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

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

Neuron 1991 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: 7.4339. Min Range: -7.4339

Max Act: 7.4339. Min Act: -0.1700

Data Index: 44242 (C4 (Web Text))

Max Activating Token Index: 866

Text #1

Max Range: 7.4339. Min Range: -7.4339

Max Act: 7.2152. Min Act: -0.1700

Data Index: 187920 (C4 (Web Text))

Max Activating Token Index: 461

Text #2

Max Range: 7.4339. Min Range: -7.4339

Max Act: 7.1527. Min Act: -0.1700

Data Index: 365552 (C4 (Web Text))

Max Activating Token Index: 311

Text #3

Max Range: 7.4339. Min Range: -7.4339

Max Act: 7.0902. Min Act: -0.1700

Data Index: 1183685 (C4 (Web Text))

Max Activating Token Index: 210

Text #4

Max Range: 7.4339. Min Range: -7.4339

Max Act: 7.1214. Min Act: -0.1700

Data Index: 539374 (C4 (Web Text))

Max Activating Token Index: 801

Text #5

Max Range: 7.4339. Min Range: -7.4339

Max Act: 7.0589. Min Act: -0.1700

Data Index: 902744 (C4 (Web Text))

Max Activating Token Index: 49

Text #6

Max Range: 7.4339. Min Range: -7.4339

Max Act: 7.0589. Min Act: -0.1700

Data Index: 630319 (C4 (Web Text))

Max Activating Token Index: 76

Text #7

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.9339. Min Act: -0.1700

Data Index: 1059781 (C4 (Web Text))

Max Activating Token Index: 743

Text #8

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.8714. Min Act: -0.1700

Data Index: 184786 (C4 (Web Text))

Max Activating Token Index: 897

Text #9

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.8089. Min Act: -0.1700

Data Index: 1182067 (C4 (Web Text))

Max Activating Token Index: 405

Text #10

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.7777. Min Act: -0.1700

Data Index: 1267390 (C4 (Web Text))

Max Activating Token Index: 283

Text #11

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.7777. Min Act: -0.1700

Data Index: 562522 (C4 (Web Text))

Max Activating Token Index: 670

Text #12

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.6839. Min Act: -0.1700

Data Index: 1272060 (C4 (Web Text))

Max Activating Token Index: 11

Text #13

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.6839. Min Act: -0.1700

Data Index: 972744 (C4 (Web Text))

Max Activating Token Index: 946

Text #14

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.6527. Min Act: -0.1700

Data Index: 843789 (C4 (Web Text))

Max Activating Token Index: 935

Text #15

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.7152. Min Act: -0.1700

Data Index: 425886 (C4 (Web Text))

Max Activating Token Index: 780

Text #16

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.7152. Min Act: -0.1700

Data Index: 286637 (C4 (Web Text))

Max Activating Token Index: 852

Text #17

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.6527. Min Act: -0.1700

Data Index: 972371 (C4 (Web Text))

Max Activating Token Index: 34

Text #18

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.6527. Min Act: -0.1700

Data Index: 1314381 (C4 (Web Text))

Max Activating Token Index: 62

Text #19

Max Range: 7.4339. Min Range: -7.4339

Max Act: 6.6839. Min Act: -0.1700

Data Index: 295 (C4 (Web Text))

Max Activating Token Index: 166