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

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

Neuron 554 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: 3.9903. Min Range: -3.9903

Max Act: 3.9903. Min Act: -0.1700

Data Index: 1235792 (C4 (Web Text))

Max Activating Token Index: 436

Text #1

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.9590. Min Act: -0.1700

Data Index: 74548 (C4 (Web Text))

Max Activating Token Index: 719

Text #2

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.9434. Min Act: -0.1700

Data Index: 716201 (C4 (Web Text))

Max Activating Token Index: 513

Text #3

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.9278. Min Act: -0.1700

Data Index: 1120840 (C4 (Web Text))

Max Activating Token Index: 622

Text #4

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.9590. Min Act: -0.1700

Data Index: 841933 (C4 (Web Text))

Max Activating Token Index: 644

Text #5

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.9434. Min Act: -0.1700

Data Index: 726522 (C4 (Web Text))

Max Activating Token Index: 739

Text #6

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.9121. Min Act: -0.1700

Data Index: 1527926 (Python Code)

Max Activating Token Index: 740

Text #7

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.9121. Min Act: -0.1700

Data Index: 575446 (C4 (Web Text))

Max Activating Token Index: 985

Text #8

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8965. Min Act: -0.1700

Data Index: 400062 (C4 (Web Text))

Max Activating Token Index: 787

Text #9

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.9278. Min Act: -0.1700

Data Index: 1262591 (C4 (Web Text))

Max Activating Token Index: 864

Text #10

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8965. Min Act: -0.1700

Data Index: 1322829 (C4 (Web Text))

Max Activating Token Index: 989

Text #11

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8652. Min Act: -0.1700

Data Index: 251113 (C4 (Web Text))

Max Activating Token Index: 1007

Text #12

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8809. Min Act: -0.1700

Data Index: 129794 (C4 (Web Text))

Max Activating Token Index: 867

Text #13

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8496. Min Act: -0.1700

Data Index: 74045 (C4 (Web Text))

Max Activating Token Index: 257

Text #14

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8496. Min Act: -0.1700

Data Index: 994207 (C4 (Web Text))

Max Activating Token Index: 924

Text #15

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8652. Min Act: -0.1700

Data Index: 710021 (C4 (Web Text))

Max Activating Token Index: 290

Text #16

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8809. Min Act: -0.1700

Data Index: 1604368 (Python Code)

Max Activating Token Index: 501

Text #17

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8339. Min Act: -0.1700

Data Index: 1371604 (Python Code)

Max Activating Token Index: 547

Text #18

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8339. Min Act: -0.1700

Data Index: 1559410 (Python Code)

Max Activating Token Index: 667

Text #19

Max Range: 3.9903. Min Range: -3.9903

Max Act: 3.8652. Min Act: -0.1700

Data Index: 1509321 (Python Code)

Max Activating Token Index: 721