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

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

Neuron 82 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.2688. Min Act: -0.1700

Data Index: 714804 (C4 (Web Text))

Max Activating Token Index: 381

Text #1

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.1587. Min Act: -0.1700

Data Index: 1221856 (C4 (Web Text))

Max Activating Token Index: 336

Text #2

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.1272. Min Act: -0.1700

Data Index: 786080 (C4 (Web Text))

Max Activating Token Index: 44

Text #3

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.1115. Min Act: -0.1700

Data Index: 779530 (C4 (Web Text))

Max Activating Token Index: 423

Text #4

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0799. Min Act: -0.1700

Data Index: 558493 (C4 (Web Text))

Max Activating Token Index: 893

Text #5

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0957. Min Act: -0.1700

Data Index: 1082045 (C4 (Web Text))

Max Activating Token Index: 73

Text #6

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0642. Min Act: -0.1700

Data Index: 1029922 (C4 (Web Text))

Max Activating Token Index: 218

Text #7

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0799. Min Act: -0.1700

Data Index: 1189066 (C4 (Web Text))

Max Activating Token Index: 416

Text #8

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0799. Min Act: -0.1700

Data Index: 1097798 (C4 (Web Text))

Max Activating Token Index: 469

Text #9

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0642. Min Act: -0.1700

Data Index: 156627 (C4 (Web Text))

Max Activating Token Index: 43

Text #10

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0484. Min Act: -0.1700

Data Index: 865980 (C4 (Web Text))

Max Activating Token Index: 972

Text #11

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0326. Min Act: -0.1700

Data Index: 1044825 (C4 (Web Text))

Max Activating Token Index: 1011

Text #12

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0484. Min Act: -0.1700

Data Index: 378655 (C4 (Web Text))

Max Activating Token Index: 894

Text #13

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0168. Min Act: -0.1700

Data Index: 798942 (C4 (Web Text))

Max Activating Token Index: 253

Text #14

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0168. Min Act: -0.1700

Data Index: 20551 (C4 (Web Text))

Max Activating Token Index: 74

Text #15

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0642. Min Act: -0.1700

Data Index: 685052 (C4 (Web Text))

Max Activating Token Index: 403

Text #16

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0168. Min Act: -0.1700

Data Index: 894947 (C4 (Web Text))

Max Activating Token Index: 682

Text #17

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0168. Min Act: -0.1700

Data Index: 1249195 (C4 (Web Text))

Max Activating Token Index: 118

Text #18

Max Range: 3.2688. Min Range: -3.2688

Max Act: 3.0168. Min Act: -0.1700

Data Index: 331809 (C4 (Web Text))

Max Activating Token Index: 754

Text #19

Max Range: 3.2688. Min Range: -3.2688

Max Act: 2.9535. Min Act: -0.1700

Data Index: 861945 (C4 (Web Text))

Max Activating Token Index: 673