Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1970 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.3111. Min Range: -5.3111

Max Act: 5.3111. Min Act: -0.1700

Data Index: 1557516 (Python Code)

Max Activating Token Index: 777

Text #1

Max Range: 5.3111. Min Range: -5.3111

Max Act: 5.2799. Min Act: -0.1700

Data Index: 1471567 (Python Code)

Max Activating Token Index: 54

Text #2

Max Range: 5.3111. Min Range: -5.3111

Max Act: 5.1236. Min Act: -0.1700

Data Index: 1418053 (Python Code)

Max Activating Token Index: 953

Text #3

Max Range: 5.3111. Min Range: -5.3111

Max Act: 5.0611. Min Act: -0.1700

Data Index: 1508639 (Python Code)

Max Activating Token Index: 603

Text #4

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.9674. Min Act: -0.1700

Data Index: 957308 (C4 (Web Text))

Max Activating Token Index: 186

Text #5

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.9986. Min Act: -0.1700

Data Index: 1425836 (Python Code)

Max Activating Token Index: 138

Text #6

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.9361. Min Act: -0.1700

Data Index: 1548267 (Python Code)

Max Activating Token Index: 889

Text #7

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.8736. Min Act: -0.1700

Data Index: 1095522 (C4 (Web Text))

Max Activating Token Index: 607

Text #8

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.8736. Min Act: -0.1700

Data Index: 1562951 (Python Code)

Max Activating Token Index: 346

Text #9

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.8424. Min Act: -0.1700

Data Index: 1577982 (Python Code)

Max Activating Token Index: 359

Text #10

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.8736. Min Act: -0.1700

Data Index: 1576398 (Python Code)

Max Activating Token Index: 22

Text #11

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.8736. Min Act: -0.1700

Data Index: 1609789 (Python Code)

Max Activating Token Index: 512

Text #12

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.8736. Min Act: -0.1700

Data Index: 773561 (C4 (Web Text))

Max Activating Token Index: 730

Text #13

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.8111. Min Act: -0.1700

Data Index: 246508 (C4 (Web Text))

Max Activating Token Index: 1022

Text #14

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.7486. Min Act: -0.1700

Data Index: 1392586 (Python Code)

Max Activating Token Index: 836

Text #15

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.7799. Min Act: -0.1700

Data Index: 1622950 (Python Code)

Max Activating Token Index: 397

Text #16

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.8424. Min Act: -0.1700

Data Index: 149265 (C4 (Web Text))

Max Activating Token Index: 29

Text #17

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.8111. Min Act: -0.1700

Data Index: 444185 (C4 (Web Text))

Max Activating Token Index: 643

Text #18

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.7799. Min Act: -0.1700

Data Index: 1336088 (C4 (Web Text))

Max Activating Token Index: 158

Text #19

Max Range: 5.3111. Min Range: -5.3111

Max Act: 4.8111. Min Act: -0.1700

Data Index: 1343904 (C4 (Web Text))

Max Activating Token Index: 114