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

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

Neuron 393 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: 10.3065. Min Range: -10.3065

Max Act: 10.3065. Min Act: -0.1700

Data Index: 77249 (C4 (Web Text))

Max Activating Token Index: 636

Text #1

Max Range: 10.3065. Min Range: -10.3065

Max Act: 9.6815. Min Act: -0.1700

Data Index: 1622793 (Python Code)

Max Activating Token Index: 874

Text #2

Max Range: 10.3065. Min Range: -10.3065

Max Act: 9.3690. Min Act: -0.1700

Data Index: 1410033 (Python Code)

Max Activating Token Index: 925

Text #3

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.9940. Min Act: -0.1700

Data Index: 1498941 (Python Code)

Max Activating Token Index: 563

Text #4

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.9940. Min Act: -0.1700

Data Index: 1552548 (Python Code)

Max Activating Token Index: 980

Text #5

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.9940. Min Act: -0.1700

Data Index: 1383547 (Python Code)

Max Activating Token Index: 555

Text #6

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.8065. Min Act: -0.1700

Data Index: 1440567 (Python Code)

Max Activating Token Index: 913

Text #7

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.8690. Min Act: -0.1700

Data Index: 1482532 (Python Code)

Max Activating Token Index: 799

Text #8

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.9002. Min Act: -0.1700

Data Index: 1584719 (Python Code)

Max Activating Token Index: 349

Text #9

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.9315. Min Act: -0.1700

Data Index: 1436003 (Python Code)

Max Activating Token Index: 303

Text #10

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.8690. Min Act: -0.1700

Data Index: 1640945 (Python Code)

Max Activating Token Index: 814

Text #11

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.8065. Min Act: -0.1700

Data Index: 1406331 (Python Code)

Max Activating Token Index: 370

Text #12

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.7752. Min Act: -0.1700

Data Index: 1400170 (Python Code)

Max Activating Token Index: 683

Text #13

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.8690. Min Act: -0.1700

Data Index: 1537655 (Python Code)

Max Activating Token Index: 101

Text #14

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.7752. Min Act: -0.1700

Data Index: 1468452 (Python Code)

Max Activating Token Index: 853

Text #15

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.7440. Min Act: -0.1700

Data Index: 1462427 (Python Code)

Max Activating Token Index: 930

Text #16

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.7127. Min Act: -0.1700

Data Index: 1554853 (Python Code)

Max Activating Token Index: 848

Text #17

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.7752. Min Act: -0.1700

Data Index: 1198004 (C4 (Web Text))

Max Activating Token Index: 852

Text #18

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.6815. Min Act: -0.1700

Data Index: 1594989 (Python Code)

Max Activating Token Index: 203

Text #19

Max Range: 10.3065. Min Range: -10.3065

Max Act: 8.7127. Min Act: -0.1700

Data Index: 1568395 (Python Code)

Max Activating Token Index: 893