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

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

Neuron 738 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: 3.8445. Min Range: -3.8445

Max Act: 3.8445. Min Act: -0.1700

Data Index: 938428 (C4 (Web Text))

Max Activating Token Index: 577

Text #1

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.7193. Min Act: -0.1700

Data Index: 26500 (C4 (Web Text))

Max Activating Token Index: 74

Text #2

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.7193. Min Act: -0.1700

Data Index: 1062039 (C4 (Web Text))

Max Activating Token Index: 398

Text #3

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.7506. Min Act: -0.1700

Data Index: 378948 (C4 (Web Text))

Max Activating Token Index: 120

Text #4

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.6880. Min Act: -0.1700

Data Index: 1120199 (C4 (Web Text))

Max Activating Token Index: 878

Text #5

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.6880. Min Act: -0.1700

Data Index: 178209 (C4 (Web Text))

Max Activating Token Index: 19

Text #6

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.6567. Min Act: -0.1700

Data Index: 1092769 (C4 (Web Text))

Max Activating Token Index: 69

Text #7

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.6254. Min Act: -0.1700

Data Index: 106578 (C4 (Web Text))

Max Activating Token Index: 794

Text #8

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.6254. Min Act: -0.1700

Data Index: 96355 (C4 (Web Text))

Max Activating Token Index: 243

Text #9

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.5315. Min Act: -0.1700

Data Index: 261075 (C4 (Web Text))

Max Activating Token Index: 451

Text #10

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.5315. Min Act: -0.1700

Data Index: 1052303 (C4 (Web Text))

Max Activating Token Index: 835

Text #11

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.5628. Min Act: -0.1700

Data Index: 414234 (C4 (Web Text))

Max Activating Token Index: 43

Text #12

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.4688. Min Act: -0.1700

Data Index: 339037 (C4 (Web Text))

Max Activating Token Index: 869

Text #13

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.4688. Min Act: -0.1700

Data Index: 630250 (C4 (Web Text))

Max Activating Token Index: 86

Text #14

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.4688. Min Act: -0.1700

Data Index: 210670 (C4 (Web Text))

Max Activating Token Index: 267

Text #15

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.5001. Min Act: -0.1700

Data Index: 1167850 (C4 (Web Text))

Max Activating Token Index: 48

Text #16

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.5001. Min Act: -0.1700

Data Index: 1217159 (C4 (Web Text))

Max Activating Token Index: 133

Text #17

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.4061. Min Act: -0.1700

Data Index: 1057851 (C4 (Web Text))

Max Activating Token Index: 9

Text #18

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.4374. Min Act: -0.1700

Data Index: 69765 (C4 (Web Text))

Max Activating Token Index: 109

Text #19

Max Range: 3.8445. Min Range: -3.8445

Max Act: 3.4374. Min Act: -0.1700

Data Index: 645531 (C4 (Web Text))

Max Activating Token Index: 799