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

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

Neuron 1831 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.3035. Min Act: -0.1700

Data Index: 218894 (C4 (Web Text))

Max Activating Token Index: 907

Text #1

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.2249. Min Act: -0.1700

Data Index: 1177156 (C4 (Web Text))

Max Activating Token Index: 1012

Text #2

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.1619. Min Act: -0.1700

Data Index: 1090443 (C4 (Web Text))

Max Activating Token Index: 897

Text #3

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.1462. Min Act: -0.1700

Data Index: 783785 (C4 (Web Text))

Max Activating Token Index: 612

Text #4

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0989. Min Act: -0.1700

Data Index: 223017 (C4 (Web Text))

Max Activating Token Index: 83

Text #5

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0831. Min Act: -0.1700

Data Index: 258540 (C4 (Web Text))

Max Activating Token Index: 556

Text #6

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0674. Min Act: -0.1700

Data Index: 941296 (C4 (Web Text))

Max Activating Token Index: 18

Text #7

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0516. Min Act: -0.1700

Data Index: 404051 (C4 (Web Text))

Max Activating Token Index: 32

Text #8

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0674. Min Act: -0.1700

Data Index: 154398 (C4 (Web Text))

Max Activating Token Index: 283

Text #9

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0200. Min Act: -0.1700

Data Index: 658268 (C4 (Web Text))

Max Activating Token Index: 65

Text #10

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0358. Min Act: -0.1700

Data Index: 601790 (C4 (Web Text))

Max Activating Token Index: 13

Text #11

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0200. Min Act: -0.1700

Data Index: 602289 (C4 (Web Text))

Max Activating Token Index: 169

Text #12

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0200. Min Act: -0.1700

Data Index: 1181960 (C4 (Web Text))

Max Activating Token Index: 952

Text #13

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0200. Min Act: -0.1700

Data Index: 442957 (C4 (Web Text))

Max Activating Token Index: 555

Text #14

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0200. Min Act: -0.1700

Data Index: 526943 (C4 (Web Text))

Max Activating Token Index: 645

Text #15

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0200. Min Act: -0.1700

Data Index: 315053 (C4 (Web Text))

Max Activating Token Index: 971

Text #16

Max Range: 3.3035. Min Range: -3.3035

Max Act: 2.9726. Min Act: -0.1700

Data Index: 1199743 (C4 (Web Text))

Max Activating Token Index: 34

Text #17

Max Range: 3.3035. Min Range: -3.3035

Max Act: 3.0042. Min Act: -0.1700

Data Index: 144851 (C4 (Web Text))

Max Activating Token Index: 7

Text #18

Max Range: 3.3035. Min Range: -3.3035

Max Act: 2.9884. Min Act: -0.1700

Data Index: 528622 (C4 (Web Text))

Max Activating Token Index: 860

Text #19

Max Range: 3.3035. Min Range: -3.3035

Max Act: 2.9726. Min Act: -0.1700

Data Index: 906089 (C4 (Web Text))

Max Activating Token Index: 573