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

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

Neuron 1475 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: 9.2561. Min Range: -9.2561

Max Act: 9.2561. Min Act: -0.1700

Data Index: 820284 (C4 (Web Text))

Max Activating Token Index: 655

Text #1

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.8811. Min Act: -0.1700

Data Index: 1189385 (C4 (Web Text))

Max Activating Token Index: 893

Text #2

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.7561. Min Act: -0.1700

Data Index: 483457 (C4 (Web Text))

Max Activating Token Index: 161

Text #3

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.7561. Min Act: -0.1700

Data Index: 372285 (C4 (Web Text))

Max Activating Token Index: 811

Text #4

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.7873. Min Act: -0.1700

Data Index: 410430 (C4 (Web Text))

Max Activating Token Index: 535

Text #5

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.6623. Min Act: -0.1700

Data Index: 908077 (C4 (Web Text))

Max Activating Token Index: 164

Text #6

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.7248. Min Act: -0.1700

Data Index: 1036843 (C4 (Web Text))

Max Activating Token Index: 733

Text #7

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.7248. Min Act: -0.1700

Data Index: 321724 (C4 (Web Text))

Max Activating Token Index: 805

Text #8

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.3498. Min Act: -0.1700

Data Index: 776562 (C4 (Web Text))

Max Activating Token Index: 582

Text #9

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.6623. Min Act: -0.1700

Data Index: 1174851 (C4 (Web Text))

Max Activating Token Index: 179

Text #10

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.4436. Min Act: -0.1700

Data Index: 233218 (C4 (Web Text))

Max Activating Token Index: 591

Text #11

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.5061. Min Act: -0.1700

Data Index: 1019138 (C4 (Web Text))

Max Activating Token Index: 642

Text #12

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.3498. Min Act: -0.1700

Data Index: 828911 (C4 (Web Text))

Max Activating Token Index: 352

Text #13

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.4748. Min Act: -0.1700

Data Index: 344709 (C4 (Web Text))

Max Activating Token Index: 744

Text #14

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.3186. Min Act: -0.1700

Data Index: 415634 (C4 (Web Text))

Max Activating Token Index: 1001

Text #15

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.3498. Min Act: -0.1700

Data Index: 552911 (C4 (Web Text))

Max Activating Token Index: 839

Text #16

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.3498. Min Act: -0.1700

Data Index: 616563 (C4 (Web Text))

Max Activating Token Index: 285

Text #17

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.1623. Min Act: -0.1700

Data Index: 702094 (C4 (Web Text))

Max Activating Token Index: 722

Text #18

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.2561. Min Act: -0.1700

Data Index: 537908 (C4 (Web Text))

Max Activating Token Index: 84

Text #19

Max Range: 9.2561. Min Range: -9.2561

Max Act: 8.3186. Min Act: -0.1700

Data Index: 1157621 (C4 (Web Text))

Max Activating Token Index: 482