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

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

Neuron 1442 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: 8.1004. Min Range: -8.1004

Max Act: 8.1004. Min Act: -0.1700

Data Index: 122056 (C4 (Web Text))

Max Activating Token Index: 415

Text #1

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.9754. Min Act: -0.1700

Data Index: 844938 (C4 (Web Text))

Max Activating Token Index: 975

Text #2

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.8504. Min Act: -0.1700

Data Index: 564758 (C4 (Web Text))

Max Activating Token Index: 549

Text #3

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.8504. Min Act: -0.1700

Data Index: 1224840 (C4 (Web Text))

Max Activating Token Index: 811

Text #4

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.7254. Min Act: -0.1700

Data Index: 1647374 (Python Code)

Max Activating Token Index: 376

Text #5

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.7254. Min Act: -0.1700

Data Index: 1566152 (Python Code)

Max Activating Token Index: 983

Text #6

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.5379. Min Act: -0.1700

Data Index: 1091238 (C4 (Web Text))

Max Activating Token Index: 952

Text #7

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.5379. Min Act: -0.1700

Data Index: 1444786 (Python Code)

Max Activating Token Index: 410

Text #8

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.6629. Min Act: -0.1700

Data Index: 1563261 (Python Code)

Max Activating Token Index: 826

Text #9

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.4129. Min Act: -0.1700

Data Index: 382064 (C4 (Web Text))

Max Activating Token Index: 533

Text #10

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.4754. Min Act: -0.1700

Data Index: 899808 (C4 (Web Text))

Max Activating Token Index: 954

Text #11

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.4129. Min Act: -0.1700

Data Index: 1217022 (C4 (Web Text))

Max Activating Token Index: 144

Text #12

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.3817. Min Act: -0.1700

Data Index: 1522683 (Python Code)

Max Activating Token Index: 363

Text #13

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.4129. Min Act: -0.1700

Data Index: 255854 (C4 (Web Text))

Max Activating Token Index: 721

Text #14

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.3192. Min Act: -0.1700

Data Index: 1604036 (Python Code)

Max Activating Token Index: 834

Text #15

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.3504. Min Act: -0.1700

Data Index: 573681 (C4 (Web Text))

Max Activating Token Index: 202

Text #16

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.3192. Min Act: -0.1700

Data Index: 1642905 (Python Code)

Max Activating Token Index: 379

Text #17

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.3192. Min Act: -0.1700

Data Index: 664814 (C4 (Web Text))

Max Activating Token Index: 828

Text #18

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.2879. Min Act: -0.1700

Data Index: 665740 (C4 (Web Text))

Max Activating Token Index: 757

Text #19

Max Range: 8.1004. Min Range: -8.1004

Max Act: 7.2254. Min Act: -0.1700

Data Index: 1446349 (Python Code)

Max Activating Token Index: 414