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

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

Neuron 1004 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.2858. Min Act: -0.1700

Data Index: 10020 (C4 (Web Text))

Max Activating Token Index: 370

Text #1

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.2544. Min Act: -0.1700

Data Index: 1203183 (C4 (Web Text))

Max Activating Token Index: 282

Text #2

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.2230. Min Act: -0.1700

Data Index: 1015839 (C4 (Web Text))

Max Activating Token Index: 82

Text #3

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.2072. Min Act: -0.1700

Data Index: 358946 (C4 (Web Text))

Max Activating Token Index: 685

Text #4

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.2230. Min Act: -0.1700

Data Index: 488135 (C4 (Web Text))

Max Activating Token Index: 918

Text #5

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.2230. Min Act: -0.1700

Data Index: 306556 (C4 (Web Text))

Max Activating Token Index: 200

Text #6

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.2230. Min Act: -0.1700

Data Index: 457695 (C4 (Web Text))

Max Activating Token Index: 923

Text #7

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.2072. Min Act: -0.1700

Data Index: 1212674 (C4 (Web Text))

Max Activating Token Index: 933

Text #8

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.1758. Min Act: -0.1700

Data Index: 893638 (C4 (Web Text))

Max Activating Token Index: 317

Text #9

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.2072. Min Act: -0.1700

Data Index: 719262 (C4 (Web Text))

Max Activating Token Index: 166

Text #10

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.2072. Min Act: -0.1700

Data Index: 1342277 (C4 (Web Text))

Max Activating Token Index: 112

Text #11

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.1758. Min Act: -0.1700

Data Index: 636469 (C4 (Web Text))

Max Activating Token Index: 320

Text #12

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.1285. Min Act: -0.1700

Data Index: 172656 (C4 (Web Text))

Max Activating Token Index: 920

Text #13

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.1600. Min Act: -0.1700

Data Index: 825474 (C4 (Web Text))

Max Activating Token Index: 606

Text #14

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.1758. Min Act: -0.1700

Data Index: 439552 (C4 (Web Text))

Max Activating Token Index: 63

Text #15

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.1915. Min Act: -0.1700

Data Index: 1087691 (C4 (Web Text))

Max Activating Token Index: 152

Text #16

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.1443. Min Act: -0.1700

Data Index: 191192 (C4 (Web Text))

Max Activating Token Index: 116

Text #17

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.1600. Min Act: -0.1700

Data Index: 515653 (C4 (Web Text))

Max Activating Token Index: 652

Text #18

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.1758. Min Act: -0.1700

Data Index: 206208 (C4 (Web Text))

Max Activating Token Index: 545

Text #19

Max Range: 3.2858. Min Range: -3.2858

Max Act: 3.1600. Min Act: -0.1700

Data Index: 958714 (C4 (Web Text))

Max Activating Token Index: 867