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

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

Neuron 533 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: 8.6555. Min Range: -8.6555

Max Act: 8.6555. Min Act: -0.1445

Data Index: 1255791 (C4 (Web Text))

Max Activating Token Index: 796

Text #1

Max Range: 8.6555. Min Range: -8.6555

Max Act: 8.2180. Min Act: -0.1700

Data Index: 146745 (C4 (Web Text))

Max Activating Token Index: 681

Text #2

Max Range: 8.6555. Min Range: -8.6555

Max Act: 8.0930. Min Act: -0.1695

Data Index: 529214 (C4 (Web Text))

Max Activating Token Index: 210

Text #3

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.7180. Min Act: -0.1700

Data Index: 1029670 (C4 (Web Text))

Max Activating Token Index: 674

Text #4

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.7180. Min Act: -0.1700

Data Index: 1030419 (C4 (Web Text))

Max Activating Token Index: 742

Text #5

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.5930. Min Act: -0.1700

Data Index: 395817 (C4 (Web Text))

Max Activating Token Index: 455

Text #6

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.5305. Min Act: -0.1696

Data Index: 366448 (C4 (Web Text))

Max Activating Token Index: 741

Text #7

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.4680. Min Act: -0.1700

Data Index: 640059 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.4680. Min Act: -0.1700

Data Index: 21733 (C4 (Web Text))

Max Activating Token Index: 492

Text #9

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.4680. Min Act: -0.1700

Data Index: 1221914 (C4 (Web Text))

Max Activating Token Index: 90

Text #10

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.4680. Min Act: -0.1699

Data Index: 765766 (C4 (Web Text))

Max Activating Token Index: 387

Text #11

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.2805. Min Act: -0.1700

Data Index: 965654 (C4 (Web Text))

Max Activating Token Index: 224

Text #12

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.2180. Min Act: -0.1700

Data Index: 625790 (C4 (Web Text))

Max Activating Token Index: 518

Text #13

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.1868. Min Act: -0.1700

Data Index: 145527 (C4 (Web Text))

Max Activating Token Index: 180

Text #14

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.2180. Min Act: -0.1700

Data Index: 100466 (C4 (Web Text))

Max Activating Token Index: 711

Text #15

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.1868. Min Act: -0.1700

Data Index: 697205 (C4 (Web Text))

Max Activating Token Index: 469

Text #16

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.1555. Min Act: -0.1699

Data Index: 396712 (C4 (Web Text))

Max Activating Token Index: 699

Text #17

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.1555. Min Act: -0.1699

Data Index: 848333 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.1555. Min Act: -0.1691

Data Index: 941396 (C4 (Web Text))

Max Activating Token Index: 609

Text #19

Max Range: 8.6555. Min Range: -8.6555

Max Act: 7.1243. Min Act: -0.1700

Data Index: 621063 (C4 (Web Text))

Max Activating Token Index: 680