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

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

Neuron 1917 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 4.0775. Min Range: -4.0775

Max Act: 4.0775. Min Act: -0.1700

Data Index: 1402765 (Python Code)

Max Activating Token Index: 46

Text #1

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.7960. Min Act: -0.1700

Data Index: 1599678 (Python Code)

Max Activating Token Index: 109

Text #2

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.6865. Min Act: -0.1700

Data Index: 1438890 (Python Code)

Max Activating Token Index: 292

Text #3

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.6865. Min Act: -0.1700

Data Index: 848701 (C4 (Web Text))

Max Activating Token Index: 68

Text #4

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.6865. Min Act: -0.1700

Data Index: 1594934 (Python Code)

Max Activating Token Index: 5

Text #5

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.6396. Min Act: -0.1700

Data Index: 1001158 (C4 (Web Text))

Max Activating Token Index: 375

Text #6

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.5926. Min Act: -0.1700

Data Index: 1599703 (Python Code)

Max Activating Token Index: 222

Text #7

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.5769. Min Act: -0.1700

Data Index: 23971 (C4 (Web Text))

Max Activating Token Index: 310

Text #8

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.6082. Min Act: -0.1700

Data Index: 1218338 (C4 (Web Text))

Max Activating Token Index: 869

Text #9

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.5769. Min Act: -0.1700

Data Index: 1449802 (Python Code)

Max Activating Token Index: 1012

Text #10

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.5613. Min Act: -0.1700

Data Index: 1605351 (Python Code)

Max Activating Token Index: 13

Text #11

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.5613. Min Act: -0.1700

Data Index: 659763 (C4 (Web Text))

Max Activating Token Index: 100

Text #12

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.5613. Min Act: -0.1700

Data Index: 988586 (C4 (Web Text))

Max Activating Token Index: 45

Text #13

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.5143. Min Act: -0.1700

Data Index: 1114711 (C4 (Web Text))

Max Activating Token Index: 8

Text #14

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.5299. Min Act: -0.1700

Data Index: 931099 (C4 (Web Text))

Max Activating Token Index: 258

Text #15

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.5143. Min Act: -0.1700

Data Index: 1433299 (Python Code)

Max Activating Token Index: 91

Text #16

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.4516. Min Act: -0.1700

Data Index: 1007152 (C4 (Web Text))

Max Activating Token Index: 161

Text #17

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.4829. Min Act: -0.1700

Data Index: 1539417 (Python Code)

Max Activating Token Index: 571

Text #18

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.4516. Min Act: -0.1700

Data Index: 1635170 (Python Code)

Max Activating Token Index: 187

Text #19

Max Range: 4.0775. Min Range: -4.0775

Max Act: 3.4673. Min Act: -0.1700

Data Index: 3072 (C4 (Web Text))

Max Activating Token Index: 9