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

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

Neuron 1979 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.4896. Min Act: -0.1700

Data Index: 1652495 (Python Code)

Max Activating Token Index: 1013

Text #1

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.4426. Min Act: -0.1700

Data Index: 1399526 (Python Code)

Max Activating Token Index: 432

Text #2

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.4583. Min Act: -0.1700

Data Index: 1369771 (Python Code)

Max Activating Token Index: 790

Text #3

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.4112. Min Act: -0.1700

Data Index: 1650240 (Python Code)

Max Activating Token Index: 899

Text #4

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.4112. Min Act: -0.1700

Data Index: 762672 (C4 (Web Text))

Max Activating Token Index: 917

Text #5

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3171. Min Act: -0.1700

Data Index: 1401419 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3642. Min Act: -0.1700

Data Index: 1538982 (Python Code)

Max Activating Token Index: 105

Text #7

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3642. Min Act: -0.1700

Data Index: 1621737 (Python Code)

Max Activating Token Index: 804

Text #8

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3171. Min Act: -0.1700

Data Index: 90521 (C4 (Web Text))

Max Activating Token Index: 940

Text #9

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.2857. Min Act: -0.1700

Data Index: 1437975 (Python Code)

Max Activating Token Index: 416

Text #10

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3014. Min Act: -0.1700

Data Index: 1617373 (Python Code)

Max Activating Token Index: 976

Text #11

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3171. Min Act: -0.1700

Data Index: 421365 (C4 (Web Text))

Max Activating Token Index: 286

Text #12

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3171. Min Act: -0.1700

Data Index: 1637186 (Python Code)

Max Activating Token Index: 425

Text #13

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3328. Min Act: -0.1700

Data Index: 1528932 (Python Code)

Max Activating Token Index: 667

Text #14

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.2857. Min Act: -0.1700

Data Index: 1530158 (Python Code)

Max Activating Token Index: 196

Text #15

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3171. Min Act: -0.1700

Data Index: 1622139 (Python Code)

Max Activating Token Index: 10

Text #16

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3014. Min Act: -0.1700

Data Index: 1520960 (Python Code)

Max Activating Token Index: 242

Text #17

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3014. Min Act: -0.1700

Data Index: 1626417 (Python Code)

Max Activating Token Index: 110

Text #18

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.3014. Min Act: -0.1700

Data Index: 1411704 (Python Code)

Max Activating Token Index: 133

Text #19

Max Range: 3.4896. Min Range: -3.4896

Max Act: 3.2700. Min Act: -0.1700

Data Index: 1631090 (Python Code)

Max Activating Token Index: 485