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

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

Neuron 751 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: 5.7529. Min Range: -5.7529

Max Act: 5.7529. Min Act: -0.1700

Data Index: 714904 (C4 (Web Text))

Max Activating Token Index: 579

Text #1

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.6591. Min Act: -0.1700

Data Index: 1232924 (C4 (Web Text))

Max Activating Token Index: 579

Text #2

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.6591. Min Act: -0.1700

Data Index: 386181 (C4 (Web Text))

Max Activating Token Index: 547

Text #3

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.5966. Min Act: -0.1700

Data Index: 829011 (C4 (Web Text))

Max Activating Token Index: 786

Text #4

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.4716. Min Act: -0.1700

Data Index: 466882 (C4 (Web Text))

Max Activating Token Index: 461

Text #5

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.4404. Min Act: -0.1699

Data Index: 920179 (C4 (Web Text))

Max Activating Token Index: 547

Text #6

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.4716. Min Act: -0.1700

Data Index: 43490 (C4 (Web Text))

Max Activating Token Index: 292

Text #7

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.3154. Min Act: -0.1700

Data Index: 755453 (C4 (Web Text))

Max Activating Token Index: 671

Text #8

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.3779. Min Act: -0.1700

Data Index: 879251 (C4 (Web Text))

Max Activating Token Index: 676

Text #9

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.3154. Min Act: -0.1700

Data Index: 1034481 (C4 (Web Text))

Max Activating Token Index: 59

Text #10

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.2529. Min Act: -0.1700

Data Index: 24490 (C4 (Web Text))

Max Activating Token Index: 922

Text #11

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.2529. Min Act: -0.1700

Data Index: 797449 (C4 (Web Text))

Max Activating Token Index: 245

Text #12

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.2216. Min Act: -0.1700

Data Index: 364935 (C4 (Web Text))

Max Activating Token Index: 619

Text #13

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.1279. Min Act: -0.1700

Data Index: 1269805 (C4 (Web Text))

Max Activating Token Index: 653

Text #14

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.0654. Min Act: -0.1696

Data Index: 1197384 (C4 (Web Text))

Max Activating Token Index: 599

Text #15

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.0966. Min Act: -0.1700

Data Index: 1304189 (C4 (Web Text))

Max Activating Token Index: 140

Text #16

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.0654. Min Act: -0.1700

Data Index: 397449 (C4 (Web Text))

Max Activating Token Index: 106

Text #17

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.0654. Min Act: -0.1700

Data Index: 1131330 (C4 (Web Text))

Max Activating Token Index: 339

Text #18

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.0654. Min Act: -0.1700

Data Index: 1162121 (C4 (Web Text))

Max Activating Token Index: 701

Text #19

Max Range: 5.7529. Min Range: -5.7529

Max Act: 5.0029. Min Act: -0.1700

Data Index: 121444 (C4 (Web Text))

Max Activating Token Index: 459