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

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

Neuron 383 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: 4.9544. Min Range: -4.9544

Max Act: 4.9544. Min Act: -0.1696

Data Index: 578283 (C4 (Web Text))

Max Activating Token Index: 655

Text #1

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.9388. Min Act: -0.1700

Data Index: 16409 (C4 (Web Text))

Max Activating Token Index: 804

Text #2

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.8294. Min Act: -0.1699

Data Index: 1217948 (C4 (Web Text))

Max Activating Token Index: 30

Text #3

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.8450. Min Act: -0.1662

Data Index: 425082 (C4 (Web Text))

Max Activating Token Index: 545

Text #4

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7981. Min Act: -0.1699

Data Index: 70913 (C4 (Web Text))

Max Activating Token Index: 133

Text #5

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7981. Min Act: -0.1700

Data Index: 1042141 (C4 (Web Text))

Max Activating Token Index: 260

Text #6

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7669. Min Act: -0.1699

Data Index: 28459 (C4 (Web Text))

Max Activating Token Index: 739

Text #7

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.6888. Min Act: -0.1690

Data Index: 486130 (C4 (Web Text))

Max Activating Token Index: 484

Text #8

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7669. Min Act: -0.1685

Data Index: 158572 (C4 (Web Text))

Max Activating Token Index: 689

Text #9

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7356. Min Act: -0.1700

Data Index: 455946 (C4 (Web Text))

Max Activating Token Index: 894

Text #10

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7513. Min Act: -0.1654

Data Index: 534967 (C4 (Web Text))

Max Activating Token Index: 129

Text #11

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7044. Min Act: -0.1696

Data Index: 488265 (C4 (Web Text))

Max Activating Token Index: 200

Text #12

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7356. Min Act: -0.1699

Data Index: 208909 (C4 (Web Text))

Max Activating Token Index: 897

Text #13

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7200. Min Act: -0.1597

Data Index: 845443 (C4 (Web Text))

Max Activating Token Index: 927

Text #14

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7200. Min Act: -0.1699

Data Index: 340210 (C4 (Web Text))

Max Activating Token Index: 936

Text #15

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7981. Min Act: -0.1699

Data Index: 497343 (C4 (Web Text))

Max Activating Token Index: 823

Text #16

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7200. Min Act: -0.1699

Data Index: 925773 (C4 (Web Text))

Max Activating Token Index: 602

Text #17

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.7044. Min Act: -0.1700

Data Index: 839463 (C4 (Web Text))

Max Activating Token Index: 26

Text #18

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.6731. Min Act: -0.1698

Data Index: 497931 (C4 (Web Text))

Max Activating Token Index: 951

Text #19

Max Range: 4.9544. Min Range: -4.9544

Max Act: 4.6888. Min Act: -0.1700

Data Index: 1230921 (C4 (Web Text))

Max Activating Token Index: 668