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

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

Neuron 750 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.6378. Min Act: -0.1700

Data Index: 717843 (C4 (Web Text))

Max Activating Token Index: 815

Text #1

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.5752. Min Act: -0.1700

Data Index: 869898 (C4 (Web Text))

Max Activating Token Index: 493

Text #2

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.5752. Min Act: -0.1700

Data Index: 987565 (C4 (Web Text))

Max Activating Token Index: 413

Text #3

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.5439. Min Act: -0.1700

Data Index: 622633 (C4 (Web Text))

Max Activating Token Index: 41

Text #4

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4812. Min Act: -0.1700

Data Index: 1056395 (C4 (Web Text))

Max Activating Token Index: 946

Text #5

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4812. Min Act: -0.1700

Data Index: 769308 (C4 (Web Text))

Max Activating Token Index: 446

Text #6

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4499. Min Act: -0.1700

Data Index: 431108 (C4 (Web Text))

Max Activating Token Index: 638

Text #7

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4499. Min Act: -0.1700

Data Index: 1043420 (C4 (Web Text))

Max Activating Token Index: 1000

Text #8

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4185. Min Act: -0.1700

Data Index: 108051 (C4 (Web Text))

Max Activating Token Index: 90

Text #9

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4812. Min Act: -0.1700

Data Index: 1078981 (C4 (Web Text))

Max Activating Token Index: 362

Text #10

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4499. Min Act: -0.1700

Data Index: 102364 (C4 (Web Text))

Max Activating Token Index: 503

Text #11

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4185. Min Act: -0.1700

Data Index: 819879 (C4 (Web Text))

Max Activating Token Index: 778

Text #12

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.3871. Min Act: -0.1700

Data Index: 1063330 (C4 (Web Text))

Max Activating Token Index: 284

Text #13

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4499. Min Act: -0.1700

Data Index: 826740 (C4 (Web Text))

Max Activating Token Index: 858

Text #14

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4185. Min Act: -0.1700

Data Index: 357528 (C4 (Web Text))

Max Activating Token Index: 625

Text #15

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4185. Min Act: -0.1700

Data Index: 1103323 (C4 (Web Text))

Max Activating Token Index: 149

Text #16

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.4185. Min Act: -0.1700

Data Index: 550500 (C4 (Web Text))

Max Activating Token Index: 657

Text #17

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.3558. Min Act: -0.1700

Data Index: 1204791 (C4 (Web Text))

Max Activating Token Index: 730

Text #18

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.3871. Min Act: -0.1700

Data Index: 353777 (C4 (Web Text))

Max Activating Token Index: 261

Text #19

Max Range: 3.6378. Min Range: -3.6378

Max Act: 3.3558. Min Act: -0.1700

Data Index: 159469 (C4 (Web Text))

Max Activating Token Index: 552