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

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

Neuron 676 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: 5.1719. Min Range: -5.1719

Max Act: 5.1719. Min Act: -0.1700

Data Index: 1373951 (Python Code)

Max Activating Token Index: 503

Text #1

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.9219. Min Act: -0.1700

Data Index: 751659 (C4 (Web Text))

Max Activating Token Index: 579

Text #2

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.6250. Min Act: -0.1700

Data Index: 794852 (C4 (Web Text))

Max Activating Token Index: 800

Text #3

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.6094. Min Act: -0.1700

Data Index: 922451 (C4 (Web Text))

Max Activating Token Index: 915

Text #4

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.5938. Min Act: -0.1700

Data Index: 360241 (C4 (Web Text))

Max Activating Token Index: 85

Text #5

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.5781. Min Act: -0.1700

Data Index: 1187906 (C4 (Web Text))

Max Activating Token Index: 1007

Text #6

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.6094. Min Act: -0.1700

Data Index: 585763 (C4 (Web Text))

Max Activating Token Index: 743

Text #7

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.5313. Min Act: -0.1700

Data Index: 934751 (C4 (Web Text))

Max Activating Token Index: 874

Text #8

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.5313. Min Act: -0.1700

Data Index: 1448669 (Python Code)

Max Activating Token Index: 105

Text #9

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.4687. Min Act: -0.1700

Data Index: 347711 (C4 (Web Text))

Max Activating Token Index: 767

Text #10

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.5156. Min Act: -0.1700

Data Index: 1103820 (C4 (Web Text))

Max Activating Token Index: 61

Text #11

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.5313. Min Act: -0.1700

Data Index: 904232 (C4 (Web Text))

Max Activating Token Index: 387

Text #12

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.5000. Min Act: -0.1700

Data Index: 609136 (C4 (Web Text))

Max Activating Token Index: 53

Text #13

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.4531. Min Act: -0.1700

Data Index: 988212 (C4 (Web Text))

Max Activating Token Index: 659

Text #14

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.4687. Min Act: -0.1700

Data Index: 457771 (C4 (Web Text))

Max Activating Token Index: 327

Text #15

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.4531. Min Act: -0.1700

Data Index: 967786 (C4 (Web Text))

Max Activating Token Index: 632

Text #16

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.4375. Min Act: -0.1700

Data Index: 688721 (C4 (Web Text))

Max Activating Token Index: 557

Text #17

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.4375. Min Act: -0.1700

Data Index: 922963 (C4 (Web Text))

Max Activating Token Index: 626

Text #18

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.4219. Min Act: -0.1700

Data Index: 1265194 (C4 (Web Text))

Max Activating Token Index: 145

Text #19

Max Range: 5.1719. Min Range: -5.1719

Max Act: 4.4062. Min Act: -0.1700

Data Index: 803343 (C4 (Web Text))

Max Activating Token Index: 575