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

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

Neuron 3227 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: 0.1957. Min Range: -0.1957

Max Act: 0.1867. Min Act: -0.0001

Data Index: 1161682 (C4 (Web Text))

Max Activating Token Index: 853

Text #1

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.1957. Min Act: -0.0000

Data Index: 783567 (C4 (Web Text))

Max Activating Token Index: 279

Text #2

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.1845. Min Act: -0.0000

Data Index: 1033866 (C4 (Web Text))

Max Activating Token Index: 821

Text #3

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.1867. Min Act: -0.0001

Data Index: 1198539 (C4 (Web Text))

Max Activating Token Index: 904

Text #4

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.1908. Min Act: -0.0000

Data Index: 4160 (C4 (Web Text))

Max Activating Token Index: 199

Text #5

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.1649. Min Act: -0.0000

Data Index: 452007 (C4 (Web Text))

Max Activating Token Index: 46

Text #6

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0862. Min Act: -0.0000

Data Index: 1349971 (C4 (Web Text))

Max Activating Token Index: 610

Text #7

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0820. Min Act: -0.0000

Data Index: 1275102 (C4 (Web Text))

Max Activating Token Index: 415

Text #8

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0759. Min Act: -0.0000

Data Index: 1400572 (Python Code)

Max Activating Token Index: 850

Text #9

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0771. Min Act: -0.0000

Data Index: 1388528 (Python Code)

Max Activating Token Index: 652

Text #10

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0707. Min Act: -0.0000

Data Index: 1193655 (C4 (Web Text))

Max Activating Token Index: 384

Text #11

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0698. Min Act: -0.0000

Data Index: 1543815 (Python Code)

Max Activating Token Index: 93

Text #12

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0635. Min Act: -0.0000

Data Index: 1139128 (C4 (Web Text))

Max Activating Token Index: 867

Text #13

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0623. Min Act: -0.0001

Data Index: 41242 (C4 (Web Text))

Max Activating Token Index: 440

Text #14

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0610. Min Act: -0.0000

Data Index: 1323603 (C4 (Web Text))

Max Activating Token Index: 317

Text #15

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0680. Min Act: -0.0000

Data Index: 300399 (C4 (Web Text))

Max Activating Token Index: 384

Text #16

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0833. Min Act: -0.0000

Data Index: 118577 (C4 (Web Text))

Max Activating Token Index: 60

Text #17

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0464. Min Act: -0.0000

Data Index: 1462644 (Python Code)

Max Activating Token Index: 330

Text #18

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0404. Min Act: -0.0000

Data Index: 296255 (C4 (Web Text))

Max Activating Token Index: 795

Text #19

Max Range: 0.1957. Min Range: -0.1957

Max Act: 0.0593. Min Act: -0.0000

Data Index: 1476530 (Python Code)

Max Activating Token Index: 396