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

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

Neuron 860 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.9563. Min Act: -0.1700

Data Index: 212763 (C4 (Web Text))

Max Activating Token Index: 937

Text #1

Max Range: 7.0813. Min Range: -7.0813

Max Act: 7.0813. Min Act: -0.1700

Data Index: 1352247 (C4 (Web Text))

Max Activating Token Index: 197

Text #2

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.6126. Min Act: -0.1700

Data Index: 394518 (C4 (Web Text))

Max Activating Token Index: 339

Text #3

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.6751. Min Act: -0.1700

Data Index: 398154 (C4 (Web Text))

Max Activating Token Index: 739

Text #4

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.5188. Min Act: -0.1700

Data Index: 553368 (C4 (Web Text))

Max Activating Token Index: 801

Text #5

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.5188. Min Act: -0.1700

Data Index: 844999 (C4 (Web Text))

Max Activating Token Index: 112

Text #6

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.4251. Min Act: -0.1700

Data Index: 30723 (C4 (Web Text))

Max Activating Token Index: 233

Text #7

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.4251. Min Act: -0.1700

Data Index: 766666 (C4 (Web Text))

Max Activating Token Index: 206

Text #8

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.4563. Min Act: -0.1700

Data Index: 842343 (C4 (Web Text))

Max Activating Token Index: 939

Text #9

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.3938. Min Act: -0.1700

Data Index: 1135537 (C4 (Web Text))

Max Activating Token Index: 240

Text #10

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.3626. Min Act: -0.1700

Data Index: 874443 (C4 (Web Text))

Max Activating Token Index: 374

Text #11

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.3938. Min Act: -0.1700

Data Index: 310425 (C4 (Web Text))

Max Activating Token Index: 406

Text #12

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.3626. Min Act: -0.1700

Data Index: 193893 (C4 (Web Text))

Max Activating Token Index: 455

Text #13

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.3626. Min Act: -0.1700

Data Index: 664464 (C4 (Web Text))

Max Activating Token Index: 997

Text #14

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.3626. Min Act: -0.1700

Data Index: 677916 (C4 (Web Text))

Max Activating Token Index: 913

Text #15

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.3001. Min Act: -0.1700

Data Index: 1097171 (C4 (Web Text))

Max Activating Token Index: 747

Text #16

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.3001. Min Act: -0.1700

Data Index: 208634 (C4 (Web Text))

Max Activating Token Index: 609

Text #17

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.2376. Min Act: -0.1700

Data Index: 137386 (C4 (Web Text))

Max Activating Token Index: 344

Text #18

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.3313. Min Act: -0.1700

Data Index: 463400 (C4 (Web Text))

Max Activating Token Index: 182

Text #19

Max Range: 7.0813. Min Range: -7.0813

Max Act: 6.3938. Min Act: -0.1700

Data Index: 53631 (C4 (Web Text))

Max Activating Token Index: 937