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

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

Neuron 1599 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.4523. Min Act: -0.1700

Data Index: 1006591 (C4 (Web Text))

Max Activating Token Index: 138

Text #1

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.4209. Min Act: -0.1700

Data Index: 1002346 (C4 (Web Text))

Max Activating Token Index: 591

Text #2

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.3895. Min Act: -0.1700

Data Index: 232586 (C4 (Web Text))

Max Activating Token Index: 725

Text #3

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.3582. Min Act: -0.1700

Data Index: 863615 (C4 (Web Text))

Max Activating Token Index: 955

Text #4

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.3582. Min Act: -0.1700

Data Index: 1305161 (C4 (Web Text))

Max Activating Token Index: 732

Text #5

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.3268. Min Act: -0.1700

Data Index: 990941 (C4 (Web Text))

Max Activating Token Index: 835

Text #6

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2954. Min Act: -0.1700

Data Index: 1312299 (C4 (Web Text))

Max Activating Token Index: 675

Text #7

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2954. Min Act: -0.1700

Data Index: 1299258 (C4 (Web Text))

Max Activating Token Index: 49

Text #8

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.3268. Min Act: -0.1700

Data Index: 1304736 (C4 (Web Text))

Max Activating Token Index: 608

Text #9

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.3268. Min Act: -0.1700

Data Index: 103329 (C4 (Web Text))

Max Activating Token Index: 688

Text #10

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2954. Min Act: -0.1700

Data Index: 114833 (C4 (Web Text))

Max Activating Token Index: 87

Text #11

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2954. Min Act: -0.1700

Data Index: 1196145 (C4 (Web Text))

Max Activating Token Index: 636

Text #12

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2954. Min Act: -0.1700

Data Index: 717425 (C4 (Web Text))

Max Activating Token Index: 127

Text #13

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.3268. Min Act: -0.1700

Data Index: 216400 (C4 (Web Text))

Max Activating Token Index: 280

Text #14

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2954. Min Act: -0.1700

Data Index: 831002 (C4 (Web Text))

Max Activating Token Index: 618

Text #15

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2325. Min Act: -0.1700

Data Index: 309339 (C4 (Web Text))

Max Activating Token Index: 99

Text #16

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2639. Min Act: -0.1700

Data Index: 793705 (C4 (Web Text))

Max Activating Token Index: 869

Text #17

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2325. Min Act: -0.1700

Data Index: 1130039 (C4 (Web Text))

Max Activating Token Index: 215

Text #18

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2639. Min Act: -0.1700

Data Index: 50868 (C4 (Web Text))

Max Activating Token Index: 250

Text #19

Max Range: 3.4523. Min Range: -3.4523

Max Act: 3.2325. Min Act: -0.1700

Data Index: 613295 (C4 (Web Text))

Max Activating Token Index: 605