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

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

Neuron 1930 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: 3.4509. Min Range: -3.4509

Max Act: 3.4509. Min Act: -0.1700

Data Index: 1235738 (C4 (Web Text))

Max Activating Token Index: 335

Text #1

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.2311. Min Act: -0.1700

Data Index: 270038 (C4 (Web Text))

Max Activating Token Index: 403

Text #2

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1524. Min Act: -0.1700

Data Index: 1426041 (Python Code)

Max Activating Token Index: 1016

Text #3

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1682. Min Act: -0.1700

Data Index: 1511298 (Python Code)

Max Activating Token Index: 835

Text #4

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1367. Min Act: -0.1700

Data Index: 1466460 (Python Code)

Max Activating Token Index: 996

Text #5

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1682. Min Act: -0.1700

Data Index: 1569348 (Python Code)

Max Activating Token Index: 145

Text #6

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1682. Min Act: -0.1700

Data Index: 593877 (C4 (Web Text))

Max Activating Token Index: 673

Text #7

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1367. Min Act: -0.1700

Data Index: 193311 (C4 (Web Text))

Max Activating Token Index: 956

Text #8

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1209. Min Act: -0.1700

Data Index: 1453685 (Python Code)

Max Activating Token Index: 576

Text #9

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1209. Min Act: -0.1700

Data Index: 1458503 (Python Code)

Max Activating Token Index: 970

Text #10

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1367. Min Act: -0.1700

Data Index: 1387759 (Python Code)

Max Activating Token Index: 376

Text #11

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1524. Min Act: -0.1700

Data Index: 187438 (C4 (Web Text))

Max Activating Token Index: 730

Text #12

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1052. Min Act: -0.1700

Data Index: 1077917 (C4 (Web Text))

Max Activating Token Index: 750

Text #13

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.0894. Min Act: -0.1700

Data Index: 1605333 (Python Code)

Max Activating Token Index: 246

Text #14

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1052. Min Act: -0.1700

Data Index: 1352139 (C4 (Web Text))

Max Activating Token Index: 640

Text #15

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.0894. Min Act: -0.1700

Data Index: 337636 (C4 (Web Text))

Max Activating Token Index: 413

Text #16

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1052. Min Act: -0.1700

Data Index: 1298334 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.1524. Min Act: -0.1700

Data Index: 403855 (C4 (Web Text))

Max Activating Token Index: 371

Text #18

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.0579. Min Act: -0.1700

Data Index: 1303960 (C4 (Web Text))

Max Activating Token Index: 721

Text #19

Max Range: 3.4509. Min Range: -3.4509

Max Act: 3.0579. Min Act: -0.1700

Data Index: 1116848 (C4 (Web Text))

Max Activating Token Index: 672