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

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

Neuron 1806 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: 5.0854. Min Range: -5.0854

Max Act: 5.0854. Min Act: -0.1700

Data Index: 8014 (C4 (Web Text))

Max Activating Token Index: 43

Text #1

Max Range: 5.0854. Min Range: -5.0854

Max Act: 5.0385. Min Act: -0.1700

Data Index: 1203875 (C4 (Web Text))

Max Activating Token Index: 22

Text #2

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.9760. Min Act: -0.1700

Data Index: 337313 (C4 (Web Text))

Max Activating Token Index: 899

Text #3

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.9760. Min Act: -0.1700

Data Index: 972023 (C4 (Web Text))

Max Activating Token Index: 105

Text #4

Max Range: 5.0854. Min Range: -5.0854

Max Act: 5.0073. Min Act: -0.1700

Data Index: 47024 (C4 (Web Text))

Max Activating Token Index: 754

Text #5

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.9917. Min Act: -0.1700

Data Index: 716839 (C4 (Web Text))

Max Activating Token Index: 884

Text #6

Max Range: 5.0854. Min Range: -5.0854

Max Act: 5.0073. Min Act: -0.1700

Data Index: 1278605 (C4 (Web Text))

Max Activating Token Index: 935

Text #7

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.8354. Min Act: -0.1700

Data Index: 204097 (C4 (Web Text))

Max Activating Token Index: 681

Text #8

Max Range: 5.0854. Min Range: -5.0854

Max Act: 5.0229. Min Act: -0.1700

Data Index: 377891 (C4 (Web Text))

Max Activating Token Index: 976

Text #9

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.9760. Min Act: -0.1700

Data Index: 297147 (C4 (Web Text))

Max Activating Token Index: 855

Text #10

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.7885. Min Act: -0.1700

Data Index: 738711 (C4 (Web Text))

Max Activating Token Index: 690

Text #11

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.7573. Min Act: -0.1700

Data Index: 347335 (C4 (Web Text))

Max Activating Token Index: 707

Text #12

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.6792. Min Act: -0.1700

Data Index: 418737 (C4 (Web Text))

Max Activating Token Index: 884

Text #13

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.8510. Min Act: -0.1700

Data Index: 1296176 (C4 (Web Text))

Max Activating Token Index: 805

Text #14

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.8667. Min Act: -0.1700

Data Index: 426030 (C4 (Web Text))

Max Activating Token Index: 592

Text #15

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.9135. Min Act: -0.1700

Data Index: 946185 (C4 (Web Text))

Max Activating Token Index: 660

Text #16

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.8042. Min Act: -0.1700

Data Index: 516795 (C4 (Web Text))

Max Activating Token Index: 971

Text #17

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.7417. Min Act: -0.1700

Data Index: 1041463 (C4 (Web Text))

Max Activating Token Index: 658

Text #18

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.8354. Min Act: -0.1700

Data Index: 1178921 (C4 (Web Text))

Max Activating Token Index: 331

Text #19

Max Range: 5.0854. Min Range: -5.0854

Max Act: 4.8667. Min Act: -0.1700

Data Index: 594493 (C4 (Web Text))

Max Activating Token Index: 696