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

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

Neuron 1877 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.3196. Min Act: -0.1700

Data Index: 985815 (C4 (Web Text))

Max Activating Token Index: 111

Text #1

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.2258. Min Act: -0.1700

Data Index: 399744 (C4 (Web Text))

Max Activating Token Index: 651

Text #2

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.1008. Min Act: -0.1700

Data Index: 1424130 (Python Code)

Max Activating Token Index: 58

Text #3

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.1321. Min Act: -0.1700

Data Index: 1308229 (C4 (Web Text))

Max Activating Token Index: 668

Text #4

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.1946. Min Act: -0.1700

Data Index: 1035848 (C4 (Web Text))

Max Activating Token Index: 978

Text #5

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.1008. Min Act: -0.1700

Data Index: 381503 (C4 (Web Text))

Max Activating Token Index: 855

Text #6

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.1321. Min Act: -0.1700

Data Index: 1017078 (C4 (Web Text))

Max Activating Token Index: 439

Text #7

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0696. Min Act: -0.1700

Data Index: 1072318 (C4 (Web Text))

Max Activating Token Index: 751

Text #8

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0071. Min Act: -0.1700

Data Index: 1303196 (C4 (Web Text))

Max Activating Token Index: 221

Text #9

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0696. Min Act: -0.1700

Data Index: 794780 (C4 (Web Text))

Max Activating Token Index: 911

Text #10

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0383. Min Act: -0.1700

Data Index: 797235 (C4 (Web Text))

Max Activating Token Index: 884

Text #11

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0383. Min Act: -0.1700

Data Index: 873714 (C4 (Web Text))

Max Activating Token Index: 751

Text #12

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0071. Min Act: -0.1700

Data Index: 687802 (C4 (Web Text))

Max Activating Token Index: 334

Text #13

Max Range: 5.3196. Min Range: -5.3196

Max Act: 4.9758. Min Act: -0.1700

Data Index: 1122159 (C4 (Web Text))

Max Activating Token Index: 120

Text #14

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0071. Min Act: -0.1700

Data Index: 438250 (C4 (Web Text))

Max Activating Token Index: 613

Text #15

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0071. Min Act: -0.1700

Data Index: 410449 (C4 (Web Text))

Max Activating Token Index: 510

Text #16

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0383. Min Act: -0.1700

Data Index: 343635 (C4 (Web Text))

Max Activating Token Index: 871

Text #17

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0071. Min Act: -0.1700

Data Index: 795731 (C4 (Web Text))

Max Activating Token Index: 810

Text #18

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0383. Min Act: -0.1700

Data Index: 383269 (C4 (Web Text))

Max Activating Token Index: 212

Text #19

Max Range: 5.3196. Min Range: -5.3196

Max Act: 5.0071. Min Act: -0.1700

Data Index: 1143852 (C4 (Web Text))

Max Activating Token Index: 894