Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 717 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.6503. Min Act: -0.1700

Data Index: 308842 (C4 (Web Text))

Max Activating Token Index: 10

Text #1

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.5878. Min Act: -0.1700

Data Index: 618877 (C4 (Web Text))

Max Activating Token Index: 273

Text #2

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.4003. Min Act: -0.1700

Data Index: 125835 (C4 (Web Text))

Max Activating Token Index: 833

Text #3

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.4003. Min Act: -0.1700

Data Index: 970190 (C4 (Web Text))

Max Activating Token Index: 146

Text #4

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.3378. Min Act: -0.1700

Data Index: 698037 (C4 (Web Text))

Max Activating Token Index: 79

Text #5

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.4316. Min Act: -0.1700

Data Index: 835911 (C4 (Web Text))

Max Activating Token Index: 847

Text #6

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.3066. Min Act: -0.1700

Data Index: 587930 (C4 (Web Text))

Max Activating Token Index: 613

Text #7

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.3066. Min Act: -0.1700

Data Index: 1073157 (C4 (Web Text))

Max Activating Token Index: 908

Text #8

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.2753. Min Act: -0.1700

Data Index: 658632 (C4 (Web Text))

Max Activating Token Index: 262

Text #9

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.1816. Min Act: -0.1700

Data Index: 179516 (C4 (Web Text))

Max Activating Token Index: 861

Text #10

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.2441. Min Act: -0.1700

Data Index: 381782 (C4 (Web Text))

Max Activating Token Index: 184

Text #11

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.1816. Min Act: -0.1700

Data Index: 295081 (C4 (Web Text))

Max Activating Token Index: 1004

Text #12

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.1503. Min Act: -0.1700

Data Index: 513053 (C4 (Web Text))

Max Activating Token Index: 312

Text #13

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.1503. Min Act: -0.1700

Data Index: 573293 (C4 (Web Text))

Max Activating Token Index: 11

Text #14

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.1503. Min Act: -0.1700

Data Index: 887296 (C4 (Web Text))

Max Activating Token Index: 59

Text #15

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.1191. Min Act: -0.1700

Data Index: 875761 (C4 (Web Text))

Max Activating Token Index: 14

Text #16

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.1191. Min Act: -0.1700

Data Index: 1029064 (C4 (Web Text))

Max Activating Token Index: 30

Text #17

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.0878. Min Act: -0.1700

Data Index: 1027782 (C4 (Web Text))

Max Activating Token Index: 280

Text #18

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.0566. Min Act: -0.1700

Data Index: 1086133 (C4 (Web Text))

Max Activating Token Index: 280

Text #19

Max Range: 5.6503. Min Range: -5.6503

Max Act: 5.0566. Min Act: -0.1700

Data Index: 1082993 (C4 (Web Text))

Max Activating Token Index: 263