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

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

Neuron 1058 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: 3.4636. Min Range: -3.4636

Max Act: 3.4636. Min Act: -0.1700

Data Index: 473896 (C4 (Web Text))

Max Activating Token Index: 643

Text #1

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.4165. Min Act: -0.1700

Data Index: 577619 (C4 (Web Text))

Max Activating Token Index: 661

Text #2

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3852. Min Act: -0.1700

Data Index: 846932 (C4 (Web Text))

Max Activating Token Index: 283

Text #3

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3852. Min Act: -0.1700

Data Index: 521182 (C4 (Web Text))

Max Activating Token Index: 506

Text #4

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3852. Min Act: -0.1700

Data Index: 304364 (C4 (Web Text))

Max Activating Token Index: 266

Text #5

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3695. Min Act: -0.1700

Data Index: 280548 (C4 (Web Text))

Max Activating Token Index: 999

Text #6

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3695. Min Act: -0.1700

Data Index: 883218 (C4 (Web Text))

Max Activating Token Index: 7

Text #7

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3224. Min Act: -0.1700

Data Index: 1633457 (Python Code)

Max Activating Token Index: 591

Text #8

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3852. Min Act: -0.1700

Data Index: 1094036 (C4 (Web Text))

Max Activating Token Index: 761

Text #9

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.2753. Min Act: -0.1700

Data Index: 76149 (C4 (Web Text))

Max Activating Token Index: 771

Text #10

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3224. Min Act: -0.1700

Data Index: 1428401 (Python Code)

Max Activating Token Index: 934

Text #11

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3695. Min Act: -0.1700

Data Index: 390789 (C4 (Web Text))

Max Activating Token Index: 844

Text #12

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.2753. Min Act: -0.1700

Data Index: 380936 (C4 (Web Text))

Max Activating Token Index: 616

Text #13

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3224. Min Act: -0.1700

Data Index: 894077 (C4 (Web Text))

Max Activating Token Index: 586

Text #14

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.2753. Min Act: -0.1700

Data Index: 89451 (C4 (Web Text))

Max Activating Token Index: 860

Text #15

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3381. Min Act: -0.1700

Data Index: 755040 (C4 (Web Text))

Max Activating Token Index: 644

Text #16

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.3067. Min Act: -0.1700

Data Index: 114242 (C4 (Web Text))

Max Activating Token Index: 461

Text #17

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.2910. Min Act: -0.1700

Data Index: 1267470 (C4 (Web Text))

Max Activating Token Index: 901

Text #18

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.2596. Min Act: -0.1700

Data Index: 127412 (C4 (Web Text))

Max Activating Token Index: 332

Text #19

Max Range: 3.4636. Min Range: -3.4636

Max Act: 3.2438. Min Act: -0.1700

Data Index: 309928 (C4 (Web Text))

Max Activating Token Index: 204