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

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

Neuron 804 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: 5.5998. Min Range: -5.5998

Max Act: 5.5998. Min Act: -0.1700

Data Index: 805763 (C4 (Web Text))

Max Activating Token Index: 671

Text #1

Max Range: 5.5998. Min Range: -5.5998

Max Act: 5.1310. Min Act: -0.1700

Data Index: 1294821 (C4 (Web Text))

Max Activating Token Index: 610

Text #2

Max Range: 5.5998. Min Range: -5.5998

Max Act: 5.0685. Min Act: -0.1700

Data Index: 471649 (C4 (Web Text))

Max Activating Token Index: 874

Text #3

Max Range: 5.5998. Min Range: -5.5998

Max Act: 5.0685. Min Act: -0.1700

Data Index: 364445 (C4 (Web Text))

Max Activating Token Index: 411

Text #4

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.9435. Min Act: -0.1700

Data Index: 1177309 (C4 (Web Text))

Max Activating Token Index: 334

Text #5

Max Range: 5.5998. Min Range: -5.5998

Max Act: 5.0060. Min Act: -0.1700

Data Index: 1187027 (C4 (Web Text))

Max Activating Token Index: 962

Text #6

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.9748. Min Act: -0.1700

Data Index: 948240 (C4 (Web Text))

Max Activating Token Index: 910

Text #7

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.8810. Min Act: -0.1700

Data Index: 371196 (C4 (Web Text))

Max Activating Token Index: 452

Text #8

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.8498. Min Act: -0.1700

Data Index: 523814 (C4 (Web Text))

Max Activating Token Index: 89

Text #9

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.8185. Min Act: -0.1700

Data Index: 570776 (C4 (Web Text))

Max Activating Token Index: 533

Text #10

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.8185. Min Act: -0.1700

Data Index: 555623 (C4 (Web Text))

Max Activating Token Index: 298

Text #11

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.7873. Min Act: -0.1700

Data Index: 579872 (C4 (Web Text))

Max Activating Token Index: 150

Text #12

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.7560. Min Act: -0.1700

Data Index: 1020975 (C4 (Web Text))

Max Activating Token Index: 88

Text #13

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.8498. Min Act: -0.1700

Data Index: 9606 (C4 (Web Text))

Max Activating Token Index: 187

Text #14

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.7873. Min Act: -0.1700

Data Index: 145935 (C4 (Web Text))

Max Activating Token Index: 39

Text #15

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.7873. Min Act: -0.1700

Data Index: 845358 (C4 (Web Text))

Max Activating Token Index: 208

Text #16

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.6623. Min Act: -0.1700

Data Index: 899729 (C4 (Web Text))

Max Activating Token Index: 742

Text #17

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.7248. Min Act: -0.1700

Data Index: 535377 (C4 (Web Text))

Max Activating Token Index: 630

Text #18

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.6310. Min Act: -0.1700

Data Index: 106710 (C4 (Web Text))

Max Activating Token Index: 501

Text #19

Max Range: 5.5998. Min Range: -5.5998

Max Act: 4.7248. Min Act: -0.1700

Data Index: 1293712 (C4 (Web Text))

Max Activating Token Index: 499