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

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

Neuron 321 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: 5.7094. Min Range: -5.7094

Max Act: 5.7094. Min Act: -0.1700

Data Index: 1393996 (Python Code)

Max Activating Token Index: 1020

Text #1

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.6781. Min Act: -0.1700

Data Index: 1404482 (Python Code)

Max Activating Token Index: 881

Text #2

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.6781. Min Act: -0.1700

Data Index: 1434559 (Python Code)

Max Activating Token Index: 82

Text #3

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.6469. Min Act: -0.1700

Data Index: 1432150 (Python Code)

Max Activating Token Index: 987

Text #4

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.6156. Min Act: -0.1700

Data Index: 1361542 (Python Code)

Max Activating Token Index: 1008

Text #5

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.6469. Min Act: -0.1700

Data Index: 1635410 (Python Code)

Max Activating Token Index: 334

Text #6

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5844. Min Act: -0.1700

Data Index: 1568922 (Python Code)

Max Activating Token Index: 923

Text #7

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5844. Min Act: -0.1700

Data Index: 1580130 (Python Code)

Max Activating Token Index: 159

Text #8

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5844. Min Act: -0.1700

Data Index: 1372226 (Python Code)

Max Activating Token Index: 1000

Text #9

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.6156. Min Act: -0.1700

Data Index: 1405247 (Python Code)

Max Activating Token Index: 53

Text #10

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5844. Min Act: -0.1700

Data Index: 1503701 (Python Code)

Max Activating Token Index: 906

Text #11

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.6469. Min Act: -0.1700

Data Index: 1419256 (Python Code)

Max Activating Token Index: 446

Text #12

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5844. Min Act: -0.1700

Data Index: 1502181 (Python Code)

Max Activating Token Index: 161

Text #13

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5531. Min Act: -0.1700

Data Index: 1414765 (Python Code)

Max Activating Token Index: 906

Text #14

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5531. Min Act: -0.1700

Data Index: 1605832 (Python Code)

Max Activating Token Index: 1020

Text #15

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5844. Min Act: -0.1700

Data Index: 1485808 (Python Code)

Max Activating Token Index: 864

Text #16

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5531. Min Act: -0.1700

Data Index: 1399795 (Python Code)

Max Activating Token Index: 908

Text #17

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5531. Min Act: -0.1700

Data Index: 1449571 (Python Code)

Max Activating Token Index: 999

Text #18

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5844. Min Act: -0.1700

Data Index: 1475425 (Python Code)

Max Activating Token Index: 103

Text #19

Max Range: 5.7094. Min Range: -5.7094

Max Act: 5.5844. Min Act: -0.1700

Data Index: 1443454 (Python Code)

Max Activating Token Index: 90