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

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

Neuron 400 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.6176. Min Act: -0.1700

Data Index: 581252 (C4 (Web Text))

Max Activating Token Index: 737

Text #1

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.5238. Min Act: -0.1700

Data Index: 156999 (C4 (Web Text))

Max Activating Token Index: 747

Text #2

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.4301. Min Act: -0.1700

Data Index: 1312487 (C4 (Web Text))

Max Activating Token Index: 758

Text #3

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.3988. Min Act: -0.1700

Data Index: 1225040 (C4 (Web Text))

Max Activating Token Index: 710

Text #4

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.2738. Min Act: -0.1700

Data Index: 157848 (C4 (Web Text))

Max Activating Token Index: 622

Text #5

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.3051. Min Act: -0.1700

Data Index: 37453 (C4 (Web Text))

Max Activating Token Index: 830

Text #6

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.2426. Min Act: -0.1700

Data Index: 1274799 (C4 (Web Text))

Max Activating Token Index: 516

Text #7

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.2113. Min Act: -0.1700

Data Index: 16963 (C4 (Web Text))

Max Activating Token Index: 514

Text #8

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.2426. Min Act: -0.1700

Data Index: 385837 (C4 (Web Text))

Max Activating Token Index: 642

Text #9

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.1801. Min Act: -0.1700

Data Index: 712399 (C4 (Web Text))

Max Activating Token Index: 848

Text #10

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.2113. Min Act: -0.1700

Data Index: 465507 (C4 (Web Text))

Max Activating Token Index: 800

Text #11

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.1801. Min Act: -0.1700

Data Index: 1128681 (C4 (Web Text))

Max Activating Token Index: 214

Text #12

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.0551. Min Act: -0.1700

Data Index: 397344 (C4 (Web Text))

Max Activating Token Index: 74

Text #13

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.1488. Min Act: -0.1700

Data Index: 290461 (C4 (Web Text))

Max Activating Token Index: 167

Text #14

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.0551. Min Act: -0.1700

Data Index: 1114863 (C4 (Web Text))

Max Activating Token Index: 1009

Text #15

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.1176. Min Act: -0.1700

Data Index: 1295577 (C4 (Web Text))

Max Activating Token Index: 155

Text #16

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.0238. Min Act: -0.1700

Data Index: 1127715 (C4 (Web Text))

Max Activating Token Index: 790

Text #17

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.0551. Min Act: -0.1700

Data Index: 1181278 (C4 (Web Text))

Max Activating Token Index: 617

Text #18

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.0551. Min Act: -0.1700

Data Index: 571755 (C4 (Web Text))

Max Activating Token Index: 781

Text #19

Max Range: 5.6176. Min Range: -5.6176

Max Act: 5.1176. Min Act: -0.1700

Data Index: 955093 (C4 (Web Text))

Max Activating Token Index: 577