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

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

Neuron 655 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.8549. Min Act: -0.1700

Data Index: 3792 (C4 (Web Text))

Max Activating Token Index: 285

Text #1

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.7912. Min Act: -0.1697

Data Index: 518507 (C4 (Web Text))

Max Activating Token Index: 606

Text #2

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.7753. Min Act: -0.1693

Data Index: 1145099 (C4 (Web Text))

Max Activating Token Index: 109

Text #3

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.7274. Min Act: -0.1700

Data Index: 548076 (C4 (Web Text))

Max Activating Token Index: 596

Text #4

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.7114. Min Act: -0.1699

Data Index: 863369 (C4 (Web Text))

Max Activating Token Index: 688

Text #5

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.7114. Min Act: -0.1697

Data Index: 371460 (C4 (Web Text))

Max Activating Token Index: 651

Text #6

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6954. Min Act: -0.1633

Data Index: 1320413 (C4 (Web Text))

Max Activating Token Index: 240

Text #7

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6794. Min Act: -0.1699

Data Index: 1160257 (C4 (Web Text))

Max Activating Token Index: 573

Text #8

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6794. Min Act: -0.1699

Data Index: 761287 (C4 (Web Text))

Max Activating Token Index: 887

Text #9

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6794. Min Act: -0.1700

Data Index: 1330815 (C4 (Web Text))

Max Activating Token Index: 756

Text #10

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6634. Min Act: -0.1696

Data Index: 270239 (C4 (Web Text))

Max Activating Token Index: 126

Text #11

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6634. Min Act: -0.1696

Data Index: 1258476 (C4 (Web Text))

Max Activating Token Index: 819

Text #12

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6473. Min Act: -0.1687

Data Index: 152038 (C4 (Web Text))

Max Activating Token Index: 546

Text #13

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6313. Min Act: -0.1700

Data Index: 873500 (C4 (Web Text))

Max Activating Token Index: 690

Text #14

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6313. Min Act: -0.1700

Data Index: 790658 (C4 (Web Text))

Max Activating Token Index: 232

Text #15

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6473. Min Act: -0.1692

Data Index: 593791 (C4 (Web Text))

Max Activating Token Index: 141

Text #16

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6473. Min Act: -0.1699

Data Index: 519439 (C4 (Web Text))

Max Activating Token Index: 881

Text #17

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6313. Min Act: -0.1700

Data Index: 685196 (C4 (Web Text))

Max Activating Token Index: 990

Text #18

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6313. Min Act: -0.1686

Data Index: 373390 (C4 (Web Text))

Max Activating Token Index: 141

Text #19

Max Range: 2.8549. Min Range: -2.8549

Max Act: 2.6313. Min Act: -0.1699

Data Index: 294827 (C4 (Web Text))

Max Activating Token Index: 190