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

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

Neuron 654 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: 4.0549. Min Range: -4.0549

Max Act: 4.0549. Min Act: -0.1700

Data Index: 797116 (C4 (Web Text))

Max Activating Token Index: 714

Text #1

Max Range: 4.0549. Min Range: -4.0549

Max Act: 4.0080. Min Act: -0.1700

Data Index: 35588 (C4 (Web Text))

Max Activating Token Index: 467

Text #2

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.9611. Min Act: -0.1700

Data Index: 431872 (C4 (Web Text))

Max Activating Token Index: 245

Text #3

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.9454. Min Act: -0.1700

Data Index: 1026814 (C4 (Web Text))

Max Activating Token Index: 40

Text #4

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.9611. Min Act: -0.1700

Data Index: 676365 (C4 (Web Text))

Max Activating Token Index: 796

Text #5

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.9298. Min Act: -0.1700

Data Index: 1533227 (Python Code)

Max Activating Token Index: 578

Text #6

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8985. Min Act: -0.1700

Data Index: 736574 (C4 (Web Text))

Max Activating Token Index: 6

Text #7

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8985. Min Act: -0.1700

Data Index: 1636716 (Python Code)

Max Activating Token Index: 35

Text #8

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8829. Min Act: -0.1700

Data Index: 127060 (C4 (Web Text))

Max Activating Token Index: 39

Text #9

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8672. Min Act: -0.1700

Data Index: 1400451 (Python Code)

Max Activating Token Index: 830

Text #10

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8829. Min Act: -0.1700

Data Index: 1520601 (Python Code)

Max Activating Token Index: 647

Text #11

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8829. Min Act: -0.1700

Data Index: 830544 (C4 (Web Text))

Max Activating Token Index: 37

Text #12

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8829. Min Act: -0.1700

Data Index: 1098288 (C4 (Web Text))

Max Activating Token Index: 825

Text #13

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8829. Min Act: -0.1700

Data Index: 30603 (C4 (Web Text))

Max Activating Token Index: 497

Text #14

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8672. Min Act: -0.1700

Data Index: 196229 (C4 (Web Text))

Max Activating Token Index: 83

Text #15

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8672. Min Act: -0.1700

Data Index: 976811 (C4 (Web Text))

Max Activating Token Index: 784

Text #16

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8829. Min Act: -0.1700

Data Index: 484539 (C4 (Web Text))

Max Activating Token Index: 707

Text #17

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8829. Min Act: -0.1700

Data Index: 9812 (C4 (Web Text))

Max Activating Token Index: 386

Text #18

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8516. Min Act: -0.1700

Data Index: 104878 (C4 (Web Text))

Max Activating Token Index: 839

Text #19

Max Range: 4.0549. Min Range: -4.0549

Max Act: 3.8360. Min Act: -0.1700

Data Index: 264967 (C4 (Web Text))

Max Activating Token Index: 830