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

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

Neuron 1426 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.3234. Min Act: -0.1700

Data Index: 854035 (C4 (Web Text))

Max Activating Token Index: 548

Text #1

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.3077. Min Act: -0.1700

Data Index: 209958 (C4 (Web Text))

Max Activating Token Index: 642

Text #2

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.2920. Min Act: -0.1700

Data Index: 1473530 (Python Code)

Max Activating Token Index: 980

Text #3

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.2291. Min Act: -0.1700

Data Index: 1625966 (Python Code)

Max Activating Token Index: 644

Text #4

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.1976. Min Act: -0.1700

Data Index: 1513994 (Python Code)

Max Activating Token Index: 932

Text #5

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.1031. Min Act: -0.1700

Data Index: 965602 (C4 (Web Text))

Max Activating Token Index: 825

Text #6

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.1347. Min Act: -0.1700

Data Index: 76667 (C4 (Web Text))

Max Activating Token Index: 718

Text #7

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0716. Min Act: -0.1700

Data Index: 495024 (C4 (Web Text))

Max Activating Token Index: 838

Text #8

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0874. Min Act: -0.1700

Data Index: 1539211 (Python Code)

Max Activating Token Index: 661

Text #9

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0874. Min Act: -0.1700

Data Index: 1503580 (Python Code)

Max Activating Token Index: 560

Text #10

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0400. Min Act: -0.1700

Data Index: 675777 (C4 (Web Text))

Max Activating Token Index: 379

Text #11

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0558. Min Act: -0.1700

Data Index: 1381373 (Python Code)

Max Activating Token Index: 816

Text #12

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0400. Min Act: -0.1700

Data Index: 1614307 (Python Code)

Max Activating Token Index: 826

Text #13

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0716. Min Act: -0.1700

Data Index: 1401313 (Python Code)

Max Activating Token Index: 653

Text #14

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0558. Min Act: -0.1700

Data Index: 1383314 (Python Code)

Max Activating Token Index: 445

Text #15

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0400. Min Act: -0.1700

Data Index: 268923 (C4 (Web Text))

Max Activating Token Index: 7

Text #16

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0558. Min Act: -0.1700

Data Index: 168957 (C4 (Web Text))

Max Activating Token Index: 131

Text #17

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0558. Min Act: -0.1700

Data Index: 133438 (C4 (Web Text))

Max Activating Token Index: 529

Text #18

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0558. Min Act: -0.1700

Data Index: 422998 (C4 (Web Text))

Max Activating Token Index: 819

Text #19

Max Range: 3.3234. Min Range: -3.3234

Max Act: 3.0400. Min Act: -0.1700

Data Index: 1367118 (Python Code)

Max Activating Token Index: 543