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

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

Neuron 4218 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: 4.5278. Min Range: -4.5278

Max Act: 4.5278. Min Act: -0.0000

Data Index: 180446 (C4 (Web Text))

Max Activating Token Index: 219

Text #1

Max Range: 4.5278. Min Range: -4.5278

Max Act: 3.7594. Min Act: -0.0000

Data Index: 179364 (C4 (Web Text))

Max Activating Token Index: 810

Text #2

Max Range: 4.5278. Min Range: -4.5278

Max Act: 3.6042. Min Act: -0.0000

Data Index: 1329594 (C4 (Web Text))

Max Activating Token Index: 436

Text #3

Max Range: 4.5278. Min Range: -4.5278

Max Act: 3.1508. Min Act: -0.0000

Data Index: 78336 (C4 (Web Text))

Max Activating Token Index: 703

Text #4

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.9940. Min Act: -0.0000

Data Index: 1313765 (C4 (Web Text))

Max Activating Token Index: 349

Text #5

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.8288. Min Act: -0.0001

Data Index: 737149 (C4 (Web Text))

Max Activating Token Index: 890

Text #6

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.8207. Min Act: -0.0001

Data Index: 1209121 (C4 (Web Text))

Max Activating Token Index: 382

Text #7

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.4914. Min Act: -0.0000

Data Index: 1321551 (C4 (Web Text))

Max Activating Token Index: 940

Text #8

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.7321. Min Act: -0.0000

Data Index: 918710 (C4 (Web Text))

Max Activating Token Index: 158

Text #9

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.5787. Min Act: -0.0000

Data Index: 568041 (C4 (Web Text))

Max Activating Token Index: 761

Text #10

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.8380. Min Act: -0.0000

Data Index: 1279618 (C4 (Web Text))

Max Activating Token Index: 637

Text #11

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.5454. Min Act: -0.0000

Data Index: 310708 (C4 (Web Text))

Max Activating Token Index: 70

Text #12

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.6535. Min Act: -0.0000

Data Index: 77043 (C4 (Web Text))

Max Activating Token Index: 25

Text #13

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.6148. Min Act: -0.0000

Data Index: 1043540 (C4 (Web Text))

Max Activating Token Index: 603

Text #14

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.6176. Min Act: -0.0000

Data Index: 498919 (C4 (Web Text))

Max Activating Token Index: 50

Text #15

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.4440. Min Act: -0.0000

Data Index: 860218 (C4 (Web Text))

Max Activating Token Index: 636

Text #16

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.4481. Min Act: -0.0000

Data Index: 1092321 (C4 (Web Text))

Max Activating Token Index: 16

Text #17

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.5752. Min Act: -0.0000

Data Index: 335258 (C4 (Web Text))

Max Activating Token Index: 753

Text #18

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.5795. Min Act: -0.0000

Data Index: 1087883 (C4 (Web Text))

Max Activating Token Index: 524

Text #19

Max Range: 4.5278. Min Range: -4.5278

Max Act: 2.5757. Min Act: -0.0000

Data Index: 650614 (C4 (Web Text))

Max Activating Token Index: 409