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

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

Neuron 1990 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: 3.2991. Min Range: -3.2991

Max Act: 3.2991. Min Act: -0.1700

Data Index: 226912 (C4 (Web Text))

Max Activating Token Index: 714

Text #1

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.2362. Min Act: -0.1700

Data Index: 984703 (C4 (Web Text))

Max Activating Token Index: 583

Text #2

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.2047. Min Act: -0.1700

Data Index: 1117668 (C4 (Web Text))

Max Activating Token Index: 1003

Text #3

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1733. Min Act: -0.1700

Data Index: 1182690 (C4 (Web Text))

Max Activating Token Index: 912

Text #4

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1733. Min Act: -0.1700

Data Index: 494875 (C4 (Web Text))

Max Activating Token Index: 732

Text #5

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1418. Min Act: -0.1700

Data Index: 1333769 (C4 (Web Text))

Max Activating Token Index: 806

Text #6

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1575. Min Act: -0.1700

Data Index: 909205 (C4 (Web Text))

Max Activating Token Index: 369

Text #7

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1260. Min Act: -0.1700

Data Index: 949454 (C4 (Web Text))

Max Activating Token Index: 446

Text #8

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1260. Min Act: -0.1700

Data Index: 1551299 (Python Code)

Max Activating Token Index: 845

Text #9

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1260. Min Act: -0.1700

Data Index: 981940 (C4 (Web Text))

Max Activating Token Index: 376

Text #10

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1260. Min Act: -0.1700

Data Index: 196182 (C4 (Web Text))

Max Activating Token Index: 924

Text #11

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1103. Min Act: -0.1700

Data Index: 1082488 (C4 (Web Text))

Max Activating Token Index: 753

Text #12

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1103. Min Act: -0.1700

Data Index: 675863 (C4 (Web Text))

Max Activating Token Index: 84

Text #13

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.0787. Min Act: -0.1700

Data Index: 1195438 (C4 (Web Text))

Max Activating Token Index: 547

Text #14

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.1260. Min Act: -0.1700

Data Index: 359031 (C4 (Web Text))

Max Activating Token Index: 805

Text #15

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.0629. Min Act: -0.1700

Data Index: 158126 (C4 (Web Text))

Max Activating Token Index: 911

Text #16

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.0629. Min Act: -0.1700

Data Index: 620263 (C4 (Web Text))

Max Activating Token Index: 841

Text #17

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.0629. Min Act: -0.1700

Data Index: 1187445 (C4 (Web Text))

Max Activating Token Index: 722

Text #18

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.0629. Min Act: -0.1700

Data Index: 1006171 (C4 (Web Text))

Max Activating Token Index: 897

Text #19

Max Range: 3.2991. Min Range: -3.2991

Max Act: 3.0787. Min Act: -0.1700

Data Index: 732671 (C4 (Web Text))

Max Activating Token Index: 436