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

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

Neuron 633 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.8065. Min Act: -0.1700

Data Index: 351675 (C4 (Web Text))

Max Activating Token Index: 911

Text #1

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.8065. Min Act: -0.1700

Data Index: 647666 (C4 (Web Text))

Max Activating Token Index: 675

Text #2

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7752. Min Act: -0.1700

Data Index: 640202 (C4 (Web Text))

Max Activating Token Index: 464

Text #3

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7440. Min Act: -0.1700

Data Index: 533422 (C4 (Web Text))

Max Activating Token Index: 787

Text #4

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7440. Min Act: -0.1700

Data Index: 127848 (C4 (Web Text))

Max Activating Token Index: 997

Text #5

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7127. Min Act: -0.1700

Data Index: 1094726 (C4 (Web Text))

Max Activating Token Index: 407

Text #6

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7127. Min Act: -0.1700

Data Index: 652791 (C4 (Web Text))

Max Activating Token Index: 845

Text #7

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7127. Min Act: -0.1700

Data Index: 636320 (C4 (Web Text))

Max Activating Token Index: 905

Text #8

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7127. Min Act: -0.1700

Data Index: 80247 (C4 (Web Text))

Max Activating Token Index: 790

Text #9

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7127. Min Act: -0.1700

Data Index: 610517 (C4 (Web Text))

Max Activating Token Index: 564

Text #10

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7440. Min Act: -0.1700

Data Index: 272100 (C4 (Web Text))

Max Activating Token Index: 795

Text #11

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7127. Min Act: -0.1699

Data Index: 402216 (C4 (Web Text))

Max Activating Token Index: 647

Text #12

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.6815. Min Act: -0.1700

Data Index: 592643 (C4 (Web Text))

Max Activating Token Index: 897

Text #13

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.6502. Min Act: -0.1700

Data Index: 976146 (C4 (Web Text))

Max Activating Token Index: 339

Text #14

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.7752. Min Act: -0.1700

Data Index: 421537 (C4 (Web Text))

Max Activating Token Index: 394

Text #15

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.6502. Min Act: -0.1700

Data Index: 1061050 (C4 (Web Text))

Max Activating Token Index: 633

Text #16

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.6502. Min Act: -0.1700

Data Index: 254584 (C4 (Web Text))

Max Activating Token Index: 631

Text #17

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.6502. Min Act: -0.1700

Data Index: 1280296 (C4 (Web Text))

Max Activating Token Index: 603

Text #18

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.5877. Min Act: -0.1700

Data Index: 1055555 (C4 (Web Text))

Max Activating Token Index: 492

Text #19

Max Range: 5.8065. Min Range: -5.8065

Max Act: 5.5565. Min Act: -0.1700

Data Index: 191312 (C4 (Web Text))

Max Activating Token Index: 252