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

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

Neuron 400 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: 3.8551. Min Range: -3.8551

Max Act: 3.8238. Min Act: -0.1700

Data Index: 964886 (C4 (Web Text))

Max Activating Token Index: 51

Text #1

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.8082. Min Act: -0.1700

Data Index: 1144660 (C4 (Web Text))

Max Activating Token Index: 464

Text #2

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.8551. Min Act: -0.1700

Data Index: 452248 (C4 (Web Text))

Max Activating Token Index: 952

Text #3

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6987. Min Act: -0.1700

Data Index: 879430 (C4 (Web Text))

Max Activating Token Index: 319

Text #4

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6987. Min Act: -0.1700

Data Index: 885984 (C4 (Web Text))

Max Activating Token Index: 59

Text #5

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.7143. Min Act: -0.1700

Data Index: 21519 (C4 (Web Text))

Max Activating Token Index: 120

Text #6

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.7300. Min Act: -0.1700

Data Index: 55217 (C4 (Web Text))

Max Activating Token Index: 289

Text #7

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.7143. Min Act: -0.1700

Data Index: 27213 (C4 (Web Text))

Max Activating Token Index: 303

Text #8

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6987. Min Act: -0.1700

Data Index: 619817 (C4 (Web Text))

Max Activating Token Index: 159

Text #9

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6987. Min Act: -0.1700

Data Index: 572338 (C4 (Web Text))

Max Activating Token Index: 607

Text #10

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6830. Min Act: -0.1700

Data Index: 323628 (C4 (Web Text))

Max Activating Token Index: 370

Text #11

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6987. Min Act: -0.1700

Data Index: 353966 (C4 (Web Text))

Max Activating Token Index: 508

Text #12

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6987. Min Act: -0.1700

Data Index: 996497 (C4 (Web Text))

Max Activating Token Index: 357

Text #13

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6987. Min Act: -0.1700

Data Index: 757924 (C4 (Web Text))

Max Activating Token Index: 428

Text #14

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6674. Min Act: -0.1700

Data Index: 297976 (C4 (Web Text))

Max Activating Token Index: 533

Text #15

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6674. Min Act: -0.1700

Data Index: 163514 (C4 (Web Text))

Max Activating Token Index: 759

Text #16

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6517. Min Act: -0.1700

Data Index: 216092 (C4 (Web Text))

Max Activating Token Index: 706

Text #17

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6517. Min Act: -0.1700

Data Index: 253050 (C4 (Web Text))

Max Activating Token Index: 199

Text #18

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6361. Min Act: -0.1700

Data Index: 75628 (C4 (Web Text))

Max Activating Token Index: 143

Text #19

Max Range: 3.8551. Min Range: -3.8551

Max Act: 3.6674. Min Act: -0.1700

Data Index: 992177 (C4 (Web Text))

Max Activating Token Index: 332