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

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

Neuron 1442 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: 5.4323. Min Range: -5.4323

Max Act: 5.4323. Min Act: -0.1700

Data Index: 684529 (C4 (Web Text))

Max Activating Token Index: 981

Text #1

Max Range: 5.4323. Min Range: -5.4323

Max Act: 5.1511. Min Act: -0.1700

Data Index: 1125603 (C4 (Web Text))

Max Activating Token Index: 339

Text #2

Max Range: 5.4323. Min Range: -5.4323

Max Act: 5.2136. Min Act: -0.1700

Data Index: 680163 (C4 (Web Text))

Max Activating Token Index: 861

Text #3

Max Range: 5.4323. Min Range: -5.4323

Max Act: 5.1511. Min Act: -0.1700

Data Index: 881370 (C4 (Web Text))

Max Activating Token Index: 56

Text #4

Max Range: 5.4323. Min Range: -5.4323

Max Act: 5.1198. Min Act: -0.1700

Data Index: 1117495 (C4 (Web Text))

Max Activating Token Index: 493

Text #5

Max Range: 5.4323. Min Range: -5.4323

Max Act: 5.1198. Min Act: -0.1700

Data Index: 1000199 (C4 (Web Text))

Max Activating Token Index: 423

Text #6

Max Range: 5.4323. Min Range: -5.4323

Max Act: 5.0573. Min Act: -0.1700

Data Index: 1302110 (C4 (Web Text))

Max Activating Token Index: 465

Text #7

Max Range: 5.4323. Min Range: -5.4323

Max Act: 5.0573. Min Act: -0.1700

Data Index: 205708 (C4 (Web Text))

Max Activating Token Index: 689

Text #8

Max Range: 5.4323. Min Range: -5.4323

Max Act: 5.0573. Min Act: -0.1700

Data Index: 1038714 (C4 (Web Text))

Max Activating Token Index: 416

Text #9

Max Range: 5.4323. Min Range: -5.4323

Max Act: 5.0886. Min Act: -0.1700

Data Index: 799794 (C4 (Web Text))

Max Activating Token Index: 684

Text #10

Max Range: 5.4323. Min Range: -5.4323

Max Act: 4.9948. Min Act: -0.1700

Data Index: 874770 (C4 (Web Text))

Max Activating Token Index: 64

Text #11

Max Range: 5.4323. Min Range: -5.4323

Max Act: 4.9948. Min Act: -0.1700

Data Index: 1310078 (C4 (Web Text))

Max Activating Token Index: 504

Text #12

Max Range: 5.4323. Min Range: -5.4323

Max Act: 5.0261. Min Act: -0.1700

Data Index: 731892 (C4 (Web Text))

Max Activating Token Index: 915

Text #13

Max Range: 5.4323. Min Range: -5.4323

Max Act: 4.9948. Min Act: -0.1700

Data Index: 819523 (C4 (Web Text))

Max Activating Token Index: 460

Text #14

Max Range: 5.4323. Min Range: -5.4323

Max Act: 4.9636. Min Act: -0.1700

Data Index: 35914 (C4 (Web Text))

Max Activating Token Index: 54

Text #15

Max Range: 5.4323. Min Range: -5.4323

Max Act: 4.9948. Min Act: -0.1700

Data Index: 29292 (C4 (Web Text))

Max Activating Token Index: 821

Text #16

Max Range: 5.4323. Min Range: -5.4323

Max Act: 4.9948. Min Act: -0.1700

Data Index: 994155 (C4 (Web Text))

Max Activating Token Index: 923

Text #17

Max Range: 5.4323. Min Range: -5.4323

Max Act: 4.9636. Min Act: -0.1700

Data Index: 776173 (C4 (Web Text))

Max Activating Token Index: 377

Text #18

Max Range: 5.4323. Min Range: -5.4323

Max Act: 4.9636. Min Act: -0.1700

Data Index: 1168291 (C4 (Web Text))

Max Activating Token Index: 26

Text #19

Max Range: 5.4323. Min Range: -5.4323

Max Act: 4.9323. Min Act: -0.1700

Data Index: 1186460 (C4 (Web Text))

Max Activating Token Index: 565