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

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

Neuron 55 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: 5.2309. Min Range: -5.2309

Max Act: 5.2309. Min Act: -0.1700

Data Index: 467717 (C4 (Web Text))

Max Activating Token Index: 60

Text #1

Max Range: 5.2309. Min Range: -5.2309

Max Act: 5.1996. Min Act: -0.1700

Data Index: 227005 (C4 (Web Text))

Max Activating Token Index: 817

Text #2

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.9496. Min Act: -0.1697

Data Index: 802224 (C4 (Web Text))

Max Activating Token Index: 830

Text #3

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.8559. Min Act: -0.1700

Data Index: 223754 (C4 (Web Text))

Max Activating Token Index: 82

Text #4

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.8246. Min Act: -0.1700

Data Index: 1200332 (C4 (Web Text))

Max Activating Token Index: 893

Text #5

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.8871. Min Act: -0.1700

Data Index: 399267 (C4 (Web Text))

Max Activating Token Index: 937

Text #6

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.7621. Min Act: -0.1700

Data Index: 156263 (C4 (Web Text))

Max Activating Token Index: 183

Text #7

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.7934. Min Act: -0.1700

Data Index: 905001 (C4 (Web Text))

Max Activating Token Index: 628

Text #8

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.7621. Min Act: -0.1700

Data Index: 1195883 (C4 (Web Text))

Max Activating Token Index: 127

Text #9

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.7621. Min Act: -0.1700

Data Index: 655283 (C4 (Web Text))

Max Activating Token Index: 498

Text #10

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.7465. Min Act: -0.1700

Data Index: 381752 (C4 (Web Text))

Max Activating Token Index: 1006

Text #11

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.6840. Min Act: -0.1700

Data Index: 632815 (C4 (Web Text))

Max Activating Token Index: 813

Text #12

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.7153. Min Act: -0.1700

Data Index: 600718 (C4 (Web Text))

Max Activating Token Index: 1006

Text #13

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.6684. Min Act: -0.1700

Data Index: 894334 (C4 (Web Text))

Max Activating Token Index: 193

Text #14

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.6528. Min Act: -0.1700

Data Index: 1317253 (C4 (Web Text))

Max Activating Token Index: 71

Text #15

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.6840. Min Act: -0.1700

Data Index: 350589 (C4 (Web Text))

Max Activating Token Index: 952

Text #16

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.6371. Min Act: -0.1700

Data Index: 90894 (C4 (Web Text))

Max Activating Token Index: 972

Text #17

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.6215. Min Act: -0.1700

Data Index: 92629 (C4 (Web Text))

Max Activating Token Index: 319

Text #18

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.6371. Min Act: -0.1700

Data Index: 724989 (C4 (Web Text))

Max Activating Token Index: 183

Text #19

Max Range: 5.2309. Min Range: -5.2309

Max Act: 4.6684. Min Act: -0.1700

Data Index: 1306845 (C4 (Web Text))

Max Activating Token Index: 252