1.3995. Min Range: -1.3995

Max Act: 0.9576. Min Act: -0.0000

Data Index: 1328438 (C4 (Web Text))

Max Activating Token Index: 794

Text #14

Max Range: 1.3995. Min Range: -1.3995

Max Act: 0.9434. Min Act: -0.0000

Data Index: 806106 (C4 (Web Text))

Max Activating Token Index: 673

Text #15

Max Range: 1.3995. Min Range: -1.3995

Max Act: 0.9498. Min Act: -0.0000

Data Index: 298444 (C4 (Web Text))

Max Activating Token Index: 436

Text #16

Max Range: 1.3995. Min Range: -1.3995

Max Act: 0.9766. Min Act: -0.0000

Data Index: 106758 (C4 (Web Text))

Max Activating Token Index: 878

Text #17

Max Range: 1.3995. Min Range: -1.3995

Max Act: 0.9471. Min Act: -0.0000

Data Index: 1086828 (C4 (Web Text))

Max Activating Token Index: 470

Text #18

Max Range: 1.3995. Min Range: -1.3995

Max Act: 0.9441. Min Act: -0.0000

Data Index: 366122 (C4 (Web Text))

Max Activating Token Index: 567

Text #19

Max Range: 1.3995. Min Range: -1.3995

Max Act: 0.8802. Min Act: -0.0000

Data Index: 1032984 (C4 (Web Text))

Max Activating Token Index: 245