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

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

Neuron 1394 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: 3.1794. Min Range: -3.1794

Max Act: 3.1794. Min Act: -0.1700

Data Index: 457139 (C4 (Web Text))

Max Activating Token Index: 128

Text #1

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.1321. Min Act: -0.1700

Data Index: 1009548 (C4 (Web Text))

Max Activating Token Index: 714

Text #2

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0691. Min Act: -0.1700

Data Index: 1650018 (Python Code)

Max Activating Token Index: 896

Text #3

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0375. Min Act: -0.1700

Data Index: 269937 (C4 (Web Text))

Max Activating Token Index: 815

Text #4

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0375. Min Act: -0.1700

Data Index: 702767 (C4 (Web Text))

Max Activating Token Index: 278

Text #5

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0217. Min Act: -0.1700

Data Index: 1278597 (C4 (Web Text))

Max Activating Token Index: 415

Text #6

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0375. Min Act: -0.1700

Data Index: 1612998 (Python Code)

Max Activating Token Index: 796

Text #7

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0533. Min Act: -0.1700

Data Index: 285692 (C4 (Web Text))

Max Activating Token Index: 367

Text #8

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0375. Min Act: -0.1700

Data Index: 489522 (C4 (Web Text))

Max Activating Token Index: 617

Text #9

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0533. Min Act: -0.1700

Data Index: 563278 (C4 (Web Text))

Max Activating Token Index: 909

Text #10

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0375. Min Act: -0.1700

Data Index: 9470 (C4 (Web Text))

Max Activating Token Index: 556

Text #11

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0059. Min Act: -0.1700

Data Index: 1228527 (C4 (Web Text))

Max Activating Token Index: 701

Text #12

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0059. Min Act: -0.1700

Data Index: 96615 (C4 (Web Text))

Max Activating Token Index: 784

Text #13

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0375. Min Act: -0.1700

Data Index: 245368 (C4 (Web Text))

Max Activating Token Index: 844

Text #14

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0059. Min Act: -0.1700

Data Index: 14629 (C4 (Web Text))

Max Activating Token Index: 737

Text #15

Max Range: 3.1794. Min Range: -3.1794

Max Act: 2.9743. Min Act: -0.1700

Data Index: 702578 (C4 (Web Text))

Max Activating Token Index: 878

Text #16

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0059. Min Act: -0.1700

Data Index: 350059 (C4 (Web Text))

Max Activating Token Index: 976

Text #17

Max Range: 3.1794. Min Range: -3.1794

Max Act: 2.9901. Min Act: -0.1700

Data Index: 1192118 (C4 (Web Text))

Max Activating Token Index: 652

Text #18

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0059. Min Act: -0.1700

Data Index: 884421 (C4 (Web Text))

Max Activating Token Index: 762

Text #19

Max Range: 3.1794. Min Range: -3.1794

Max Act: 3.0059. Min Act: -0.1700

Data Index: 1068745 (C4 (Web Text))

Max Activating Token Index: 861