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

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

Neuron 802 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: 3.3581. Min Range: -3.3581

Max Act: 3.3581. Min Act: -0.1700

Data Index: 251820 (C4 (Web Text))

Max Activating Token Index: 950

Text #1

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.3267. Min Act: -0.1700

Data Index: 1030946 (C4 (Web Text))

Max Activating Token Index: 680

Text #2

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.2953. Min Act: -0.1700

Data Index: 1039101 (C4 (Web Text))

Max Activating Token Index: 956

Text #3

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.2639. Min Act: -0.1700

Data Index: 1107967 (C4 (Web Text))

Max Activating Token Index: 615

Text #4

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.2325. Min Act: -0.1700

Data Index: 963730 (C4 (Web Text))

Max Activating Token Index: 688

Text #5

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.1695. Min Act: -0.1700

Data Index: 613116 (C4 (Web Text))

Max Activating Token Index: 480

Text #6

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.2010. Min Act: -0.1700

Data Index: 1123012 (C4 (Web Text))

Max Activating Token Index: 748

Text #7

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.2010. Min Act: -0.1700

Data Index: 750744 (C4 (Web Text))

Max Activating Token Index: 162

Text #8

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.2010. Min Act: -0.1699

Data Index: 554926 (C4 (Web Text))

Max Activating Token Index: 142

Text #9

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.1695. Min Act: -0.1700

Data Index: 1066727 (C4 (Web Text))

Max Activating Token Index: 1016

Text #10

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.1065. Min Act: -0.1700

Data Index: 1049784 (C4 (Web Text))

Max Activating Token Index: 207

Text #11

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.1380. Min Act: -0.1700

Data Index: 505975 (C4 (Web Text))

Max Activating Token Index: 387

Text #12

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.1380. Min Act: -0.1700

Data Index: 964625 (C4 (Web Text))

Max Activating Token Index: 912

Text #13

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.1065. Min Act: -0.1700

Data Index: 170549 (C4 (Web Text))

Max Activating Token Index: 442

Text #14

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.1065. Min Act: -0.1700

Data Index: 699443 (C4 (Web Text))

Max Activating Token Index: 178

Text #15

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.1380. Min Act: -0.1700

Data Index: 493270 (C4 (Web Text))

Max Activating Token Index: 946

Text #16

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.1065. Min Act: -0.1700

Data Index: 433962 (C4 (Web Text))

Max Activating Token Index: 358

Text #17

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.0750. Min Act: -0.1700

Data Index: 40977 (C4 (Web Text))

Max Activating Token Index: 205

Text #18

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.0434. Min Act: -0.1700

Data Index: 1045945 (C4 (Web Text))

Max Activating Token Index: 787

Text #19

Max Range: 3.3581. Min Range: -3.3581

Max Act: 3.0750. Min Act: -0.1700

Data Index: 723485 (C4 (Web Text))

Max Activating Token Index: 196