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

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

Neuron 2031 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: 3.4941. Min Range: -3.4941

Max Act: 3.4941. Min Act: -0.1700

Data Index: 125794 (C4 (Web Text))

Max Activating Token Index: 60

Text #1

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.4941. Min Act: -0.1700

Data Index: 140836 (C4 (Web Text))

Max Activating Token Index: 505

Text #2

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3843. Min Act: -0.1700

Data Index: 725473 (C4 (Web Text))

Max Activating Token Index: 656

Text #3

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3843. Min Act: -0.1700

Data Index: 388072 (C4 (Web Text))

Max Activating Token Index: 513

Text #4

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3529. Min Act: -0.1700

Data Index: 1178027 (C4 (Web Text))

Max Activating Token Index: 594

Text #5

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3686. Min Act: -0.1700

Data Index: 1175863 (C4 (Web Text))

Max Activating Token Index: 175

Text #6

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3215. Min Act: -0.1700

Data Index: 1077201 (C4 (Web Text))

Max Activating Token Index: 736

Text #7

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3529. Min Act: -0.1700

Data Index: 1099404 (C4 (Web Text))

Max Activating Token Index: 321

Text #8

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3529. Min Act: -0.1700

Data Index: 19652 (C4 (Web Text))

Max Activating Token Index: 491

Text #9

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3372. Min Act: -0.1700

Data Index: 1326660 (C4 (Web Text))

Max Activating Token Index: 28

Text #10

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3215. Min Act: -0.1700

Data Index: 981544 (C4 (Web Text))

Max Activating Token Index: 800

Text #11

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3215. Min Act: -0.1700

Data Index: 1295624 (C4 (Web Text))

Max Activating Token Index: 24

Text #12

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3058. Min Act: -0.1700

Data Index: 667448 (C4 (Web Text))

Max Activating Token Index: 188

Text #13

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3215. Min Act: -0.1700

Data Index: 1216633 (C4 (Web Text))

Max Activating Token Index: 18

Text #14

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.3058. Min Act: -0.1700

Data Index: 1163399 (C4 (Web Text))

Max Activating Token Index: 736

Text #15

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.2901. Min Act: -0.1700

Data Index: 892198 (C4 (Web Text))

Max Activating Token Index: 417

Text #16

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.2901. Min Act: -0.1700

Data Index: 1342054 (C4 (Web Text))

Max Activating Token Index: 351

Text #17

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.2744. Min Act: -0.1700

Data Index: 872152 (C4 (Web Text))

Max Activating Token Index: 432

Text #18

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.2744. Min Act: -0.1700

Data Index: 1167002 (C4 (Web Text))

Max Activating Token Index: 650

Text #19

Max Range: 3.4941. Min Range: -3.4941

Max Act: 3.2901. Min Act: -0.1700

Data Index: 647263 (C4 (Web Text))

Max Activating Token Index: 475