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

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

Neuron 1146 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: 3.6805. Min Range: -3.6805

Max Act: 3.6805. Min Act: -0.1700

Data Index: 913519 (C4 (Web Text))

Max Activating Token Index: 393

Text #1

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.6335. Min Act: -0.1700

Data Index: 726031 (C4 (Web Text))

Max Activating Token Index: 800

Text #2

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.6179. Min Act: -0.1700

Data Index: 961120 (C4 (Web Text))

Max Activating Token Index: 485

Text #3

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5865. Min Act: -0.1700

Data Index: 257461 (C4 (Web Text))

Max Activating Token Index: 55

Text #4

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5865. Min Act: -0.1700

Data Index: 532884 (C4 (Web Text))

Max Activating Token Index: 67

Text #5

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5865. Min Act: -0.1700

Data Index: 314288 (C4 (Web Text))

Max Activating Token Index: 251

Text #6

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5865. Min Act: -0.1700

Data Index: 727040 (C4 (Web Text))

Max Activating Token Index: 591

Text #7

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5865. Min Act: -0.1700

Data Index: 199173 (C4 (Web Text))

Max Activating Token Index: 746

Text #8

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5552. Min Act: -0.1700

Data Index: 99494 (C4 (Web Text))

Max Activating Token Index: 283

Text #9

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5865. Min Act: -0.1700

Data Index: 192357 (C4 (Web Text))

Max Activating Token Index: 868

Text #10

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5396. Min Act: -0.1700

Data Index: 331360 (C4 (Web Text))

Max Activating Token Index: 132

Text #11

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5396. Min Act: -0.1700

Data Index: 642407 (C4 (Web Text))

Max Activating Token Index: 545

Text #12

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5396. Min Act: -0.1700

Data Index: 519808 (C4 (Web Text))

Max Activating Token Index: 408

Text #13

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5239. Min Act: -0.1700

Data Index: 1175802 (C4 (Web Text))

Max Activating Token Index: 50

Text #14

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5082. Min Act: -0.1700

Data Index: 1339611 (C4 (Web Text))

Max Activating Token Index: 926

Text #15

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5396. Min Act: -0.1700

Data Index: 631286 (C4 (Web Text))

Max Activating Token Index: 238

Text #16

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5239. Min Act: -0.1700

Data Index: 1449020 (Python Code)

Max Activating Token Index: 235

Text #17

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5082. Min Act: -0.1700

Data Index: 592790 (C4 (Web Text))

Max Activating Token Index: 712

Text #18

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5082. Min Act: -0.1700

Data Index: 266543 (C4 (Web Text))

Max Activating Token Index: 122

Text #19

Max Range: 3.6805. Min Range: -3.6805

Max Act: 3.5082. Min Act: -0.1700

Data Index: 143393 (C4 (Web Text))

Max Activating Token Index: 227