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

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

Neuron 1896 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.1765. Min Range: -5.1765

Max Act: 5.1765. Min Act: -0.1700

Data Index: 1407432 (Python Code)

Max Activating Token Index: 24

Text #1

Max Range: 5.1765. Min Range: -5.1765

Max Act: 5.0202. Min Act: -0.1700

Data Index: 1426956 (Python Code)

Max Activating Token Index: 484

Text #2

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.9577. Min Act: -0.1700

Data Index: 1530961 (Python Code)

Max Activating Token Index: 1018

Text #3

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.9890. Min Act: -0.1700

Data Index: 1431270 (Python Code)

Max Activating Token Index: 745

Text #4

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8952. Min Act: -0.1700

Data Index: 1500849 (Python Code)

Max Activating Token Index: 977

Text #5

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8952. Min Act: -0.1700

Data Index: 1210696 (C4 (Web Text))

Max Activating Token Index: 620

Text #6

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.9265. Min Act: -0.1699

Data Index: 1551962 (Python Code)

Max Activating Token Index: 666

Text #7

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8327. Min Act: -0.1700

Data Index: 887873 (C4 (Web Text))

Max Activating Token Index: 821

Text #8

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8640. Min Act: -0.1699

Data Index: 1639264 (Python Code)

Max Activating Token Index: 442

Text #9

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8640. Min Act: -0.1700

Data Index: 1443360 (Python Code)

Max Activating Token Index: 264

Text #10

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8327. Min Act: -0.1700

Data Index: 1435457 (Python Code)

Max Activating Token Index: 612

Text #11

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8015. Min Act: -0.1700

Data Index: 1653038 (Python Code)

Max Activating Token Index: 382

Text #12

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8015. Min Act: -0.1700

Data Index: 1443329 (Python Code)

Max Activating Token Index: 148

Text #13

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8015. Min Act: -0.1700

Data Index: 1605929 (Python Code)

Max Activating Token Index: 945

Text #14

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.7702. Min Act: -0.1700

Data Index: 1582144 (Python Code)

Max Activating Token Index: 378

Text #15

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8015. Min Act: -0.1700

Data Index: 1241245 (C4 (Web Text))

Max Activating Token Index: 717

Text #16

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.7390. Min Act: -0.1700

Data Index: 986688 (C4 (Web Text))

Max Activating Token Index: 901

Text #17

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.7702. Min Act: -0.1700

Data Index: 1440516 (Python Code)

Max Activating Token Index: 283

Text #18

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.8015. Min Act: -0.1700

Data Index: 1408808 (Python Code)

Max Activating Token Index: 254

Text #19

Max Range: 5.1765. Min Range: -5.1765

Max Act: 4.7702. Min Act: -0.1700

Data Index: 1590759 (Python Code)

Max Activating Token Index: 81