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

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

Neuron 1887 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: 3.8229. Min Range: -3.8229

Max Act: 3.8229. Min Act: -0.1700

Data Index: 568900 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.7916. Min Act: -0.1700

Data Index: 580655 (C4 (Web Text))

Max Activating Token Index: 921

Text #2

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.7603. Min Act: -0.1700

Data Index: 1032622 (C4 (Web Text))

Max Activating Token Index: 749

Text #3

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.7290. Min Act: -0.1700

Data Index: 696728 (C4 (Web Text))

Max Activating Token Index: 658

Text #4

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6977. Min Act: -0.1700

Data Index: 1029809 (C4 (Web Text))

Max Activating Token Index: 979

Text #5

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.7290. Min Act: -0.1700

Data Index: 940946 (C4 (Web Text))

Max Activating Token Index: 945

Text #6

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6977. Min Act: -0.1700

Data Index: 931592 (C4 (Web Text))

Max Activating Token Index: 943

Text #7

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6977. Min Act: -0.1700

Data Index: 563871 (C4 (Web Text))

Max Activating Token Index: 740

Text #8

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6195. Min Act: -0.1700

Data Index: 709114 (C4 (Web Text))

Max Activating Token Index: 836

Text #9

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6351. Min Act: -0.1700

Data Index: 450485 (C4 (Web Text))

Max Activating Token Index: 971

Text #10

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6351. Min Act: -0.1700

Data Index: 317367 (C4 (Web Text))

Max Activating Token Index: 740

Text #11

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6038. Min Act: -0.1700

Data Index: 356381 (C4 (Web Text))

Max Activating Token Index: 870

Text #12

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6195. Min Act: -0.1700

Data Index: 925616 (C4 (Web Text))

Max Activating Token Index: 580

Text #13

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6195. Min Act: -0.1700

Data Index: 1065582 (C4 (Web Text))

Max Activating Token Index: 653

Text #14

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.5725. Min Act: -0.1700

Data Index: 1056832 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6664. Min Act: -0.1700

Data Index: 1036728 (C4 (Web Text))

Max Activating Token Index: 771

Text #16

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6351. Min Act: -0.1700

Data Index: 391480 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6664. Min Act: -0.1700

Data Index: 898579 (C4 (Web Text))

Max Activating Token Index: 499

Text #18

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.5725. Min Act: -0.1700

Data Index: 1249656 (C4 (Web Text))

Max Activating Token Index: 750

Text #19

Max Range: 3.8229. Min Range: -3.8229

Max Act: 3.6351. Min Act: -0.1700

Data Index: 466818 (C4 (Web Text))

Max Activating Token Index: 828