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

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

Neuron 2012 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: 3.8772. Min Range: -3.8772

Max Act: 3.8772. Min Act: -0.1700

Data Index: 280864 (C4 (Web Text))

Max Activating Token Index: 865

Text #1

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.8460. Min Act: -0.1700

Data Index: 557485 (C4 (Web Text))

Max Activating Token Index: 859

Text #2

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7678. Min Act: -0.1700

Data Index: 693577 (C4 (Web Text))

Max Activating Token Index: 732

Text #3

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7834. Min Act: -0.1700

Data Index: 266145 (C4 (Web Text))

Max Activating Token Index: 755

Text #4

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.8147. Min Act: -0.1700

Data Index: 78065 (C4 (Web Text))

Max Activating Token Index: 729

Text #5

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7678. Min Act: -0.1700

Data Index: 1269044 (C4 (Web Text))

Max Activating Token Index: 865

Text #6

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7990. Min Act: -0.1700

Data Index: 263675 (C4 (Web Text))

Max Activating Token Index: 816

Text #7

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.8147. Min Act: -0.1700

Data Index: 1092548 (C4 (Web Text))

Max Activating Token Index: 700

Text #8

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7990. Min Act: -0.1700

Data Index: 122227 (C4 (Web Text))

Max Activating Token Index: 701

Text #9

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7678. Min Act: -0.1700

Data Index: 1319503 (C4 (Web Text))

Max Activating Token Index: 878

Text #10

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7678. Min Act: -0.1700

Data Index: 1097077 (C4 (Web Text))

Max Activating Token Index: 130

Text #11

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7678. Min Act: -0.1700

Data Index: 1122858 (C4 (Web Text))

Max Activating Token Index: 699

Text #12

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7678. Min Act: -0.1700

Data Index: 531880 (C4 (Web Text))

Max Activating Token Index: 700

Text #13

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7365. Min Act: -0.1700

Data Index: 519954 (C4 (Web Text))

Max Activating Token Index: 790

Text #14

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7678. Min Act: -0.1700

Data Index: 988313 (C4 (Web Text))

Max Activating Token Index: 623

Text #15

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7208. Min Act: -0.1700

Data Index: 432551 (C4 (Web Text))

Max Activating Token Index: 685

Text #16

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7208. Min Act: -0.1700

Data Index: 172510 (C4 (Web Text))

Max Activating Token Index: 691

Text #17

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7365. Min Act: -0.1700

Data Index: 814616 (C4 (Web Text))

Max Activating Token Index: 130

Text #18

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7208. Min Act: -0.1700

Data Index: 623869 (C4 (Web Text))

Max Activating Token Index: 727

Text #19

Max Range: 3.8772. Min Range: -3.8772

Max Act: 3.7365. Min Act: -0.1700

Data Index: 1234560 (C4 (Web Text))

Max Activating Token Index: 128