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

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

Neuron 4316 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.8437. Min Act: -0.0000

Data Index: 853645 (C4 (Web Text))

Max Activating Token Index: 50

Text #1

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.8089. Min Act: -0.0000

Data Index: 259306 (C4 (Web Text))

Max Activating Token Index: 9

Text #2

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.8106. Min Act: -0.0001

Data Index: 757052 (C4 (Web Text))

Max Activating Token Index: 822

Text #3

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7790. Min Act: -0.0000

Data Index: 85577 (C4 (Web Text))

Max Activating Token Index: 119

Text #4

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7702. Min Act: -0.0000

Data Index: 325889 (C4 (Web Text))

Max Activating Token Index: 808

Text #5

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.8241. Min Act: -0.0000

Data Index: 48596 (C4 (Web Text))

Max Activating Token Index: 921

Text #6

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7687. Min Act: -0.0001

Data Index: 16041 (C4 (Web Text))

Max Activating Token Index: 863

Text #7

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7655. Min Act: -0.0000

Data Index: 653991 (C4 (Web Text))

Max Activating Token Index: 37

Text #8

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7428. Min Act: -0.0000

Data Index: 949912 (C4 (Web Text))

Max Activating Token Index: 917

Text #9

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7613. Min Act: -0.0000

Data Index: 213997 (C4 (Web Text))

Max Activating Token Index: 371

Text #10

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7602. Min Act: -0.0001

Data Index: 829110 (C4 (Web Text))

Max Activating Token Index: 384

Text #11

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7319. Min Act: -0.0000

Data Index: 223853 (C4 (Web Text))

Max Activating Token Index: 350

Text #12

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7010. Min Act: -0.0001

Data Index: 721953 (C4 (Web Text))

Max Activating Token Index: 838

Text #13

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7524. Min Act: -0.0001

Data Index: 785961 (C4 (Web Text))

Max Activating Token Index: 651

Text #14

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7605. Min Act: -0.0001

Data Index: 314388 (C4 (Web Text))

Max Activating Token Index: 133

Text #15

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7625. Min Act: -0.0000

Data Index: 641326 (C4 (Web Text))

Max Activating Token Index: 544

Text #16

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.6969. Min Act: -0.0000

Data Index: 1325652 (C4 (Web Text))

Max Activating Token Index: 123

Text #17

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7467. Min Act: -0.0000

Data Index: 873454 (C4 (Web Text))

Max Activating Token Index: 878

Text #18

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7528. Min Act: -0.0000

Data Index: 212503 (C4 (Web Text))

Max Activating Token Index: 156

Text #19

Max Range: 0.8437. Min Range: -0.8437

Max Act: 0.7252. Min Act: -0.0000

Data Index: 295817 (C4 (Web Text))

Max Activating Token Index: 972