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

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

Neuron 1445 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.1667. Min Range: -5.1667

Max Act: 5.1667. Min Act: -0.1700

Data Index: 560408 (C4 (Web Text))

Max Activating Token Index: 353

Text #1

Max Range: 5.1667. Min Range: -5.1667

Max Act: 5.0730. Min Act: -0.1700

Data Index: 621623 (C4 (Web Text))

Max Activating Token Index: 1004

Text #2

Max Range: 5.1667. Min Range: -5.1667

Max Act: 5.0730. Min Act: -0.1700

Data Index: 715326 (C4 (Web Text))

Max Activating Token Index: 978

Text #3

Max Range: 5.1667. Min Range: -5.1667

Max Act: 5.0417. Min Act: -0.1700

Data Index: 587621 (C4 (Web Text))

Max Activating Token Index: 204

Text #4

Max Range: 5.1667. Min Range: -5.1667

Max Act: 5.0417. Min Act: -0.1700

Data Index: 1208580 (C4 (Web Text))

Max Activating Token Index: 640

Text #5

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.9480. Min Act: -0.1700

Data Index: 1333582 (C4 (Web Text))

Max Activating Token Index: 44

Text #6

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.8855. Min Act: -0.1700

Data Index: 478085 (C4 (Web Text))

Max Activating Token Index: 706

Text #7

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.8542. Min Act: -0.1700

Data Index: 1011070 (C4 (Web Text))

Max Activating Token Index: 413

Text #8

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.8230. Min Act: -0.1700

Data Index: 1121176 (C4 (Web Text))

Max Activating Token Index: 287

Text #9

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.9167. Min Act: -0.1700

Data Index: 1182258 (C4 (Web Text))

Max Activating Token Index: 816

Text #10

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.8542. Min Act: -0.1700

Data Index: 458647 (C4 (Web Text))

Max Activating Token Index: 1001

Text #11

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.8230. Min Act: -0.1700

Data Index: 243371 (C4 (Web Text))

Max Activating Token Index: 294

Text #12

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.8230. Min Act: -0.1700

Data Index: 371848 (C4 (Web Text))

Max Activating Token Index: 284

Text #13

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.7917. Min Act: -0.1700

Data Index: 938759 (C4 (Web Text))

Max Activating Token Index: 286

Text #14

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.8230. Min Act: -0.1700

Data Index: 1187728 (C4 (Web Text))

Max Activating Token Index: 119

Text #15

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.8230. Min Act: -0.1700

Data Index: 360091 (C4 (Web Text))

Max Activating Token Index: 824

Text #16

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.7605. Min Act: -0.1700

Data Index: 807548 (C4 (Web Text))

Max Activating Token Index: 409

Text #17

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.7605. Min Act: -0.1700

Data Index: 714772 (C4 (Web Text))

Max Activating Token Index: 804

Text #18

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.7292. Min Act: -0.1700

Data Index: 549127 (C4 (Web Text))

Max Activating Token Index: 767

Text #19

Max Range: 5.1667. Min Range: -5.1667

Max Act: 4.6980. Min Act: -0.1700

Data Index: 407496 (C4 (Web Text))

Max Activating Token Index: 367