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

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

Neuron 728 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.5274. Min Act: -0.1700

Data Index: 991209 (C4 (Web Text))

Max Activating Token Index: 832

Text #1

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.5117. Min Act: -0.1700

Data Index: 346865 (C4 (Web Text))

Max Activating Token Index: 903

Text #2

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.5117. Min Act: -0.1700

Data Index: 748840 (C4 (Web Text))

Max Activating Token Index: 891

Text #3

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4647. Min Act: -0.1700

Data Index: 618999 (C4 (Web Text))

Max Activating Token Index: 892

Text #4

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4490. Min Act: -0.1699

Data Index: 977707 (C4 (Web Text))

Max Activating Token Index: 856

Text #5

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4647. Min Act: -0.1700

Data Index: 943206 (C4 (Web Text))

Max Activating Token Index: 701

Text #6

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4490. Min Act: -0.1700

Data Index: 674236 (C4 (Web Text))

Max Activating Token Index: 974

Text #7

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4490. Min Act: -0.1700

Data Index: 20874 (C4 (Web Text))

Max Activating Token Index: 749

Text #8

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4333. Min Act: -0.1700

Data Index: 852696 (C4 (Web Text))

Max Activating Token Index: 857

Text #9

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4804. Min Act: -0.1700

Data Index: 952106 (C4 (Web Text))

Max Activating Token Index: 864

Text #10

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4020. Min Act: -0.1700

Data Index: 1246953 (C4 (Web Text))

Max Activating Token Index: 747

Text #11

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4020. Min Act: -0.1700

Data Index: 1552374 (Python Code)

Max Activating Token Index: 459

Text #12

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4177. Min Act: -0.1700

Data Index: 1214682 (C4 (Web Text))

Max Activating Token Index: 1019

Text #13

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.3863. Min Act: -0.1699

Data Index: 130752 (C4 (Web Text))

Max Activating Token Index: 773

Text #14

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4020. Min Act: -0.1700

Data Index: 129192 (C4 (Web Text))

Max Activating Token Index: 738

Text #15

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4020. Min Act: -0.1700

Data Index: 203278 (C4 (Web Text))

Max Activating Token Index: 845

Text #16

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4333. Min Act: -0.1700

Data Index: 620025 (C4 (Web Text))

Max Activating Token Index: 771

Text #17

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.4177. Min Act: -0.1700

Data Index: 981200 (C4 (Web Text))

Max Activating Token Index: 521

Text #18

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.3549. Min Act: -0.1700

Data Index: 1092570 (C4 (Web Text))

Max Activating Token Index: 1014

Text #19

Max Range: 3.5274. Min Range: -3.5274

Max Act: 3.3549. Min Act: -0.1699

Data Index: 768 (C4 (Web Text))

Max Activating Token Index: 568