Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1728 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.3902. Min Act: -0.1700

Data Index: 936805 (C4 (Web Text))

Max Activating Token Index: 709

Text #1

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.2960. Min Act: -0.1700

Data Index: 247007 (C4 (Web Text))

Max Activating Token Index: 683

Text #2

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.2332. Min Act: -0.1700

Data Index: 1438109 (Python Code)

Max Activating Token Index: 172

Text #3

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.2332. Min Act: -0.1700

Data Index: 1509392 (Python Code)

Max Activating Token Index: 684

Text #4

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.2017. Min Act: -0.1700

Data Index: 21310 (C4 (Web Text))

Max Activating Token Index: 972

Text #5

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.1702. Min Act: -0.1700

Data Index: 1409997 (Python Code)

Max Activating Token Index: 279

Text #6

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.1387. Min Act: -0.1700

Data Index: 992079 (C4 (Web Text))

Max Activating Token Index: 998

Text #7

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.1387. Min Act: -0.1700

Data Index: 1360688 (Python Code)

Max Activating Token Index: 310

Text #8

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.1230. Min Act: -0.1700

Data Index: 1479630 (Python Code)

Max Activating Token Index: 897

Text #9

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.1387. Min Act: -0.1700

Data Index: 1000531 (C4 (Web Text))

Max Activating Token Index: 606

Text #10

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.0914. Min Act: -0.1700

Data Index: 1542046 (Python Code)

Max Activating Token Index: 535

Text #11

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.1072. Min Act: -0.1700

Data Index: 244947 (C4 (Web Text))

Max Activating Token Index: 476

Text #12

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.1387. Min Act: -0.1700

Data Index: 167988 (C4 (Web Text))

Max Activating Token Index: 394

Text #13

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.0757. Min Act: -0.1700

Data Index: 725257 (C4 (Web Text))

Max Activating Token Index: 563

Text #14

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.1230. Min Act: -0.1700

Data Index: 1536422 (Python Code)

Max Activating Token Index: 805

Text #15

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.0757. Min Act: -0.1700

Data Index: 115055 (C4 (Web Text))

Max Activating Token Index: 669

Text #16

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.0914. Min Act: -0.1700

Data Index: 1161960 (C4 (Web Text))

Max Activating Token Index: 602

Text #17

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.0757. Min Act: -0.1700

Data Index: 602625 (C4 (Web Text))

Max Activating Token Index: 847

Text #18

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.0757. Min Act: -0.1700

Data Index: 675539 (C4 (Web Text))

Max Activating Token Index: 583

Text #19

Max Range: 3.3902. Min Range: -3.3902

Max Act: 3.0914. Min Act: -0.1700

Data Index: 583883 (C4 (Web Text))

Max Activating Token Index: 675