Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 17 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.3502. Min Act: -0.1700

Data Index: 760640 (C4 (Web Text))

Max Activating Token Index: 470

Text #1

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.2877. Min Act: -0.1700

Data Index: 781932 (C4 (Web Text))

Max Activating Token Index: 264

Text #2

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.1940. Min Act: -0.1700

Data Index: 966780 (C4 (Web Text))

Max Activating Token Index: 23

Text #3

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.0690. Min Act: -0.1700

Data Index: 166891 (C4 (Web Text))

Max Activating Token Index: 924

Text #4

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.1315. Min Act: -0.1700

Data Index: 156834 (C4 (Web Text))

Max Activating Token Index: 366

Text #5

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.1940. Min Act: -0.1700

Data Index: 192426 (C4 (Web Text))

Max Activating Token Index: 274

Text #6

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.1627. Min Act: -0.1700

Data Index: 444251 (C4 (Web Text))

Max Activating Token Index: 593

Text #7

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.1315. Min Act: -0.1700

Data Index: 219910 (C4 (Web Text))

Max Activating Token Index: 373

Text #8

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.1002. Min Act: -0.1700

Data Index: 291999 (C4 (Web Text))

Max Activating Token Index: 290

Text #9

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.1315. Min Act: -0.1700

Data Index: 1245651 (C4 (Web Text))

Max Activating Token Index: 764

Text #10

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.0377. Min Act: -0.1700

Data Index: 809877 (C4 (Web Text))

Max Activating Token Index: 777

Text #11

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.0065. Min Act: -0.1700

Data Index: 1065388 (C4 (Web Text))

Max Activating Token Index: 834

Text #12

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.0065. Min Act: -0.1700

Data Index: 248037 (C4 (Web Text))

Max Activating Token Index: 577

Text #13

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.0690. Min Act: -0.1700

Data Index: 875650 (C4 (Web Text))

Max Activating Token Index: 784

Text #14

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.1002. Min Act: -0.1700

Data Index: 1304238 (C4 (Web Text))

Max Activating Token Index: 921

Text #15

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.0065. Min Act: -0.1700

Data Index: 800310 (C4 (Web Text))

Max Activating Token Index: 269

Text #16

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.0377. Min Act: -0.1700

Data Index: 639450 (C4 (Web Text))

Max Activating Token Index: 928

Text #17

Max Range: 5.3502. Min Range: -5.3502

Max Act: 4.9752. Min Act: -0.1700

Data Index: 791531 (C4 (Web Text))

Max Activating Token Index: 814

Text #18

Max Range: 5.3502. Min Range: -5.3502

Max Act: 5.0377. Min Act: -0.1700

Data Index: 322992 (C4 (Web Text))

Max Activating Token Index: 467

Text #19

Max Range: 5.3502. Min Range: -5.3502

Max Act: 4.9752. Min Act: -0.1700

Data Index: 946598 (C4 (Web Text))

Max Activating Token Index: 794