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

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

Neuron 1649 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.0783. Min Range: -4.0783

Max Act: 4.0471. Min Act: -0.1700

Data Index: 981181 (C4 (Web Text))

Max Activating Token Index: 957

Text #1

Max Range: 4.0783. Min Range: -4.0783

Max Act: 4.0783. Min Act: -0.1700

Data Index: 513459 (C4 (Web Text))

Max Activating Token Index: 558

Text #2

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.8282. Min Act: -0.1700

Data Index: 584393 (C4 (Web Text))

Max Activating Token Index: 78

Text #3

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.7656. Min Act: -0.1700

Data Index: 357979 (C4 (Web Text))

Max Activating Token Index: 191

Text #4

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.7187. Min Act: -0.1700

Data Index: 642032 (C4 (Web Text))

Max Activating Token Index: 973

Text #5

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.7343. Min Act: -0.1700

Data Index: 781258 (C4 (Web Text))

Max Activating Token Index: 1008

Text #6

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.7030. Min Act: -0.1700

Data Index: 493434 (C4 (Web Text))

Max Activating Token Index: 856

Text #7

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6874. Min Act: -0.1700

Data Index: 1023434 (C4 (Web Text))

Max Activating Token Index: 693

Text #8

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6248. Min Act: -0.1700

Data Index: 728747 (C4 (Web Text))

Max Activating Token Index: 543

Text #9

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6561. Min Act: -0.1700

Data Index: 1211484 (C4 (Web Text))

Max Activating Token Index: 868

Text #10

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6404. Min Act: -0.1700

Data Index: 202917 (C4 (Web Text))

Max Activating Token Index: 122

Text #11

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6248. Min Act: -0.1700

Data Index: 531547 (C4 (Web Text))

Max Activating Token Index: 468

Text #12

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6874. Min Act: -0.1700

Data Index: 1091288 (C4 (Web Text))

Max Activating Token Index: 412

Text #13

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6091. Min Act: -0.1700

Data Index: 1260905 (C4 (Web Text))

Max Activating Token Index: 725

Text #14

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6091. Min Act: -0.1700

Data Index: 1047305 (C4 (Web Text))

Max Activating Token Index: 206

Text #15

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6248. Min Act: -0.1700

Data Index: 813462 (C4 (Web Text))

Max Activating Token Index: 393

Text #16

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.5621. Min Act: -0.1700

Data Index: 1636591 (Python Code)

Max Activating Token Index: 726

Text #17

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6248. Min Act: -0.1700

Data Index: 548729 (C4 (Web Text))

Max Activating Token Index: 937

Text #18

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6404. Min Act: -0.1700

Data Index: 329779 (C4 (Web Text))

Max Activating Token Index: 681

Text #19

Max Range: 4.0783. Min Range: -4.0783

Max Act: 3.6248. Min Act: -0.1700

Data Index: 60559 (C4 (Web Text))

Max Activating Token Index: 400