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

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

Neuron 1412 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.7918. Min Act: -0.1700

Data Index: 148420 (C4 (Web Text))

Max Activating Token Index: 809

Text #1

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.7606. Min Act: -0.1700

Data Index: 1610250 (Python Code)

Max Activating Token Index: 361

Text #2

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.7293. Min Act: -0.1700

Data Index: 773571 (C4 (Web Text))

Max Activating Token Index: 181

Text #3

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6981. Min Act: -0.1700

Data Index: 1609992 (Python Code)

Max Activating Token Index: 590

Text #4

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6981. Min Act: -0.1700

Data Index: 508326 (C4 (Web Text))

Max Activating Token Index: 521

Text #5

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6668. Min Act: -0.1700

Data Index: 1408940 (Python Code)

Max Activating Token Index: 485

Text #6

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6668. Min Act: -0.1700

Data Index: 211861 (C4 (Web Text))

Max Activating Token Index: 706

Text #7

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6356. Min Act: -0.1700

Data Index: 885777 (C4 (Web Text))

Max Activating Token Index: 1004

Text #8

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6668. Min Act: -0.1700

Data Index: 840235 (C4 (Web Text))

Max Activating Token Index: 8

Text #9

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6668. Min Act: -0.1700

Data Index: 871496 (C4 (Web Text))

Max Activating Token Index: 9

Text #10

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6668. Min Act: -0.1700

Data Index: 303191 (C4 (Web Text))

Max Activating Token Index: 352

Text #11

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6668. Min Act: -0.1700

Data Index: 115001 (C4 (Web Text))

Max Activating Token Index: 200

Text #12

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6356. Min Act: -0.1700

Data Index: 1091762 (C4 (Web Text))

Max Activating Token Index: 18

Text #13

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6356. Min Act: -0.1700

Data Index: 465332 (C4 (Web Text))

Max Activating Token Index: 909

Text #14

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6043. Min Act: -0.1700

Data Index: 820216 (C4 (Web Text))

Max Activating Token Index: 973

Text #15

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6356. Min Act: -0.1700

Data Index: 1656140 (Python Code)

Max Activating Token Index: 57

Text #16

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6356. Min Act: -0.1700

Data Index: 103237 (C4 (Web Text))

Max Activating Token Index: 818

Text #17

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6043. Min Act: -0.1700

Data Index: 873263 (C4 (Web Text))

Max Activating Token Index: 47

Text #18

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6043. Min Act: -0.1700

Data Index: 171997 (C4 (Web Text))

Max Activating Token Index: 923

Text #19

Max Range: 5.7918. Min Range: -5.7918

Max Act: 5.6043. Min Act: -0.1700

Data Index: 298036 (C4 (Web Text))

Max Activating Token Index: 204