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

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

Neuron 1893 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: 4.6642. Min Range: -4.6642

Max Act: 4.6642. Min Act: -0.1700

Data Index: 1196675 (C4 (Web Text))

Max Activating Token Index: 664

Text #1

Max Range: 4.6642. Min Range: -4.6642

Max Act: 4.5079. Min Act: -0.1700

Data Index: 611389 (C4 (Web Text))

Max Activating Token Index: 146

Text #2

Max Range: 4.6642. Min Range: -4.6642

Max Act: 4.4454. Min Act: -0.1700

Data Index: 175638 (C4 (Web Text))

Max Activating Token Index: 773

Text #3

Max Range: 4.6642. Min Range: -4.6642

Max Act: 4.0078. Min Act: -0.1700

Data Index: 69701 (C4 (Web Text))

Max Activating Token Index: 984

Text #4

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.9140. Min Act: -0.1700

Data Index: 482313 (C4 (Web Text))

Max Activating Token Index: 879

Text #5

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.9140. Min Act: -0.1700

Data Index: 802681 (C4 (Web Text))

Max Activating Token Index: 320

Text #6

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.9140. Min Act: -0.1700

Data Index: 539456 (C4 (Web Text))

Max Activating Token Index: 441

Text #7

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8827. Min Act: -0.1700

Data Index: 319299 (C4 (Web Text))

Max Activating Token Index: 989

Text #8

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8827. Min Act: -0.1700

Data Index: 947497 (C4 (Web Text))

Max Activating Token Index: 1010

Text #9

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.9140. Min Act: -0.1700

Data Index: 1158778 (C4 (Web Text))

Max Activating Token Index: 881

Text #10

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.9140. Min Act: -0.1700

Data Index: 18737 (C4 (Web Text))

Max Activating Token Index: 972

Text #11

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8827. Min Act: -0.1700

Data Index: 789751 (C4 (Web Text))

Max Activating Token Index: 858

Text #12

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8827. Min Act: -0.1700

Data Index: 75108 (C4 (Web Text))

Max Activating Token Index: 900

Text #13

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8827. Min Act: -0.1700

Data Index: 302858 (C4 (Web Text))

Max Activating Token Index: 981

Text #14

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8827. Min Act: -0.1700

Data Index: 172461 (C4 (Web Text))

Max Activating Token Index: 687

Text #15

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8827. Min Act: -0.1700

Data Index: 559641 (C4 (Web Text))

Max Activating Token Index: 947

Text #16

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8827. Min Act: -0.1700

Data Index: 1321406 (C4 (Web Text))

Max Activating Token Index: 151

Text #17

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8827. Min Act: -0.1700

Data Index: 20922 (C4 (Web Text))

Max Activating Token Index: 974

Text #18

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8515. Min Act: -0.1700

Data Index: 672102 (C4 (Web Text))

Max Activating Token Index: 72

Text #19

Max Range: 4.6642. Min Range: -4.6642

Max Act: 3.8515. Min Act: -0.1700

Data Index: 231353 (C4 (Web Text))

Max Activating Token Index: 135