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

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

Neuron 1447 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: 7.6125. Min Range: -7.6125

Max Act: 7.6125. Min Act: -0.1700

Data Index: 1589365 (Python Code)

Max Activating Token Index: 527

Text #1

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.5187. Min Act: -0.1699

Data Index: 1580422 (Python Code)

Max Activating Token Index: 797

Text #2

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.3937. Min Act: -0.1700

Data Index: 1570209 (Python Code)

Max Activating Token Index: 82

Text #3

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.3625. Min Act: -0.1700

Data Index: 1562364 (Python Code)

Max Activating Token Index: 640

Text #4

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.4250. Min Act: -0.1700

Data Index: 1511865 (Python Code)

Max Activating Token Index: 582

Text #5

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.3312. Min Act: -0.1700

Data Index: 1500318 (Python Code)

Max Activating Token Index: 1005

Text #6

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.3937. Min Act: -0.1700

Data Index: 1577980 (Python Code)

Max Activating Token Index: 129

Text #7

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.3312. Min Act: -0.1700

Data Index: 1534470 (Python Code)

Max Activating Token Index: 1006

Text #8

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.3625. Min Act: -0.1700

Data Index: 1641227 (Python Code)

Max Activating Token Index: 36

Text #9

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.3937. Min Act: -0.1700

Data Index: 1540414 (Python Code)

Max Activating Token Index: 1011

Text #10

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.2062. Min Act: -0.1700

Data Index: 1496051 (Python Code)

Max Activating Token Index: 756

Text #11

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.3000. Min Act: -0.1700

Data Index: 1608795 (Python Code)

Max Activating Token Index: 492

Text #12

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.3000. Min Act: -0.1700

Data Index: 1373268 (Python Code)

Max Activating Token Index: 997

Text #13

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.2687. Min Act: -0.1700

Data Index: 1473660 (Python Code)

Max Activating Token Index: 514

Text #14

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.2375. Min Act: -0.1700

Data Index: 1404558 (Python Code)

Max Activating Token Index: 896

Text #15

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.2062. Min Act: -0.1700

Data Index: 1421010 (Python Code)

Max Activating Token Index: 588

Text #16

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.2062. Min Act: -0.1700

Data Index: 1364285 (Python Code)

Max Activating Token Index: 119

Text #17

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.2062. Min Act: -0.1700

Data Index: 506126 (C4 (Web Text))

Max Activating Token Index: 720

Text #18

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.1125. Min Act: -0.1700

Data Index: 1474591 (Python Code)

Max Activating Token Index: 705

Text #19

Max Range: 7.6125. Min Range: -7.6125

Max Act: 7.1437. Min Act: -0.1700

Data Index: 1506420 (Python Code)

Max Activating Token Index: 171