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

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

Neuron 24 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: 7.1042. Min Range: -7.1042

Max Act: 7.1042. Min Act: -0.1700

Data Index: 1317486 (C4 (Web Text))

Max Activating Token Index: 862

Text #1

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.9792. Min Act: -0.1700

Data Index: 157390 (C4 (Web Text))

Max Activating Token Index: 785

Text #2

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.8229. Min Act: -0.1700

Data Index: 1546902 (Python Code)

Max Activating Token Index: 61

Text #3

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.7604. Min Act: -0.1699

Data Index: 274722 (C4 (Web Text))

Max Activating Token Index: 67

Text #4

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.6667. Min Act: -0.1700

Data Index: 241657 (C4 (Web Text))

Max Activating Token Index: 982

Text #5

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.6979. Min Act: -0.1700

Data Index: 859023 (C4 (Web Text))

Max Activating Token Index: 178

Text #6

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.6667. Min Act: -0.1700

Data Index: 525866 (C4 (Web Text))

Max Activating Token Index: 730

Text #7

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.6354. Min Act: -0.1700

Data Index: 1323574 (C4 (Web Text))

Max Activating Token Index: 904

Text #8

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.6354. Min Act: -0.1699

Data Index: 694225 (C4 (Web Text))

Max Activating Token Index: 341

Text #9

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.6667. Min Act: -0.1700

Data Index: 640765 (C4 (Web Text))

Max Activating Token Index: 58

Text #10

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.6354. Min Act: -0.1700

Data Index: 650755 (C4 (Web Text))

Max Activating Token Index: 355

Text #11

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.6042. Min Act: -0.1700

Data Index: 822647 (C4 (Web Text))

Max Activating Token Index: 43

Text #12

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.5729. Min Act: -0.1699

Data Index: 780907 (C4 (Web Text))

Max Activating Token Index: 41

Text #13

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.6042. Min Act: -0.1700

Data Index: 1347758 (C4 (Web Text))

Max Activating Token Index: 732

Text #14

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.6042. Min Act: -0.1700

Data Index: 518353 (C4 (Web Text))

Max Activating Token Index: 220

Text #15

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.5417. Min Act: -0.1700

Data Index: 1423316 (Python Code)

Max Activating Token Index: 32

Text #16

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.5417. Min Act: -0.1700

Data Index: 35208 (C4 (Web Text))

Max Activating Token Index: 654

Text #17

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.4792. Min Act: -0.1698

Data Index: 80867 (C4 (Web Text))

Max Activating Token Index: 716

Text #18

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.4479. Min Act: -0.1699

Data Index: 593948 (C4 (Web Text))

Max Activating Token Index: 896

Text #19

Max Range: 7.1042. Min Range: -7.1042

Max Act: 6.5729. Min Act: -0.1700

Data Index: 840263 (C4 (Web Text))

Max Activating Token Index: 719