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

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

Neuron 1740 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: 5.7030. Min Range: -5.7030

Max Act: 5.7030. Min Act: -0.1700

Data Index: 1532694 (Python Code)

Max Activating Token Index: 952

Text #1

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.4218. Min Act: -0.1700

Data Index: 1415493 (Python Code)

Max Activating Token Index: 34

Text #2

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.1405. Min Act: -0.1700

Data Index: 897625 (C4 (Web Text))

Max Activating Token Index: 698

Text #3

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.2030. Min Act: -0.1700

Data Index: 1371963 (Python Code)

Max Activating Token Index: 679

Text #4

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.1405. Min Act: -0.1700

Data Index: 896636 (C4 (Web Text))

Max Activating Token Index: 229

Text #5

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.1093. Min Act: -0.1700

Data Index: 1567135 (Python Code)

Max Activating Token Index: 444

Text #6

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.1093. Min Act: -0.1700

Data Index: 1542669 (Python Code)

Max Activating Token Index: 28

Text #7

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.1718. Min Act: -0.1700

Data Index: 137369 (C4 (Web Text))

Max Activating Token Index: 647

Text #8

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.1093. Min Act: -0.1700

Data Index: 1534531 (Python Code)

Max Activating Token Index: 595

Text #9

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.1405. Min Act: -0.1700

Data Index: 1501886 (Python Code)

Max Activating Token Index: 256

Text #10

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.1405. Min Act: -0.1700

Data Index: 1389489 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.0780. Min Act: -0.1700

Data Index: 451024 (C4 (Web Text))

Max Activating Token Index: 718

Text #12

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.0780. Min Act: -0.1700

Data Index: 1449593 (Python Code)

Max Activating Token Index: 255

Text #13

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.0780. Min Act: -0.1700

Data Index: 584642 (C4 (Web Text))

Max Activating Token Index: 371

Text #14

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.0468. Min Act: -0.1700

Data Index: 1509234 (Python Code)

Max Activating Token Index: 157

Text #15

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.0468. Min Act: -0.1700

Data Index: 1481577 (Python Code)

Max Activating Token Index: 4

Text #16

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.0468. Min Act: -0.1700

Data Index: 1106989 (C4 (Web Text))

Max Activating Token Index: 466

Text #17

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.0780. Min Act: -0.1700

Data Index: 973603 (C4 (Web Text))

Max Activating Token Index: 375

Text #18

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.0155. Min Act: -0.1700

Data Index: 1380647 (Python Code)

Max Activating Token Index: 261

Text #19

Max Range: 5.7030. Min Range: -5.7030

Max Act: 5.0155. Min Act: -0.1700

Data Index: 1193520 (C4 (Web Text))

Max Activating Token Index: 403