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

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

Neuron 335 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.3718. Min Act: -0.1700

Data Index: 1096227 (C4 (Web Text))

Max Activating Token Index: 893

Text #1

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.3405. Min Act: -0.1700

Data Index: 652013 (C4 (Web Text))

Max Activating Token Index: 353

Text #2

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.3875. Min Act: -0.1700

Data Index: 1147828 (C4 (Web Text))

Max Activating Token Index: 574

Text #3

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.3718. Min Act: -0.1700

Data Index: 501763 (C4 (Web Text))

Max Activating Token Index: 794

Text #4

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.3405. Min Act: -0.1700

Data Index: 177366 (C4 (Web Text))

Max Activating Token Index: 790

Text #5

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2933. Min Act: -0.1700

Data Index: 693110 (C4 (Web Text))

Max Activating Token Index: 32

Text #6

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2776. Min Act: -0.1700

Data Index: 105210 (C4 (Web Text))

Max Activating Token Index: 356

Text #7

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2148. Min Act: -0.1700

Data Index: 773535 (C4 (Web Text))

Max Activating Token Index: 707

Text #8

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2148. Min Act: -0.1700

Data Index: 164066 (C4 (Web Text))

Max Activating Token Index: 475

Text #9

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2305. Min Act: -0.1700

Data Index: 682796 (C4 (Web Text))

Max Activating Token Index: 605

Text #10

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2933. Min Act: -0.1700

Data Index: 175516 (C4 (Web Text))

Max Activating Token Index: 695

Text #11

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.1990. Min Act: -0.1700

Data Index: 90837 (C4 (Web Text))

Max Activating Token Index: 849

Text #12

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2776. Min Act: -0.1700

Data Index: 819845 (C4 (Web Text))

Max Activating Token Index: 138

Text #13

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2619. Min Act: -0.1700

Data Index: 1654209 (Python Code)

Max Activating Token Index: 183

Text #14

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2305. Min Act: -0.1700

Data Index: 817003 (C4 (Web Text))

Max Activating Token Index: 54

Text #15

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.1833. Min Act: -0.1700

Data Index: 478964 (C4 (Web Text))

Max Activating Token Index: 518

Text #16

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.1990. Min Act: -0.1700

Data Index: 114583 (C4 (Web Text))

Max Activating Token Index: 292

Text #17

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2148. Min Act: -0.1700

Data Index: 1092413 (C4 (Web Text))

Max Activating Token Index: 90

Text #18

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.1833. Min Act: -0.1700

Data Index: 781763 (C4 (Web Text))

Max Activating Token Index: 682

Text #19

Max Range: 3.3875. Min Range: -3.3875

Max Act: 3.2776. Min Act: -0.1700

Data Index: 674002 (C4 (Web Text))

Max Activating Token Index: 658