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

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

Neuron 1906 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.2698. Min Act: -0.1700

Data Index: 830647 (C4 (Web Text))

Max Activating Token Index: 666

Text #1

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.2698. Min Act: -0.1700

Data Index: 1520596 (Python Code)

Max Activating Token Index: 856

Text #2

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.1597. Min Act: -0.1700

Data Index: 112059 (C4 (Web Text))

Max Activating Token Index: 292

Text #3

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.1754. Min Act: -0.1700

Data Index: 1646738 (Python Code)

Max Activating Token Index: 364

Text #4

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.1282. Min Act: -0.1700

Data Index: 790915 (C4 (Web Text))

Max Activating Token Index: 791

Text #5

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0809. Min Act: -0.1700

Data Index: 790433 (C4 (Web Text))

Max Activating Token Index: 947

Text #6

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0966. Min Act: -0.1700

Data Index: 1442774 (Python Code)

Max Activating Token Index: 977

Text #7

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0809. Min Act: -0.1700

Data Index: 1226569 (C4 (Web Text))

Max Activating Token Index: 900

Text #8

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0966. Min Act: -0.1700

Data Index: 1552534 (Python Code)

Max Activating Token Index: 273

Text #9

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0177. Min Act: -0.1700

Data Index: 1428732 (Python Code)

Max Activating Token Index: 619

Text #10

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0335. Min Act: -0.1700

Data Index: 1610074 (Python Code)

Max Activating Token Index: 328

Text #11

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0177. Min Act: -0.1700

Data Index: 465760 (C4 (Web Text))

Max Activating Token Index: 709

Text #12

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0177. Min Act: -0.1700

Data Index: 1428055 (Python Code)

Max Activating Token Index: 736

Text #13

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0493. Min Act: -0.1700

Data Index: 1404832 (Python Code)

Max Activating Token Index: 379

Text #14

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0335. Min Act: -0.1700

Data Index: 1379879 (Python Code)

Max Activating Token Index: 472

Text #15

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0335. Min Act: -0.1700

Data Index: 1492295 (Python Code)

Max Activating Token Index: 851

Text #16

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0335. Min Act: -0.1700

Data Index: 1377930 (Python Code)

Max Activating Token Index: 210

Text #17

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0335. Min Act: -0.1700

Data Index: 1647387 (Python Code)

Max Activating Token Index: 675

Text #18

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0019. Min Act: -0.1700

Data Index: 1561177 (Python Code)

Max Activating Token Index: 120

Text #19

Max Range: 3.2698. Min Range: -3.2698

Max Act: 3.0177. Min Act: -0.1700

Data Index: 992244 (C4 (Web Text))

Max Activating Token Index: 206