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

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

Neuron 389 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.2527. Min Act: -0.1700

Data Index: 879856 (C4 (Web Text))

Max Activating Token Index: 1009

Text #1

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.2527. Min Act: -0.1700

Data Index: 450063 (C4 (Web Text))

Max Activating Token Index: 468

Text #2

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.2684. Min Act: -0.1700

Data Index: 942878 (C4 (Web Text))

Max Activating Token Index: 726

Text #3

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.2527. Min Act: -0.1700

Data Index: 614918 (C4 (Web Text))

Max Activating Token Index: 630

Text #4

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.2213. Min Act: -0.1700

Data Index: 1074573 (C4 (Web Text))

Max Activating Token Index: 689

Text #5

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.2527. Min Act: -0.1700

Data Index: 730274 (C4 (Web Text))

Max Activating Token Index: 725

Text #6

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.2213. Min Act: -0.1700

Data Index: 285291 (C4 (Web Text))

Max Activating Token Index: 954

Text #7

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1898. Min Act: -0.1700

Data Index: 870743 (C4 (Web Text))

Max Activating Token Index: 737

Text #8

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1898. Min Act: -0.1700

Data Index: 88119 (C4 (Web Text))

Max Activating Token Index: 645

Text #9

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1741. Min Act: -0.1700

Data Index: 448152 (C4 (Web Text))

Max Activating Token Index: 521

Text #10

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1898. Min Act: -0.1700

Data Index: 265376 (C4 (Web Text))

Max Activating Token Index: 62

Text #11

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1741. Min Act: -0.1700

Data Index: 155915 (C4 (Web Text))

Max Activating Token Index: 750

Text #12

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1741. Min Act: -0.1700

Data Index: 1100324 (C4 (Web Text))

Max Activating Token Index: 500

Text #13

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1741. Min Act: -0.1700

Data Index: 54401 (C4 (Web Text))

Max Activating Token Index: 638

Text #14

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1583. Min Act: -0.1700

Data Index: 263731 (C4 (Web Text))

Max Activating Token Index: 710

Text #15

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1583. Min Act: -0.1700

Data Index: 120949 (C4 (Web Text))

Max Activating Token Index: 825

Text #16

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1741. Min Act: -0.1700

Data Index: 174356 (C4 (Web Text))

Max Activating Token Index: 790

Text #17

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1741. Min Act: -0.1700

Data Index: 1369586 (Python Code)

Max Activating Token Index: 86

Text #18

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1426. Min Act: -0.1700

Data Index: 373649 (C4 (Web Text))

Max Activating Token Index: 476

Text #19

Max Range: 3.2684. Min Range: -3.2684

Max Act: 3.1426. Min Act: -0.1700

Data Index: 805574 (C4 (Web Text))

Max Activating Token Index: 500