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

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

Neuron 951 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.3404. Min Range: -8.3404

Max Act: 8.3404. Min Act: -0.1699

Data Index: 1372470 (Python Code)

Max Activating Token Index: 244

Text #1

Max Range: 8.3404. Min Range: -8.3404

Max Act: 7.7154. Min Act: -0.1700

Data Index: 229849 (C4 (Web Text))

Max Activating Token Index: 106

Text #2

Max Range: 8.3404. Min Range: -8.3404

Max Act: 7.6529. Min Act: -0.1700

Data Index: 1323574 (C4 (Web Text))

Max Activating Token Index: 903

Text #3

Max Range: 8.3404. Min Range: -8.3404

Max Act: 7.6842. Min Act: -0.1697

Data Index: 1317486 (C4 (Web Text))

Max Activating Token Index: 863

Text #4

Max Range: 8.3404. Min Range: -8.3404

Max Act: 7.4967. Min Act: -0.1696

Data Index: 157390 (C4 (Web Text))

Max Activating Token Index: 785

Text #5

Max Range: 8.3404. Min Range: -8.3404

Max Act: 7.3404. Min Act: -0.1700

Data Index: 1075609 (C4 (Web Text))

Max Activating Token Index: 506

Text #6

Max Range: 8.3404. Min Range: -8.3404

Max Act: 7.3092. Min Act: -0.1699

Data Index: 165014 (C4 (Web Text))

Max Activating Token Index: 701

Text #7

Max Range: 8.3404. Min Range: -8.3404

Max Act: 7.0279. Min Act: -0.1700

Data Index: 1049527 (C4 (Web Text))

Max Activating Token Index: 782

Text #8

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.9342. Min Act: -0.1700

Data Index: 1423316 (Python Code)

Max Activating Token Index: 32

Text #9

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.9342. Min Act: -0.1686

Data Index: 694225 (C4 (Web Text))

Max Activating Token Index: 828

Text #10

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.7467. Min Act: -0.1699

Data Index: 35208 (C4 (Web Text))

Max Activating Token Index: 654

Text #11

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.7779. Min Act: -0.1698

Data Index: 241657 (C4 (Web Text))

Max Activating Token Index: 1008

Text #12

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.4967. Min Act: -0.1698

Data Index: 1232359 (C4 (Web Text))

Max Activating Token Index: 389

Text #13

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.4967. Min Act: -0.1698

Data Index: 145543 (C4 (Web Text))

Max Activating Token Index: 607

Text #14

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.4967. Min Act: -0.1700

Data Index: 1147649 (C4 (Web Text))

Max Activating Token Index: 509

Text #15

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.4342. Min Act: -0.1700

Data Index: 746682 (C4 (Web Text))

Max Activating Token Index: 900

Text #16

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.4654. Min Act: -0.1669

Data Index: 1193272 (C4 (Web Text))

Max Activating Token Index: 167

Text #17

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.4342. Min Act: -0.1700

Data Index: 1601652 (Python Code)

Max Activating Token Index: 149

Text #18

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.4029. Min Act: -0.1700

Data Index: 1157139 (C4 (Web Text))

Max Activating Token Index: 824

Text #19

Max Range: 8.3404. Min Range: -8.3404

Max Act: 6.3717. Min Act: -0.1700

Data Index: 1466298 (Python Code)

Max Activating Token Index: 417