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

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

Neuron 2032 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: 3.9862. Min Range: -3.9862

Max Act: 3.9862. Min Act: -0.1700

Data Index: 930278 (C4 (Web Text))

Max Activating Token Index: 100

Text #1

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.9862. Min Act: -0.1700

Data Index: 1148898 (C4 (Web Text))

Max Activating Token Index: 328

Text #2

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.9237. Min Act: -0.1700

Data Index: 745366 (C4 (Web Text))

Max Activating Token Index: 882

Text #3

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.7360. Min Act: -0.1700

Data Index: 1286068 (C4 (Web Text))

Max Activating Token Index: 831

Text #4

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.7360. Min Act: -0.1699

Data Index: 451995 (C4 (Web Text))

Max Activating Token Index: 276

Text #5

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.7673. Min Act: -0.1700

Data Index: 1035098 (C4 (Web Text))

Max Activating Token Index: 572

Text #6

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.7360. Min Act: -0.1700

Data Index: 250334 (C4 (Web Text))

Max Activating Token Index: 352

Text #7

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.7047. Min Act: -0.1700

Data Index: 233497 (C4 (Web Text))

Max Activating Token Index: 386

Text #8

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.7047. Min Act: -0.1700

Data Index: 549793 (C4 (Web Text))

Max Activating Token Index: 261

Text #9

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.7047. Min Act: -0.1700

Data Index: 537593 (C4 (Web Text))

Max Activating Token Index: 957

Text #10

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.7360. Min Act: -0.1700

Data Index: 89852 (C4 (Web Text))

Max Activating Token Index: 856

Text #11

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.7047. Min Act: -0.1700

Data Index: 1247673 (C4 (Web Text))

Max Activating Token Index: 53

Text #12

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.6734. Min Act: -0.1700

Data Index: 1277870 (C4 (Web Text))

Max Activating Token Index: 896

Text #13

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.6734. Min Act: -0.1700

Data Index: 1344041 (C4 (Web Text))

Max Activating Token Index: 471

Text #14

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.6421. Min Act: -0.1700

Data Index: 330643 (C4 (Web Text))

Max Activating Token Index: 830

Text #15

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.6734. Min Act: -0.1700

Data Index: 101629 (C4 (Web Text))

Max Activating Token Index: 584

Text #16

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.6734. Min Act: -0.1700

Data Index: 629204 (C4 (Web Text))

Max Activating Token Index: 69

Text #17

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.6421. Min Act: -0.1700

Data Index: 92169 (C4 (Web Text))

Max Activating Token Index: 633

Text #18

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.6108. Min Act: -0.1700

Data Index: 731892 (C4 (Web Text))

Max Activating Token Index: 890

Text #19

Max Range: 3.9862. Min Range: -3.9862

Max Act: 3.6734. Min Act: -0.1700

Data Index: 963344 (C4 (Web Text))

Max Activating Token Index: 696