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

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

Neuron 666 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: 3.6867. Min Range: -3.6867

Max Act: 3.6867. Min Act: -0.1700

Data Index: 204422 (C4 (Web Text))

Max Activating Token Index: 444

Text #1

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.6867. Min Act: -0.1700

Data Index: 808614 (C4 (Web Text))

Max Activating Token Index: 540

Text #2

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.6867. Min Act: -0.1700

Data Index: 356487 (C4 (Web Text))

Max Activating Token Index: 853

Text #3

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.6867. Min Act: -0.1700

Data Index: 199711 (C4 (Web Text))

Max Activating Token Index: 708

Text #4

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.6554. Min Act: -0.1700

Data Index: 1073062 (C4 (Web Text))

Max Activating Token Index: 576

Text #5

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.6554. Min Act: -0.1700

Data Index: 352566 (C4 (Web Text))

Max Activating Token Index: 282

Text #6

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.6241. Min Act: -0.1700

Data Index: 203296 (C4 (Web Text))

Max Activating Token Index: 666

Text #7

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5771. Min Act: -0.1700

Data Index: 890875 (C4 (Web Text))

Max Activating Token Index: 341

Text #8

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5928. Min Act: -0.1700

Data Index: 1208097 (C4 (Web Text))

Max Activating Token Index: 797

Text #9

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5771. Min Act: -0.1700

Data Index: 461008 (C4 (Web Text))

Max Activating Token Index: 768

Text #10

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5928. Min Act: -0.1700

Data Index: 1174662 (C4 (Web Text))

Max Activating Token Index: 98

Text #11

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5614. Min Act: -0.1700

Data Index: 474497 (C4 (Web Text))

Max Activating Token Index: 580

Text #12

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5928. Min Act: -0.1700

Data Index: 548623 (C4 (Web Text))

Max Activating Token Index: 546

Text #13

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5928. Min Act: -0.1700

Data Index: 1063432 (C4 (Web Text))

Max Activating Token Index: 74

Text #14

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5614. Min Act: -0.1700

Data Index: 358688 (C4 (Web Text))

Max Activating Token Index: 80

Text #15

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5771. Min Act: -0.1700

Data Index: 638132 (C4 (Web Text))

Max Activating Token Index: 669

Text #16

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5458. Min Act: -0.1700

Data Index: 1295179 (C4 (Web Text))

Max Activating Token Index: 580

Text #17

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5928. Min Act: -0.1700

Data Index: 416462 (C4 (Web Text))

Max Activating Token Index: 670

Text #18

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5771. Min Act: -0.1700

Data Index: 277618 (C4 (Web Text))

Max Activating Token Index: 540

Text #19

Max Range: 3.6867. Min Range: -3.6867

Max Act: 3.5458. Min Act: -0.1700

Data Index: 196916 (C4 (Web Text))

Max Activating Token Index: 970