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

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

Neuron 639 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: 7.0795. Min Range: -7.0795

Max Act: 7.0795. Min Act: -0.1700

Data Index: 191730 (C4 (Web Text))

Max Activating Token Index: 893

Text #1

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.8920. Min Act: -0.1700

Data Index: 544285 (C4 (Web Text))

Max Activating Token Index: 187

Text #2

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.7045. Min Act: -0.1700

Data Index: 142819 (C4 (Web Text))

Max Activating Token Index: 959

Text #3

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.9857. Min Act: -0.1700

Data Index: 1012813 (C4 (Web Text))

Max Activating Token Index: 926

Text #4

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.7357. Min Act: -0.1700

Data Index: 450704 (C4 (Web Text))

Max Activating Token Index: 721

Text #5

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.7357. Min Act: -0.1700

Data Index: 1357536 (C4 (Web Text))

Max Activating Token Index: 622

Text #6

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.5795. Min Act: -0.1700

Data Index: 88831 (C4 (Web Text))

Max Activating Token Index: 882

Text #7

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.7045. Min Act: -0.1700

Data Index: 1146296 (C4 (Web Text))

Max Activating Token Index: 759

Text #8

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.6420. Min Act: -0.1700

Data Index: 442964 (C4 (Web Text))

Max Activating Token Index: 789

Text #9

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.6732. Min Act: -0.1700

Data Index: 310816 (C4 (Web Text))

Max Activating Token Index: 268

Text #10

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.6107. Min Act: -0.1700

Data Index: 129489 (C4 (Web Text))

Max Activating Token Index: 784

Text #11

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.6107. Min Act: -0.1700

Data Index: 138122 (C4 (Web Text))

Max Activating Token Index: 19

Text #12

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.4545. Min Act: -0.1700

Data Index: 1401729 (Python Code)

Max Activating Token Index: 443

Text #13

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.5170. Min Act: -0.1700

Data Index: 54929 (C4 (Web Text))

Max Activating Token Index: 644

Text #14

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.4545. Min Act: -0.1700

Data Index: 829232 (C4 (Web Text))

Max Activating Token Index: 695

Text #15

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.5170. Min Act: -0.1700

Data Index: 542596 (C4 (Web Text))

Max Activating Token Index: 815

Text #16

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.4545. Min Act: -0.1700

Data Index: 969401 (C4 (Web Text))

Max Activating Token Index: 882

Text #17

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.4232. Min Act: -0.1700

Data Index: 169438 (C4 (Web Text))

Max Activating Token Index: 671

Text #18

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.5482. Min Act: -0.1700

Data Index: 852308 (C4 (Web Text))

Max Activating Token Index: 308

Text #19

Max Range: 7.0795. Min Range: -7.0795

Max Act: 6.5482. Min Act: -0.1700

Data Index: 670665 (C4 (Web Text))

Max Activating Token Index: 447