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

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

Neuron 1307 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3264. Min Range: -5.3264

Max Act: 5.3264. Min Act: -0.1700

Data Index: 1124530 (C4 (Web Text))

Max Activating Token Index: 277

Text #1

Max Range: 5.3264. Min Range: -5.3264

Max Act: 5.1702. Min Act: -0.1700

Data Index: 227746 (C4 (Web Text))

Max Activating Token Index: 906

Text #2

Max Range: 5.3264. Min Range: -5.3264

Max Act: 5.0764. Min Act: -0.1700

Data Index: 161500 (C4 (Web Text))

Max Activating Token Index: 629

Text #3

Max Range: 5.3264. Min Range: -5.3264

Max Act: 5.0452. Min Act: -0.1700

Data Index: 489405 (C4 (Web Text))

Max Activating Token Index: 376

Text #4

Max Range: 5.3264. Min Range: -5.3264

Max Act: 5.1077. Min Act: -0.1700

Data Index: 1059149 (C4 (Web Text))

Max Activating Token Index: 538

Text #5

Max Range: 5.3264. Min Range: -5.3264

Max Act: 5.0139. Min Act: -0.1700

Data Index: 531542 (C4 (Web Text))

Max Activating Token Index: 16

Text #6

Max Range: 5.3264. Min Range: -5.3264

Max Act: 5.0452. Min Act: -0.1700

Data Index: 1253478 (C4 (Web Text))

Max Activating Token Index: 550

Text #7

Max Range: 5.3264. Min Range: -5.3264

Max Act: 5.0139. Min Act: -0.1700

Data Index: 1048007 (C4 (Web Text))

Max Activating Token Index: 193

Text #8

Max Range: 5.3264. Min Range: -5.3264

Max Act: 5.0139. Min Act: -0.1700

Data Index: 838298 (C4 (Web Text))

Max Activating Token Index: 39

Text #9

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.8889. Min Act: -0.1700

Data Index: 261106 (C4 (Web Text))

Max Activating Token Index: 168

Text #10

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.9202. Min Act: -0.1700

Data Index: 90385 (C4 (Web Text))

Max Activating Token Index: 746

Text #11

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.8889. Min Act: -0.1700

Data Index: 1232829 (C4 (Web Text))

Max Activating Token Index: 324

Text #12

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.8889. Min Act: -0.1700

Data Index: 1200606 (C4 (Web Text))

Max Activating Token Index: 454

Text #13

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.9202. Min Act: -0.1700

Data Index: 1256585 (C4 (Web Text))

Max Activating Token Index: 610

Text #14

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.9202. Min Act: -0.1700

Data Index: 262101 (C4 (Web Text))

Max Activating Token Index: 855

Text #15

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.7952. Min Act: -0.1700

Data Index: 40427 (C4 (Web Text))

Max Activating Token Index: 816

Text #16

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.8577. Min Act: -0.1700

Data Index: 58495 (C4 (Web Text))

Max Activating Token Index: 339

Text #17

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.9202. Min Act: -0.1700

Data Index: 1031033 (C4 (Web Text))

Max Activating Token Index: 68

Text #18

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.8264. Min Act: -0.1700

Data Index: 940842 (C4 (Web Text))

Max Activating Token Index: 549

Text #19

Max Range: 5.3264. Min Range: -5.3264

Max Act: 4.7952. Min Act: -0.1700

Data Index: 70126 (C4 (Web Text))

Max Activating Token Index: 986