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

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

Neuron 1365 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: 3.8595. Min Range: -3.8595

Max Act: 3.8595. Min Act: -0.1700

Data Index: 1213243 (C4 (Web Text))

Max Activating Token Index: 516

Text #1

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.8595. Min Act: -0.1700

Data Index: 748974 (C4 (Web Text))

Max Activating Token Index: 834

Text #2

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7343. Min Act: -0.1700

Data Index: 96163 (C4 (Web Text))

Max Activating Token Index: 545

Text #3

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7656. Min Act: -0.1700

Data Index: 1046807 (C4 (Web Text))

Max Activating Token Index: 170

Text #4

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.8595. Min Act: -0.1700

Data Index: 870526 (C4 (Web Text))

Max Activating Token Index: 464

Text #5

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.8282. Min Act: -0.1700

Data Index: 1615582 (Python Code)

Max Activating Token Index: 251

Text #6

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7969. Min Act: -0.1700

Data Index: 5445 (C4 (Web Text))

Max Activating Token Index: 867

Text #7

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7969. Min Act: -0.1700

Data Index: 774642 (C4 (Web Text))

Max Activating Token Index: 344

Text #8

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7656. Min Act: -0.1700

Data Index: 1138888 (C4 (Web Text))

Max Activating Token Index: 710

Text #9

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7969. Min Act: -0.1700

Data Index: 259482 (C4 (Web Text))

Max Activating Token Index: 725

Text #10

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7656. Min Act: -0.1700

Data Index: 988531 (C4 (Web Text))

Max Activating Token Index: 542

Text #11

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7343. Min Act: -0.1700

Data Index: 1412513 (Python Code)

Max Activating Token Index: 34

Text #12

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7343. Min Act: -0.1700

Data Index: 16123 (C4 (Web Text))

Max Activating Token Index: 437

Text #13

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7343. Min Act: -0.1700

Data Index: 324005 (C4 (Web Text))

Max Activating Token Index: 146

Text #14

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7030. Min Act: -0.1700

Data Index: 53389 (C4 (Web Text))

Max Activating Token Index: 567

Text #15

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.6717. Min Act: -0.1700

Data Index: 305462 (C4 (Web Text))

Max Activating Token Index: 118

Text #16

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7343. Min Act: -0.1700

Data Index: 1089091 (C4 (Web Text))

Max Activating Token Index: 445

Text #17

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7656. Min Act: -0.1700

Data Index: 201491 (C4 (Web Text))

Max Activating Token Index: 701

Text #18

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7343. Min Act: -0.1700

Data Index: 859876 (C4 (Web Text))

Max Activating Token Index: 639

Text #19

Max Range: 3.8595. Min Range: -3.8595

Max Act: 3.7343. Min Act: -0.1700

Data Index: 1207865 (C4 (Web Text))

Max Activating Token Index: 825