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

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

Neuron 1709 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: 7.3875. Min Range: -7.3875

Max Act: 7.3875. Min Act: -0.1700

Data Index: 311066 (C4 (Web Text))

Max Activating Token Index: 680

Text #1

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.3563. Min Act: -0.1700

Data Index: 470280 (C4 (Web Text))

Max Activating Token Index: 797

Text #2

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.2938. Min Act: -0.1700

Data Index: 1029537 (C4 (Web Text))

Max Activating Token Index: 570

Text #3

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.1688. Min Act: -0.1700

Data Index: 153398 (C4 (Web Text))

Max Activating Token Index: 563

Text #4

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.1688. Min Act: -0.1700

Data Index: 342026 (C4 (Web Text))

Max Activating Token Index: 127

Text #5

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.2313. Min Act: -0.1700

Data Index: 324602 (C4 (Web Text))

Max Activating Token Index: 477

Text #6

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.0750. Min Act: -0.1700

Data Index: 1328571 (C4 (Web Text))

Max Activating Token Index: 555

Text #7

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.1688. Min Act: -0.1700

Data Index: 1209949 (C4 (Web Text))

Max Activating Token Index: 935

Text #8

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.1063. Min Act: -0.1700

Data Index: 533455 (C4 (Web Text))

Max Activating Token Index: 409

Text #9

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.0438. Min Act: -0.1700

Data Index: 1143875 (C4 (Web Text))

Max Activating Token Index: 362

Text #10

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.1063. Min Act: -0.1700

Data Index: 980189 (C4 (Web Text))

Max Activating Token Index: 219

Text #11

Max Range: 7.3875. Min Range: -7.3875

Max Act: 7.0125. Min Act: -0.1700

Data Index: 389427 (C4 (Web Text))

Max Activating Token Index: 109

Text #12

Max Range: 7.3875. Min Range: -7.3875

Max Act: 6.9813. Min Act: -0.1700

Data Index: 714364 (C4 (Web Text))

Max Activating Token Index: 307

Text #13

Max Range: 7.3875. Min Range: -7.3875

Max Act: 6.7313. Min Act: -0.1700

Data Index: 601083 (C4 (Web Text))

Max Activating Token Index: 482

Text #14

Max Range: 7.3875. Min Range: -7.3875

Max Act: 6.9500. Min Act: -0.1700

Data Index: 1256848 (C4 (Web Text))

Max Activating Token Index: 478

Text #15

Max Range: 7.3875. Min Range: -7.3875

Max Act: 6.9188. Min Act: -0.1700

Data Index: 97990 (C4 (Web Text))

Max Activating Token Index: 650

Text #16

Max Range: 7.3875. Min Range: -7.3875

Max Act: 6.8875. Min Act: -0.1700

Data Index: 569123 (C4 (Web Text))

Max Activating Token Index: 339

Text #17

Max Range: 7.3875. Min Range: -7.3875

Max Act: 6.4188. Min Act: -0.1700

Data Index: 1270987 (C4 (Web Text))

Max Activating Token Index: 318

Text #18

Max Range: 7.3875. Min Range: -7.3875

Max Act: 6.2938. Min Act: -0.1700

Data Index: 1252317 (C4 (Web Text))

Max Activating Token Index: 45

Text #19

Max Range: 7.3875. Min Range: -7.3875

Max Act: 6.2625. Min Act: -0.1700

Data Index: 1308743 (C4 (Web Text))

Max Activating Token Index: 826