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

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

Neuron 665 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.4634. Min Range: -5.4634

Max Act: 5.4634. Min Act: -0.1700

Data Index: 33017 (C4 (Web Text))

Max Activating Token Index: 45

Text #1

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.9946. Min Act: -0.1700

Data Index: 1164096 (C4 (Web Text))

Max Activating Token Index: 162

Text #2

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.9321. Min Act: -0.1699

Data Index: 45879 (C4 (Web Text))

Max Activating Token Index: 603

Text #3

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.9946. Min Act: -0.1699

Data Index: 1003475 (C4 (Web Text))

Max Activating Token Index: 551

Text #4

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.9009. Min Act: -0.1700

Data Index: 994807 (C4 (Web Text))

Max Activating Token Index: 306

Text #5

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.9321. Min Act: -0.1700

Data Index: 620584 (C4 (Web Text))

Max Activating Token Index: 713

Text #6

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.9321. Min Act: -0.1698

Data Index: 665919 (C4 (Web Text))

Max Activating Token Index: 164

Text #7

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.9009. Min Act: -0.1700

Data Index: 407249 (C4 (Web Text))

Max Activating Token Index: 405

Text #8

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.7759. Min Act: -0.1700

Data Index: 450655 (C4 (Web Text))

Max Activating Token Index: 854

Text #9

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.8071. Min Act: -0.1699

Data Index: 1065602 (C4 (Web Text))

Max Activating Token Index: 571

Text #10

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.7759. Min Act: -0.1699

Data Index: 602231 (C4 (Web Text))

Max Activating Token Index: 618

Text #11

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.7759. Min Act: -0.1700

Data Index: 597903 (C4 (Web Text))

Max Activating Token Index: 635

Text #12

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.6821. Min Act: -0.1700

Data Index: 812472 (C4 (Web Text))

Max Activating Token Index: 221

Text #13

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.7134. Min Act: -0.1700

Data Index: 27333 (C4 (Web Text))

Max Activating Token Index: 558

Text #14

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.6821. Min Act: -0.1700

Data Index: 1106339 (C4 (Web Text))

Max Activating Token Index: 569

Text #15

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.6509. Min Act: -0.1700

Data Index: 83150 (C4 (Web Text))

Max Activating Token Index: 591

Text #16

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.6509. Min Act: -0.1700

Data Index: 1082664 (C4 (Web Text))

Max Activating Token Index: 844

Text #17

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.6509. Min Act: -0.1700

Data Index: 1079287 (C4 (Web Text))

Max Activating Token Index: 793

Text #18

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.5884. Min Act: -0.1700

Data Index: 217143 (C4 (Web Text))

Max Activating Token Index: 329

Text #19

Max Range: 5.4634. Min Range: -5.4634

Max Act: 4.6196. Min Act: -0.1700

Data Index: 722976 (C4 (Web Text))

Max Activating Token Index: 698