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

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

Neuron 33 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.3627. Min Act: -0.1700

Data Index: 1065368 (C4 (Web Text))

Max Activating Token Index: 791

Text #1

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.2213. Min Act: -0.1700

Data Index: 810282 (C4 (Web Text))

Max Activating Token Index: 32

Text #2

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.2213. Min Act: -0.1700

Data Index: 307780 (C4 (Web Text))

Max Activating Token Index: 114

Text #3

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1899. Min Act: -0.1700

Data Index: 492267 (C4 (Web Text))

Max Activating Token Index: 866

Text #4

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.2056. Min Act: -0.1700

Data Index: 132634 (C4 (Web Text))

Max Activating Token Index: 132

Text #5

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1899. Min Act: -0.1700

Data Index: 427172 (C4 (Web Text))

Max Activating Token Index: 582

Text #6

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1584. Min Act: -0.1700

Data Index: 206868 (C4 (Web Text))

Max Activating Token Index: 425

Text #7

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1584. Min Act: -0.1700

Data Index: 1244993 (C4 (Web Text))

Max Activating Token Index: 864

Text #8

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1899. Min Act: -0.1700

Data Index: 786841 (C4 (Web Text))

Max Activating Token Index: 628

Text #9

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1899. Min Act: -0.1700

Data Index: 625681 (C4 (Web Text))

Max Activating Token Index: 820

Text #10

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1741. Min Act: -0.1700

Data Index: 445450 (C4 (Web Text))

Max Activating Token Index: 984

Text #11

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1584. Min Act: -0.1700

Data Index: 579270 (C4 (Web Text))

Max Activating Token Index: 622

Text #12

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1269. Min Act: -0.1700

Data Index: 468645 (C4 (Web Text))

Max Activating Token Index: 1017

Text #13

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1426. Min Act: -0.1700

Data Index: 322183 (C4 (Web Text))

Max Activating Token Index: 339

Text #14

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1426. Min Act: -0.1700

Data Index: 401253 (C4 (Web Text))

Max Activating Token Index: 208

Text #15

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1111. Min Act: -0.1700

Data Index: 554093 (C4 (Web Text))

Max Activating Token Index: 423

Text #16

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1111. Min Act: -0.1700

Data Index: 389725 (C4 (Web Text))

Max Activating Token Index: 478

Text #17

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1269. Min Act: -0.1700

Data Index: 323809 (C4 (Web Text))

Max Activating Token Index: 46

Text #18

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.1269. Min Act: -0.1700

Data Index: 584036 (C4 (Web Text))

Max Activating Token Index: 22

Text #19

Max Range: 3.3627. Min Range: -3.3627

Max Act: 3.0954. Min Act: -0.1700

Data Index: 1602227 (Python Code)

Max Activating Token Index: 682