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

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

Neuron 1622 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: 7.8723. Min Range: -7.8723

Max Act: 7.8723. Min Act: -0.1699

Data Index: 571208 (C4 (Web Text))

Max Activating Token Index: 195

Text #1

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.6536. Min Act: -0.1700

Data Index: 48542 (C4 (Web Text))

Max Activating Token Index: 644

Text #2

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.5286. Min Act: -0.1700

Data Index: 795958 (C4 (Web Text))

Max Activating Token Index: 79

Text #3

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.6223. Min Act: -0.1700

Data Index: 407494 (C4 (Web Text))

Max Activating Token Index: 125

Text #4

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.5911. Min Act: -0.1700

Data Index: 1204473 (C4 (Web Text))

Max Activating Token Index: 238

Text #5

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.6223. Min Act: -0.1700

Data Index: 1046922 (C4 (Web Text))

Max Activating Token Index: 732

Text #6

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.4973. Min Act: -0.1700

Data Index: 1069866 (C4 (Web Text))

Max Activating Token Index: 528

Text #7

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.4348. Min Act: -0.1700

Data Index: 810764 (C4 (Web Text))

Max Activating Token Index: 928

Text #8

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.4348. Min Act: -0.1700

Data Index: 236992 (C4 (Web Text))

Max Activating Token Index: 359

Text #9

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.3411. Min Act: -0.1700

Data Index: 457166 (C4 (Web Text))

Max Activating Token Index: 547

Text #10

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.4036. Min Act: -0.1700

Data Index: 915370 (C4 (Web Text))

Max Activating Token Index: 826

Text #11

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.3723. Min Act: -0.1700

Data Index: 384313 (C4 (Web Text))

Max Activating Token Index: 533

Text #12

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.2161. Min Act: -0.1700

Data Index: 930829 (C4 (Web Text))

Max Activating Token Index: 425

Text #13

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.2786. Min Act: -0.1700

Data Index: 326461 (C4 (Web Text))

Max Activating Token Index: 662

Text #14

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.2161. Min Act: -0.1691

Data Index: 198897 (C4 (Web Text))

Max Activating Token Index: 557

Text #15

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.2786. Min Act: -0.1700

Data Index: 1103934 (C4 (Web Text))

Max Activating Token Index: 430

Text #16

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.2161. Min Act: -0.1700

Data Index: 666481 (C4 (Web Text))

Max Activating Token Index: 342

Text #17

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.1536. Min Act: -0.1700

Data Index: 1650123 (Python Code)

Max Activating Token Index: 676

Text #18

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.1848. Min Act: -0.1700

Data Index: 352930 (C4 (Web Text))

Max Activating Token Index: 84

Text #19

Max Range: 7.8723. Min Range: -7.8723

Max Act: 7.2161. Min Act: -0.1700

Data Index: 686295 (C4 (Web Text))

Max Activating Token Index: 590