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

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

Neuron 172 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: 7.8669. Min Range: -7.8669

Max Act: 7.8669. Min Act: -0.1700

Data Index: 1418028 (Python Code)

Max Activating Token Index: 749

Text #1

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.8669. Min Act: -0.1700

Data Index: 1508569 (Python Code)

Max Activating Token Index: 1012

Text #2

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.5544. Min Act: -0.1700

Data Index: 1377151 (Python Code)

Max Activating Token Index: 715

Text #3

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.4919. Min Act: -0.1700

Data Index: 1591700 (Python Code)

Max Activating Token Index: 514

Text #4

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.5231. Min Act: -0.1700

Data Index: 1477062 (Python Code)

Max Activating Token Index: 635

Text #5

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.4919. Min Act: -0.1700

Data Index: 1511886 (Python Code)

Max Activating Token Index: 254

Text #6

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.3981. Min Act: -0.1700

Data Index: 1415738 (Python Code)

Max Activating Token Index: 279

Text #7

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.3044. Min Act: -0.1700

Data Index: 1437324 (Python Code)

Max Activating Token Index: 512

Text #8

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.3356. Min Act: -0.1700

Data Index: 1561236 (Python Code)

Max Activating Token Index: 441

Text #9

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.3044. Min Act: -0.1700

Data Index: 1561970 (Python Code)

Max Activating Token Index: 928

Text #10

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.3044. Min Act: -0.1700

Data Index: 1434501 (Python Code)

Max Activating Token Index: 626

Text #11

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.1794. Min Act: -0.1700

Data Index: 1550358 (Python Code)

Max Activating Token Index: 106

Text #12

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.1794. Min Act: -0.1700

Data Index: 1511278 (Python Code)

Max Activating Token Index: 731

Text #13

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.1794. Min Act: -0.1700

Data Index: 1510584 (Python Code)

Max Activating Token Index: 985

Text #14

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.2106. Min Act: -0.1700

Data Index: 1597961 (Python Code)

Max Activating Token Index: 993

Text #15

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.1481. Min Act: -0.1700

Data Index: 1447827 (Python Code)

Max Activating Token Index: 980

Text #16

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.1481. Min Act: -0.1700

Data Index: 1638072 (Python Code)

Max Activating Token Index: 559

Text #17

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.1481. Min Act: -0.1700

Data Index: 1391502 (Python Code)

Max Activating Token Index: 822

Text #18

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.1169. Min Act: -0.1700

Data Index: 1585189 (Python Code)

Max Activating Token Index: 369

Text #19

Max Range: 7.8669. Min Range: -7.8669

Max Act: 7.1481. Min Act: -0.1700

Data Index: 1482523 (Python Code)

Max Activating Token Index: 818