Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2023 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.0555. Min Range: -5.0555

Max Act: 5.0555. Min Act: -0.1700

Data Index: 1075980 (C4 (Web Text))

Max Activating Token Index: 850

Text #1

Max Range: 5.0555. Min Range: -5.0555

Max Act: 5.0242. Min Act: -0.1700

Data Index: 106982 (C4 (Web Text))

Max Activating Token Index: 701

Text #2

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.9617. Min Act: -0.1700

Data Index: 190495 (C4 (Web Text))

Max Activating Token Index: 929

Text #3

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8680. Min Act: -0.1700

Data Index: 747436 (C4 (Web Text))

Max Activating Token Index: 884

Text #4

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8992. Min Act: -0.1700

Data Index: 1150568 (C4 (Web Text))

Max Activating Token Index: 886

Text #5

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8680. Min Act: -0.1700

Data Index: 304593 (C4 (Web Text))

Max Activating Token Index: 870

Text #6

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8680. Min Act: -0.1700

Data Index: 167011 (C4 (Web Text))

Max Activating Token Index: 990

Text #7

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8680. Min Act: -0.1700

Data Index: 443686 (C4 (Web Text))

Max Activating Token Index: 63

Text #8

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8992. Min Act: -0.1700

Data Index: 461493 (C4 (Web Text))

Max Activating Token Index: 978

Text #9

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8680. Min Act: -0.1700

Data Index: 1037193 (C4 (Web Text))

Max Activating Token Index: 1005

Text #10

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8367. Min Act: -0.1700

Data Index: 931847 (C4 (Web Text))

Max Activating Token Index: 796

Text #11

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.9305. Min Act: -0.1700

Data Index: 378694 (C4 (Web Text))

Max Activating Token Index: 213

Text #12

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8680. Min Act: -0.1700

Data Index: 946544 (C4 (Web Text))

Max Activating Token Index: 76

Text #13

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8367. Min Act: -0.1700

Data Index: 366606 (C4 (Web Text))

Max Activating Token Index: 118

Text #14

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.7742. Min Act: -0.1700

Data Index: 634786 (C4 (Web Text))

Max Activating Token Index: 894

Text #15

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8055. Min Act: -0.1700

Data Index: 1293527 (C4 (Web Text))

Max Activating Token Index: 1017

Text #16

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8055. Min Act: -0.1700

Data Index: 757836 (C4 (Web Text))

Max Activating Token Index: 360

Text #17

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.7742. Min Act: -0.1700

Data Index: 1019277 (C4 (Web Text))

Max Activating Token Index: 117

Text #18

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.7430. Min Act: -0.1700

Data Index: 280580 (C4 (Web Text))

Max Activating Token Index: 951

Text #19

Max Range: 5.0555. Min Range: -5.0555

Max Act: 4.8055. Min Act: -0.1700

Data Index: 320504 (C4 (Web Text))

Max Activating Token Index: 779