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

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

Neuron 1921 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.2416. Min Act: -0.1700

Data Index: 1275692 (C4 (Web Text))

Max Activating Token Index: 14

Text #1

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.2416. Min Act: -0.1700

Data Index: 456808 (C4 (Web Text))

Max Activating Token Index: 924

Text #2

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.2259. Min Act: -0.1700

Data Index: 417905 (C4 (Web Text))

Max Activating Token Index: 275

Text #3

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.2259. Min Act: -0.1700

Data Index: 469991 (C4 (Web Text))

Max Activating Token Index: 690

Text #4

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1787. Min Act: -0.1700

Data Index: 33260 (C4 (Web Text))

Max Activating Token Index: 790

Text #5

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1787. Min Act: -0.1700

Data Index: 822994 (C4 (Web Text))

Max Activating Token Index: 829

Text #6

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1472. Min Act: -0.1700

Data Index: 472722 (C4 (Web Text))

Max Activating Token Index: 813

Text #7

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1787. Min Act: -0.1700

Data Index: 406130 (C4 (Web Text))

Max Activating Token Index: 45

Text #8

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1472. Min Act: -0.1700

Data Index: 523160 (C4 (Web Text))

Max Activating Token Index: 50

Text #9

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1787. Min Act: -0.1700

Data Index: 1306414 (C4 (Web Text))

Max Activating Token Index: 18

Text #10

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1315. Min Act: -0.1700

Data Index: 661595 (C4 (Web Text))

Max Activating Token Index: 546

Text #11

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1157. Min Act: -0.1700

Data Index: 857394 (C4 (Web Text))

Max Activating Token Index: 517

Text #12

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1315. Min Act: -0.1700

Data Index: 339642 (C4 (Web Text))

Max Activating Token Index: 169

Text #13

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1157. Min Act: -0.1700

Data Index: 138901 (C4 (Web Text))

Max Activating Token Index: 11

Text #14

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.1157. Min Act: -0.1700

Data Index: 398686 (C4 (Web Text))

Max Activating Token Index: 275

Text #15

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.0999. Min Act: -0.1700

Data Index: 1049356 (C4 (Web Text))

Max Activating Token Index: 194

Text #16

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.0842. Min Act: -0.1700

Data Index: 1213594 (C4 (Web Text))

Max Activating Token Index: 56

Text #17

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.0684. Min Act: -0.1700

Data Index: 404516 (C4 (Web Text))

Max Activating Token Index: 14

Text #18

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.0526. Min Act: -0.1700

Data Index: 260716 (C4 (Web Text))

Max Activating Token Index: 48

Text #19

Max Range: 3.2416. Min Range: -3.2416

Max Act: 3.0684. Min Act: -0.1700

Data Index: 78859 (C4 (Web Text))

Max Activating Token Index: 142