Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1993 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 7.2823. Min Range: -7.2823

Max Act: 7.2823. Min Act: -0.1694

Data Index: 510163 (C4 (Web Text))

Max Activating Token Index: 638

Text #1

Max Range: 7.2823. Min Range: -7.2823

Max Act: 7.2198. Min Act: -0.1700

Data Index: 941026 (C4 (Web Text))

Max Activating Token Index: 422

Text #2

Max Range: 7.2823. Min Range: -7.2823

Max Act: 7.1886. Min Act: -0.1699

Data Index: 742866 (C4 (Web Text))

Max Activating Token Index: 621

Text #3

Max Range: 7.2823. Min Range: -7.2823

Max Act: 7.1886. Min Act: -0.1700

Data Index: 931431 (C4 (Web Text))

Max Activating Token Index: 95

Text #4

Max Range: 7.2823. Min Range: -7.2823

Max Act: 7.0011. Min Act: -0.1700

Data Index: 496437 (C4 (Web Text))

Max Activating Token Index: 925

Text #5

Max Range: 7.2823. Min Range: -7.2823

Max Act: 7.0636. Min Act: -0.1699

Data Index: 1087594 (C4 (Web Text))

Max Activating Token Index: 68

Text #6

Max Range: 7.2823. Min Range: -7.2823

Max Act: 7.0323. Min Act: -0.1700

Data Index: 179127 (C4 (Web Text))

Max Activating Token Index: 641

Text #7

Max Range: 7.2823. Min Range: -7.2823

Max Act: 7.0323. Min Act: -0.1700

Data Index: 237585 (C4 (Web Text))

Max Activating Token Index: 76

Text #8

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.9698. Min Act: -0.1697

Data Index: 717800 (C4 (Web Text))

Max Activating Token Index: 569

Text #9

Max Range: 7.2823. Min Range: -7.2823

Max Act: 7.0011. Min Act: -0.1696

Data Index: 1036301 (C4 (Web Text))

Max Activating Token Index: 456

Text #10

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.9386. Min Act: -0.1700

Data Index: 559662 (C4 (Web Text))

Max Activating Token Index: 37

Text #11

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.9386. Min Act: -0.1700

Data Index: 63000 (C4 (Web Text))

Max Activating Token Index: 135

Text #12

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.9698. Min Act: -0.1699

Data Index: 68662 (C4 (Web Text))

Max Activating Token Index: 403

Text #13

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.9698. Min Act: -0.1700

Data Index: 858763 (C4 (Web Text))

Max Activating Token Index: 356

Text #14

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.8761. Min Act: -0.1699

Data Index: 355239 (C4 (Web Text))

Max Activating Token Index: 139

Text #15

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.9073. Min Act: -0.1700

Data Index: 586500 (C4 (Web Text))

Max Activating Token Index: 466

Text #16

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.9386. Min Act: -0.1699

Data Index: 617623 (C4 (Web Text))

Max Activating Token Index: 774

Text #17

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.9073. Min Act: -0.1699

Data Index: 1272058 (C4 (Web Text))

Max Activating Token Index: 111

Text #18

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.8761. Min Act: -0.1700

Data Index: 211780 (C4 (Web Text))

Max Activating Token Index: 785

Text #19

Max Range: 7.2823. Min Range: -7.2823

Max Act: 6.8761. Min Act: -0.1698

Data Index: 159413 (C4 (Web Text))

Max Activating Token Index: 447