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

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

Neuron 1182 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: 8.5563. Min Range: -8.5563

Max Act: 8.5563. Min Act: -0.1699

Data Index: 1479732 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.4001. Min Act: -0.1700

Data Index: 1119288 (C4 (Web Text))

Max Activating Token Index: 4

Text #2

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.4001. Min Act: -0.1645

Data Index: 1058062 (C4 (Web Text))

Max Activating Token Index: 4

Text #3

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.3063. Min Act: -0.0998

Data Index: 146224 (C4 (Web Text))

Max Activating Token Index: 4

Text #4

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.3063. Min Act: 0.2090

Data Index: 1011494 (C4 (Web Text))

Max Activating Token Index: 4

Text #5

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.3063. Min Act: -0.1438

Data Index: 325958 (C4 (Web Text))

Max Activating Token Index: 9

Text #6

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.3063. Min Act: -0.1674

Data Index: 643859 (C4 (Web Text))

Max Activating Token Index: 5

Text #7

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.3063. Min Act: -0.0899

Data Index: 1312742 (C4 (Web Text))

Max Activating Token Index: 4

Text #8

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.3688. Min Act: -0.1699

Data Index: 1551905 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.2751. Min Act: -0.0937

Data Index: 477489 (C4 (Web Text))

Max Activating Token Index: 5

Text #10

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.2126. Min Act: -0.1683

Data Index: 902995 (C4 (Web Text))

Max Activating Token Index: 5

Text #11

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.1501. Min Act: -0.1697

Data Index: 398977 (C4 (Web Text))

Max Activating Token Index: 4

Text #12

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.1188. Min Act: -0.1699

Data Index: 325911 (C4 (Web Text))

Max Activating Token Index: 4

Text #13

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.1813. Min Act: 0.0242

Data Index: 14489 (C4 (Web Text))

Max Activating Token Index: 5

Text #14

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.0876. Min Act: -0.1666

Data Index: 1257319 (C4 (Web Text))

Max Activating Token Index: 8

Text #15

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.0563. Min Act: -0.1380

Data Index: 556664 (C4 (Web Text))

Max Activating Token Index: 7

Text #16

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.0563. Min Act: -0.1045

Data Index: 777256 (C4 (Web Text))

Max Activating Token Index: 7

Text #17

Max Range: 8.5563. Min Range: -8.5563

Max Act: 7.9938. Min Act: 0.0098

Data Index: 1222114 (C4 (Web Text))

Max Activating Token Index: 9

Text #18

Max Range: 8.5563. Min Range: -8.5563

Max Act: 8.0251. Min Act: -0.1699

Data Index: 1130368 (C4 (Web Text))

Max Activating Token Index: 7

Text #19

Max Range: 8.5563. Min Range: -8.5563

Max Act: 7.9938. Min Act: -0.1112

Data Index: 246435 (C4 (Web Text))

Max Activating Token Index: 4