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

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

Neuron 397 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: 5.7498. Min Range: -5.7498

Max Act: 5.7498. Min Act: -0.1699

Data Index: 172345 (C4 (Web Text))

Max Activating Token Index: 920

Text #1

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.4998. Min Act: -0.1700

Data Index: 566721 (C4 (Web Text))

Max Activating Token Index: 799

Text #2

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.3748. Min Act: -0.1700

Data Index: 84914 (C4 (Web Text))

Max Activating Token Index: 179

Text #3

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.3436. Min Act: -0.1700

Data Index: 752338 (C4 (Web Text))

Max Activating Token Index: 124

Text #4

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.2498. Min Act: -0.1700

Data Index: 497234 (C4 (Web Text))

Max Activating Token Index: 956

Text #5

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.1873. Min Act: -0.1700

Data Index: 209058 (C4 (Web Text))

Max Activating Token Index: 1009

Text #6

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.2186. Min Act: -0.1700

Data Index: 958020 (C4 (Web Text))

Max Activating Token Index: 359

Text #7

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.1873. Min Act: -0.1700

Data Index: 1263935 (C4 (Web Text))

Max Activating Token Index: 218

Text #8

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.0936. Min Act: -0.1700

Data Index: 319336 (C4 (Web Text))

Max Activating Token Index: 600

Text #9

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.0623. Min Act: -0.1700

Data Index: 1042841 (C4 (Web Text))

Max Activating Token Index: 194

Text #10

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.0623. Min Act: -0.1700

Data Index: 626135 (C4 (Web Text))

Max Activating Token Index: 784

Text #11

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.0311. Min Act: -0.1696

Data Index: 219667 (C4 (Web Text))

Max Activating Token Index: 960

Text #12

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.0311. Min Act: -0.1700

Data Index: 938613 (C4 (Web Text))

Max Activating Token Index: 691

Text #13

Max Range: 5.7498. Min Range: -5.7498

Max Act: 4.9998. Min Act: -0.1700

Data Index: 30475 (C4 (Web Text))

Max Activating Token Index: 162

Text #14

Max Range: 5.7498. Min Range: -5.7498

Max Act: 5.0623. Min Act: -0.1700

Data Index: 909773 (C4 (Web Text))

Max Activating Token Index: 120

Text #15

Max Range: 5.7498. Min Range: -5.7498

Max Act: 4.9686. Min Act: -0.1700

Data Index: 1221415 (C4 (Web Text))

Max Activating Token Index: 28

Text #16

Max Range: 5.7498. Min Range: -5.7498

Max Act: 4.9373. Min Act: -0.1700

Data Index: 1064781 (C4 (Web Text))

Max Activating Token Index: 996

Text #17

Max Range: 5.7498. Min Range: -5.7498

Max Act: 4.9686. Min Act: -0.1700

Data Index: 675738 (C4 (Web Text))

Max Activating Token Index: 304

Text #18

Max Range: 5.7498. Min Range: -5.7498

Max Act: 4.9061. Min Act: -0.1700

Data Index: 105508 (C4 (Web Text))

Max Activating Token Index: 213

Text #19

Max Range: 5.7498. Min Range: -5.7498

Max Act: 4.9373. Min Act: -0.1700

Data Index: 670734 (C4 (Web Text))

Max Activating Token Index: 780