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

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

Neuron 36 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: 5.5836. Min Range: -5.5836

Max Act: 5.5836. Min Act: -0.1700

Data Index: 557300 (C4 (Web Text))

Max Activating Token Index: 927

Text #1

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.4586. Min Act: -0.1700

Data Index: 93344 (C4 (Web Text))

Max Activating Token Index: 501

Text #2

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.3961. Min Act: -0.1700

Data Index: 693749 (C4 (Web Text))

Max Activating Token Index: 918

Text #3

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.3649. Min Act: -0.1700

Data Index: 396972 (C4 (Web Text))

Max Activating Token Index: 493

Text #4

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.3024. Min Act: -0.1700

Data Index: 245311 (C4 (Web Text))

Max Activating Token Index: 169

Text #5

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.3336. Min Act: -0.1700

Data Index: 1169223 (C4 (Web Text))

Max Activating Token Index: 748

Text #6

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.3336. Min Act: -0.1700

Data Index: 1278347 (C4 (Web Text))

Max Activating Token Index: 672

Text #7

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.2711. Min Act: -0.1700

Data Index: 595545 (C4 (Web Text))

Max Activating Token Index: 780

Text #8

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.2399. Min Act: -0.1700

Data Index: 1189571 (C4 (Web Text))

Max Activating Token Index: 39

Text #9

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.2399. Min Act: -0.1700

Data Index: 472939 (C4 (Web Text))

Max Activating Token Index: 887

Text #10

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.1774. Min Act: -0.1700

Data Index: 159999 (C4 (Web Text))

Max Activating Token Index: 563

Text #11

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.1774. Min Act: -0.1700

Data Index: 416653 (C4 (Web Text))

Max Activating Token Index: 894

Text #12

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.1774. Min Act: -0.1700

Data Index: 1310537 (C4 (Web Text))

Max Activating Token Index: 115

Text #13

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.2399. Min Act: -0.1700

Data Index: 1045654 (C4 (Web Text))

Max Activating Token Index: 865

Text #14

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.1149. Min Act: -0.1700

Data Index: 686712 (C4 (Web Text))

Max Activating Token Index: 232

Text #15

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.1461. Min Act: -0.1700

Data Index: 638086 (C4 (Web Text))

Max Activating Token Index: 151

Text #16

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.1461. Min Act: -0.1700

Data Index: 784200 (C4 (Web Text))

Max Activating Token Index: 288

Text #17

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.1461. Min Act: -0.1700

Data Index: 714964 (C4 (Web Text))

Max Activating Token Index: 542

Text #18

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.1149. Min Act: -0.1700

Data Index: 68361 (C4 (Web Text))

Max Activating Token Index: 454

Text #19

Max Range: 5.5836. Min Range: -5.5836

Max Act: 5.1149. Min Act: -0.1700

Data Index: 216314 (C4 (Web Text))

Max Activating Token Index: 513