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

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

Neuron 715 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.1824. Min Range: -5.1824

Max Act: 5.1824. Min Act: -0.1700

Data Index: 942998 (C4 (Web Text))

Max Activating Token Index: 160

Text #1

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.9949. Min Act: -0.1700

Data Index: 77853 (C4 (Web Text))

Max Activating Token Index: 152

Text #2

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.9636. Min Act: -0.1700

Data Index: 959314 (C4 (Web Text))

Max Activating Token Index: 73

Text #3

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.9324. Min Act: -0.1700

Data Index: 269369 (C4 (Web Text))

Max Activating Token Index: 49

Text #4

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.8073. Min Act: -0.1700

Data Index: 533465 (C4 (Web Text))

Max Activating Token Index: 1016

Text #5

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.9011. Min Act: -0.1700

Data Index: 1177381 (C4 (Web Text))

Max Activating Token Index: 72

Text #6

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.8699. Min Act: -0.1700

Data Index: 637723 (C4 (Web Text))

Max Activating Token Index: 338

Text #7

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.8386. Min Act: -0.1700

Data Index: 734925 (C4 (Web Text))

Max Activating Token Index: 213

Text #8

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.8699. Min Act: -0.1700

Data Index: 607665 (C4 (Web Text))

Max Activating Token Index: 136

Text #9

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.7761. Min Act: -0.1700

Data Index: 339172 (C4 (Web Text))

Max Activating Token Index: 656

Text #10

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.8073. Min Act: -0.1700

Data Index: 1182419 (C4 (Web Text))

Max Activating Token Index: 20

Text #11

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.8073. Min Act: -0.1700

Data Index: 308557 (C4 (Web Text))

Max Activating Token Index: 512

Text #12

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.8073. Min Act: -0.1700

Data Index: 734893 (C4 (Web Text))

Max Activating Token Index: 792

Text #13

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.8073. Min Act: -0.1700

Data Index: 37457 (C4 (Web Text))

Max Activating Token Index: 22

Text #14

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.6823. Min Act: -0.1700

Data Index: 1191584 (C4 (Web Text))

Max Activating Token Index: 847

Text #15

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.7761. Min Act: -0.1700

Data Index: 314370 (C4 (Web Text))

Max Activating Token Index: 39

Text #16

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.7448. Min Act: -0.1700

Data Index: 939995 (C4 (Web Text))

Max Activating Token Index: 914

Text #17

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.7136. Min Act: -0.1700

Data Index: 633379 (C4 (Web Text))

Max Activating Token Index: 56

Text #18

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.7448. Min Act: -0.1700

Data Index: 824668 (C4 (Web Text))

Max Activating Token Index: 201

Text #19

Max Range: 5.1824. Min Range: -5.1824

Max Act: 4.7761. Min Act: -0.1700

Data Index: 274409 (C4 (Web Text))

Max Activating Token Index: 315