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

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

Neuron 638 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.9012. Min Act: -0.1700

Data Index: 767638 (C4 (Web Text))

Max Activating Token Index: 866

Text #1

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.9012. Min Act: -0.1700

Data Index: 485127 (C4 (Web Text))

Max Activating Token Index: 163

Text #2

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.8074. Min Act: -0.1700

Data Index: 1179658 (C4 (Web Text))

Max Activating Token Index: 969

Text #3

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.8700. Min Act: -0.1700

Data Index: 798360 (C4 (Web Text))

Max Activating Token Index: 305

Text #4

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7761. Min Act: -0.1700

Data Index: 222436 (C4 (Web Text))

Max Activating Token Index: 400

Text #5

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7448. Min Act: -0.1700

Data Index: 1351306 (C4 (Web Text))

Max Activating Token Index: 443

Text #6

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.8074. Min Act: -0.1700

Data Index: 1022800 (C4 (Web Text))

Max Activating Token Index: 81

Text #7

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.8074. Min Act: -0.1700

Data Index: 1014298 (C4 (Web Text))

Max Activating Token Index: 97

Text #8

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7761. Min Act: -0.1700

Data Index: 33829 (C4 (Web Text))

Max Activating Token Index: 341

Text #9

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7292. Min Act: -0.1700

Data Index: 316445 (C4 (Web Text))

Max Activating Token Index: 515

Text #10

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.8387. Min Act: -0.1700

Data Index: 978944 (C4 (Web Text))

Max Activating Token Index: 737

Text #11

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7761. Min Act: -0.1700

Data Index: 564086 (C4 (Web Text))

Max Activating Token Index: 500

Text #12

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7448. Min Act: -0.1700

Data Index: 1359342 (C4 (Web Text))

Max Activating Token Index: 760

Text #13

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7292. Min Act: -0.1700

Data Index: 628839 (C4 (Web Text))

Max Activating Token Index: 787

Text #14

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7448. Min Act: -0.1700

Data Index: 1338751 (C4 (Web Text))

Max Activating Token Index: 564

Text #15

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7135. Min Act: -0.1700

Data Index: 376410 (C4 (Web Text))

Max Activating Token Index: 881

Text #16

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7448. Min Act: -0.1700

Data Index: 1112381 (C4 (Web Text))

Max Activating Token Index: 739

Text #17

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.7292. Min Act: -0.1700

Data Index: 882718 (C4 (Web Text))

Max Activating Token Index: 344

Text #18

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.6979. Min Act: -0.1700

Data Index: 854078 (C4 (Web Text))

Max Activating Token Index: 35

Text #19

Max Range: 3.9012. Min Range: -3.9012

Max Act: 3.6509. Min Act: -0.1700

Data Index: 164002 (C4 (Web Text))

Max Activating Token Index: 861