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

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

Neuron 37 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.7983. Min Act: -0.1700

Data Index: 1335142 (C4 (Web Text))

Max Activating Token Index: 817

Text #1

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.7045. Min Act: -0.1700

Data Index: 403021 (C4 (Web Text))

Max Activating Token Index: 1012

Text #2

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6732. Min Act: -0.1700

Data Index: 1084824 (C4 (Web Text))

Max Activating Token Index: 687

Text #3

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6419. Min Act: -0.1700

Data Index: 885372 (C4 (Web Text))

Max Activating Token Index: 971

Text #4

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6732. Min Act: -0.1700

Data Index: 360787 (C4 (Web Text))

Max Activating Token Index: 193

Text #5

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6732. Min Act: -0.1700

Data Index: 1171785 (C4 (Web Text))

Max Activating Token Index: 173

Text #6

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6732. Min Act: -0.1700

Data Index: 1238238 (C4 (Web Text))

Max Activating Token Index: 204

Text #7

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6106. Min Act: -0.1700

Data Index: 693530 (C4 (Web Text))

Max Activating Token Index: 974

Text #8

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6106. Min Act: -0.1700

Data Index: 283600 (C4 (Web Text))

Max Activating Token Index: 840

Text #9

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6419. Min Act: -0.1700

Data Index: 1315097 (C4 (Web Text))

Max Activating Token Index: 216

Text #10

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6106. Min Act: -0.1700

Data Index: 1009792 (C4 (Web Text))

Max Activating Token Index: 65

Text #11

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6419. Min Act: -0.1700

Data Index: 497148 (C4 (Web Text))

Max Activating Token Index: 648

Text #12

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6419. Min Act: -0.1700

Data Index: 1013417 (C4 (Web Text))

Max Activating Token Index: 413

Text #13

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6419. Min Act: -0.1700

Data Index: 748533 (C4 (Web Text))

Max Activating Token Index: 888

Text #14

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6419. Min Act: -0.1700

Data Index: 1105665 (C4 (Web Text))

Max Activating Token Index: 561

Text #15

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6419. Min Act: -0.1700

Data Index: 339543 (C4 (Web Text))

Max Activating Token Index: 826

Text #16

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6106. Min Act: -0.1700

Data Index: 858099 (C4 (Web Text))

Max Activating Token Index: 589

Text #17

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.5793. Min Act: -0.1700

Data Index: 465759 (C4 (Web Text))

Max Activating Token Index: 681

Text #18

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.6106. Min Act: -0.1700

Data Index: 586366 (C4 (Web Text))

Max Activating Token Index: 280

Text #19

Max Range: 3.7983. Min Range: -3.7983

Max Act: 3.5793. Min Act: -0.1700

Data Index: 550544 (C4 (Web Text))

Max Activating Token Index: 573