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

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

Neuron 1888 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.4036. Min Range: -5.4036

Max Act: 5.3411. Min Act: -0.1700

Data Index: 1214907 (C4 (Web Text))

Max Activating Token Index: 901

Text #1

Max Range: 5.4036. Min Range: -5.4036

Max Act: 5.4036. Min Act: -0.1700

Data Index: 702977 (C4 (Web Text))

Max Activating Token Index: 354

Text #2

Max Range: 5.4036. Min Range: -5.4036

Max Act: 5.1849. Min Act: -0.1700

Data Index: 679834 (C4 (Web Text))

Max Activating Token Index: 389

Text #3

Max Range: 5.4036. Min Range: -5.4036

Max Act: 5.0911. Min Act: -0.1700

Data Index: 220066 (C4 (Web Text))

Max Activating Token Index: 945

Text #4

Max Range: 5.4036. Min Range: -5.4036

Max Act: 5.1224. Min Act: -0.1700

Data Index: 755637 (C4 (Web Text))

Max Activating Token Index: 924

Text #5

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.9974. Min Act: -0.1700

Data Index: 89005 (C4 (Web Text))

Max Activating Token Index: 297

Text #6

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.9974. Min Act: -0.1700

Data Index: 1096767 (C4 (Web Text))

Max Activating Token Index: 351

Text #7

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.9661. Min Act: -0.1700

Data Index: 531703 (C4 (Web Text))

Max Activating Token Index: 929

Text #8

Max Range: 5.4036. Min Range: -5.4036

Max Act: 5.0286. Min Act: -0.1700

Data Index: 341486 (C4 (Web Text))

Max Activating Token Index: 380

Text #9

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.9661. Min Act: -0.1700

Data Index: 1055007 (C4 (Web Text))

Max Activating Token Index: 459

Text #10

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.9036. Min Act: -0.1700

Data Index: 566829 (C4 (Web Text))

Max Activating Token Index: 724

Text #11

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.9661. Min Act: -0.1700

Data Index: 1197710 (C4 (Web Text))

Max Activating Token Index: 542

Text #12

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.9036. Min Act: -0.1700

Data Index: 1140404 (C4 (Web Text))

Max Activating Token Index: 584

Text #13

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.9036. Min Act: -0.1700

Data Index: 204751 (C4 (Web Text))

Max Activating Token Index: 1006

Text #14

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.8724. Min Act: -0.1700

Data Index: 338704 (C4 (Web Text))

Max Activating Token Index: 839

Text #15

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.9974. Min Act: -0.1700

Data Index: 77373 (C4 (Web Text))

Max Activating Token Index: 729

Text #16

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.7474. Min Act: -0.1700

Data Index: 1058457 (C4 (Web Text))

Max Activating Token Index: 700

Text #17

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.9036. Min Act: -0.1700

Data Index: 1123171 (C4 (Web Text))

Max Activating Token Index: 492

Text #18

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.8099. Min Act: -0.1700

Data Index: 637228 (C4 (Web Text))

Max Activating Token Index: 953

Text #19

Max Range: 5.4036. Min Range: -5.4036

Max Act: 4.8099. Min Act: -0.1700

Data Index: 919331 (C4 (Web Text))

Max Activating Token Index: 524