Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 790 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.9217. Min Act: -0.1700

Data Index: 14969 (C4 (Web Text))

Max Activating Token Index: 337

Text #1

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.7966. Min Act: -0.1700

Data Index: 902866 (C4 (Web Text))

Max Activating Token Index: 178

Text #2

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.7340. Min Act: -0.1700

Data Index: 955835 (C4 (Web Text))

Max Activating Token Index: 393

Text #3

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.7653. Min Act: -0.1700

Data Index: 919780 (C4 (Web Text))

Max Activating Token Index: 862

Text #4

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.7340. Min Act: -0.1700

Data Index: 696495 (C4 (Web Text))

Max Activating Token Index: 498

Text #5

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.7966. Min Act: -0.1700

Data Index: 1055772 (C4 (Web Text))

Max Activating Token Index: 519

Text #6

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.7027. Min Act: -0.1700

Data Index: 1591294 (Python Code)

Max Activating Token Index: 420

Text #7

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6871. Min Act: -0.1700

Data Index: 354833 (C4 (Web Text))

Max Activating Token Index: 444

Text #8

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6871. Min Act: -0.1700

Data Index: 719634 (C4 (Web Text))

Max Activating Token Index: 228

Text #9

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.7027. Min Act: -0.1700

Data Index: 1634461 (Python Code)

Max Activating Token Index: 369

Text #10

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6558. Min Act: -0.1700

Data Index: 1574717 (Python Code)

Max Activating Token Index: 774

Text #11

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6401. Min Act: -0.1700

Data Index: 1485449 (Python Code)

Max Activating Token Index: 471

Text #12

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6714. Min Act: -0.1700

Data Index: 33594 (C4 (Web Text))

Max Activating Token Index: 897

Text #13

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6245. Min Act: -0.1700

Data Index: 1435627 (Python Code)

Max Activating Token Index: 373

Text #14

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6245. Min Act: -0.1700

Data Index: 1415127 (Python Code)

Max Activating Token Index: 461

Text #15

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6401. Min Act: -0.1700

Data Index: 1417363 (Python Code)

Max Activating Token Index: 564

Text #16

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6088. Min Act: -0.1700

Data Index: 295453 (C4 (Web Text))

Max Activating Token Index: 812

Text #17

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6245. Min Act: -0.1700

Data Index: 725137 (C4 (Web Text))

Max Activating Token Index: 655

Text #18

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6245. Min Act: -0.1700

Data Index: 735856 (C4 (Web Text))

Max Activating Token Index: 868

Text #19

Max Range: 3.9217. Min Range: -3.9217

Max Act: 3.6401. Min Act: -0.1700

Data Index: 138552 (C4 (Web Text))

Max Activating Token Index: 556